From michael.foord at resolversystems.com  Mon Dec  1 14:13:43 2008
From: michael.foord at resolversystems.com (Michael Foord)
Date: Mon, 01 Dec 2008 13:13:43 +0000
Subject: [IronPython] Change to COM Interop in IronPython 2
Message-ID: <4933E307.8040909@resolversystems.com>

Hello guys,

In some of our Resolver One installer tests we do some COM interop to 
check that the targets of the shortcuts the installer creates are correct.

It looks like the way COM interop is done in IronPython 2 has changed a 
bit. We are now getting an exception thrown when previously None was 
returned to us, and I'd like to check that this is the correct / 
expected behaviour.

The code use to look like this:

   def GetLinkTarget(filename):
       import clr
       import sys
       sys.path.append(Path.GetFullPath('TestUtils'))
       clr.AddReferenceToFile('Interop.Shell32.dll')
       from Interop.Shell32 import ShellClass
       shell = ShellClass()
       folder = shell.NameSpace(Path.GetDirectoryName(filename))
       folderItem = folder.ParseName(Path.GetFileName(filename))
       if folderItem is not None:
           return folderItem.GetLink.Path

In moving to IronPython 2 we've had to change the last line to:

    return folderItem.GetLink().Path()

If we call the function with a filename that doesn't exist, 
folder.ParseName(...) used to return None. Now we get the following 
exception:

Traceback (most recent call last):
 File "FunctionalTests\UnitTests\InstallerTestTest.py", line 97, in 
testGetLinkTarget2
   self.assertNone(InstallerTest.GetLinkTarget(
 File "C:\shared\Ipy2\FunctionalTests\InstallerTest.py", line 67, in 
GetLinkTarget
   folderItem = folder.ParseName(Path.GetFileName(filename))
TypeError: Value cannot be null.
Parameter name: pUnk

All the best,

Michael Foord

-- 
http://www.ironpythoninaction.com/
http://www.voidspace.org.uk/blog




From fuzzyman at voidspace.org.uk  Mon Dec  1 16:41:22 2008
From: fuzzyman at voidspace.org.uk (Michael Foord)
Date: Mon, 01 Dec 2008 15:41:22 +0000
Subject: [IronPython] __init__ ignored in favor of base
 class	constructor? - blocker
In-Reply-To: <4817b6fc0811300742i210e9a0at834b57803dd2e899@mail.gmail.com>
References: <4817b6fc0811291904v4c284026o3fa7093096a056b@mail.gmail.com>	<49329565.8030301@voidspace.org.uk>
	<4817b6fc0811300742i210e9a0at834b57803dd2e899@mail.gmail.com>
Message-ID: <493405A2.7030508@voidspace.org.uk>

Dan Eloff wrote:
> On Sun, Nov 30, 2008 at 8:30 AM, Michael Foord
> <fuzzyman at voidspace.org.uk> wrote:
>   
>> Hello Dan,
>>
>> I can't find this assembly to experiment - but when you subclass a .NET type
>> in IronPython you override the constructor by overriding '__new__' and not
>> '__init__'. Perhaps this is the cause of your problem? (If the default
>> constructor takes two arguments then merely supplying an __init__ method
>> that only takes one won't stop the default two argument constructor being
>> called.)
>>
>>     
>
> I don't think I've ever subclassed a .NET type that has a default
> constructor taking an argument, hopefully the problem here is with me
> and not IronPython :) I thought __new__ might be the one to use, but I
> got these errors with __new__:
>
> class MyTheme(Theme):
>     def __new__(cls):
>         return Theme.__new__(cls,
> Application.GetResourceStream(Uri('theme.xaml', UriKind.Relative)))
>
> TypeError: default __new__ does not take parameters.
>
>   

Looks like the subclassing bug *again*. :-)

(i.e. it looks like  Theme.__new__(...) is actually being diverted back 
to MyTheme.__new__(...))

Michael

> If I use:
>
> class MyTheme(Theme):
>     def __new__(cls):
>         return Theme.__new__(cls)
>
> TypeError: MyTheme() takes at least 2 arguments (1 given)
>
> Back to square one. Do you have some idea what I'm doing wrong here?
> The assembly is from the Silverlight Toolkit:
> http://www.codeplex.com/Silverlight
>
> -Dan
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>   


-- 
http://www.ironpythoninaction.com/
http://www.voidspace.org.uk/blog




From dan.eloff at gmail.com  Mon Dec  1 20:15:02 2008
From: dan.eloff at gmail.com (Dan Eloff)
Date: Mon, 1 Dec 2008 14:15:02 -0500
Subject: [IronPython] __init__ ignored in favor of base class
	constructor? - blocker
In-Reply-To: <493405A2.7030508@voidspace.org.uk>
References: <4817b6fc0811291904v4c284026o3fa7093096a056b@mail.gmail.com>
	<49329565.8030301@voidspace.org.uk>
	<4817b6fc0811300742i210e9a0at834b57803dd2e899@mail.gmail.com>
	<493405A2.7030508@voidspace.org.uk>
Message-ID: <4817b6fc0812011115y62037e8bpa8f8cbecd25e71a5@mail.gmail.com>

On Mon, Dec 1, 2008 at 10:41 AM, Michael Foord
<fuzzyman at voidspace.org.uk> wrote:
>
> Looks like the subclassing bug *again*. :-)
>
> (i.e. it looks like  Theme.__new__(...) is actually being diverted back to
> MyTheme.__new__(...))
>
> Michael
>

I suspect this is the same bug you resolver guys have been reporting
recently, but I reported this at
http://www.codeplex.com/IronPython/WorkItem/View.aspx?WorkItemId=20021
just to be thorough.

-Dan


From dinov at microsoft.com  Mon Dec  1 20:22:43 2008
From: dinov at microsoft.com (Dino Viehland)
Date: Mon, 1 Dec 2008 11:22:43 -0800
Subject: [IronPython] __init__ ignored in favor of base
	class	constructor? - blocker
In-Reply-To: <4817b6fc0812011115y62037e8bpa8f8cbecd25e71a5@mail.gmail.com>
References: <4817b6fc0811291904v4c284026o3fa7093096a056b@mail.gmail.com>
	<49329565.8030301@voidspace.org.uk>
	<4817b6fc0811300742i210e9a0at834b57803dd2e899@mail.gmail.com>
	<493405A2.7030508@voidspace.org.uk>
	<4817b6fc0812011115y62037e8bpa8f8cbecd25e71a5@mail.gmail.com>
Message-ID: <350E7D38B6D819428718949920EC2355564A33293C@NA-EXMSG-C102.redmond.corp.microsoft.com>

Thanks for opening this - I'm going to look at this and the other subclassing issues and see how scary the fix(es) are.

-----Original Message-----
From: users-bounces at lists.ironpython.com [mailto:users-bounces at lists.ironpython.com] On Behalf Of Dan Eloff
Sent: Monday, December 01, 2008 11:15 AM
To: Discussion of IronPython
Subject: Re: [IronPython] __init__ ignored in favor of base class constructor? - blocker

On Mon, Dec 1, 2008 at 10:41 AM, Michael Foord
<fuzzyman at voidspace.org.uk> wrote:
>
> Looks like the subclassing bug *again*. :-)
>
> (i.e. it looks like  Theme.__new__(...) is actually being diverted back to
> MyTheme.__new__(...))
>
> Michael
>

I suspect this is the same bug you resolver guys have been reporting
recently, but I reported this at
http://www.codeplex.com/IronPython/WorkItem/View.aspx?WorkItemId=20021
just to be thorough.

-Dan
_______________________________________________
Users mailing list
Users at lists.ironpython.com
http://lists.ironpython.com/listinfo.cgi/users-ironpython.com


From dinov at microsoft.com  Mon Dec  1 20:26:22 2008
From: dinov at microsoft.com (Dino Viehland)
Date: Mon, 1 Dec 2008 11:26:22 -0800
Subject: [IronPython] [Hosting] Cannot convert callable class to delegate
In-Reply-To: <ea89d0e50811221033v9409e07l25868ebccbdcccdc@mail.gmail.com>
References: <ea89d0e50811221033v9409e07l25868ebccbdcccdc@mail.gmail.com>
Message-ID: <350E7D38B6D819428718949920EC2355564A332947@NA-EXMSG-C102.redmond.corp.microsoft.com>

Thanks for reporting & opening the bug.  The fix for this is likely fairly easy but I don't think we'd spin a new 2.0 build for just this issue because it isn't really blocking anything as you have a work around.

That being said it would certainly be something we'd fix for 2.0.1 and we don't expect the time between 2.0 and 2.0.1 to be particularly long.  But if we were to spin a new 2.0 build for some other reason (which I'm a little worried we might need to do) we might just include this fix.

-----Original Message-----
From: users-bounces at lists.ironpython.com [mailto:users-bounces at lists.ironpython.com] On Behalf Of Jeff Hardy
Sent: Saturday, November 22, 2008 10:34 AM
To: Discussion of IronPython
Subject: [IronPython] [Hosting] Cannot convert callable class to delegate

Hi,
Attempting to convert a class with a __call__() member to a delagte
causes an ArgumentNullException; presumably it should create a
delegate, as it is a valid callable object. It can be worked around be
using a function that forwards to the class, but it's an annoyance to
have to do that.

I have opened #19724
(http://www.codeplex.com/IronPython/WorkItem/View.aspx?WorkItemId=19724);
I mention it here because I would really like to see it fixed for 2.0.

-Jeff
_______________________________________________
Users mailing list
Users at lists.ironpython.com
http://lists.ironpython.com/listinfo.cgi/users-ironpython.com


From dinov at microsoft.com  Mon Dec  1 20:43:45 2008
From: dinov at microsoft.com (Dino Viehland)
Date: Mon, 1 Dec 2008 11:43:45 -0800
Subject: [IronPython] __init__ ignored in favor of base
	class	constructor? - blocker
In-Reply-To: <4817b6fc0811300742i210e9a0at834b57803dd2e899@mail.gmail.com>
References: <4817b6fc0811291904v4c284026o3fa7093096a056b@mail.gmail.com>
	<49329565.8030301@voidspace.org.uk>
	<4817b6fc0811300742i210e9a0at834b57803dd2e899@mail.gmail.com>
Message-ID: <350E7D38B6D819428718949920EC2355564A33296B@NA-EXMSG-C102.redmond.corp.microsoft.com>

Ok, I think the problem here is two-fold.  First there's no public default constructor - only a private one.  So you need to pick between the Stream overload or the StreamResourceInfo, Uri overload.  You're apparently trying to choose the Stream overload.

Which brings me to the 2nd problem.  Quite surprisingly GetResourceStream doesn't seem to return a stream - it returns a StreamResourceInfo.  So ultimately there are no applicable methods for us to call.

I think you want:

class MyTheme(Theme):
    def __new__(cls):
        return Theme.__new__(cls, Application.GetResourceStream(Uri('theme.xaml', UriKind.Relative)).Stream)

And then we need to continue to work on improving our error messages which remain particularly bad in some spots :).

-----Original Message-----
From: users-bounces at lists.ironpython.com [mailto:users-bounces at lists.ironpython.com] On Behalf Of Dan Eloff
Sent: Sunday, November 30, 2008 7:43 AM
To: Discussion of IronPython
Subject: Re: [IronPython] __init__ ignored in favor of base class constructor? - blocker

On Sun, Nov 30, 2008 at 8:30 AM, Michael Foord
<fuzzyman at voidspace.org.uk> wrote:
> Hello Dan,
>
> I can't find this assembly to experiment - but when you subclass a .NET type
> in IronPython you override the constructor by overriding '__new__' and not
> '__init__'. Perhaps this is the cause of your problem? (If the default
> constructor takes two arguments then merely supplying an __init__ method
> that only takes one won't stop the default two argument constructor being
> called.)
>

I don't think I've ever subclassed a .NET type that has a default
constructor taking an argument, hopefully the problem here is with me
and not IronPython :) I thought __new__ might be the one to use, but I
got these errors with __new__:

class MyTheme(Theme):
    def __new__(cls):
        return Theme.__new__(cls,
Application.GetResourceStream(Uri('theme.xaml', UriKind.Relative)))

TypeError: default __new__ does not take parameters.

If I use:

class MyTheme(Theme):
    def __new__(cls):
        return Theme.__new__(cls)

TypeError: MyTheme() takes at least 2 arguments (1 given)

Back to square one. Do you have some idea what I'm doing wrong here?
The assembly is from the Silverlight Toolkit:
http://www.codeplex.com/Silverlight

-Dan
_______________________________________________
Users mailing list
Users at lists.ironpython.com
http://lists.ironpython.com/listinfo.cgi/users-ironpython.com


From vernondcole at gmail.com  Mon Dec  1 21:09:29 2008
From: vernondcole at gmail.com (Vernon Cole)
Date: Mon, 1 Dec 2008 13:09:29 -0700
Subject: [IronPython] Announcing IronPython 2.0 RC2
In-Reply-To: <B0F32F33670C5D47BB6D5D315C32CD66221E44FD61@NA-EXMSG-C108.redmond.corp.microsoft.com>
References: <AclQJAOUdQygCQEeTeS7NNg7+S41rw==>
	<B0F32F33670C5D47BB6D5D315C32CD66221E44FD61@NA-EXMSG-C108.redmond.corp.microsoft.com>
Message-ID: <b7eb4a700812011209k22990778he4092e52cbb86842@mail.gmail.com>

Congratulations and "Well Done" to the Iron Python team!
This release has taken Iron Python from an experimental curiosity to a
viable work engine (IMHO).

I ran my test suite and found only one remaining restriction -- Binary data
cannot be passed as a parameter. (The same error shows up four times in
testing because six different environments are used. It passes all tests
under mySQL which does not attempt Binary.)

I am enclosing an edited copy of the test results to document the result I
found most surprising: Iron Python was faster than CPython in each test
case!

The test was set up as follows:
Workstation: A mid-range HP minitower with a dual-core 2.4GHz CPU and 2 GB
ram running XP Pro.
ACCESS database built on the local C: drive.
MS SQL database on a venerable SQL2000 server somewhere over the fiber.
mySQL on an old Ubuntu workstation under my desk.

Each test is run once with datetime.datetime dates and once with time.gmtime
dates. (Unrelated Question: Why is datetime.datetime so *much* slower???)
Then a generic test for dbapi 2.0 compliance is run.

V v v v v v v v v Begin CPython Tests v v v v v v v v v v v v
...Lib\site-packages\adodbapi\tests>adodbapitest.py

2.6 (r26:66721, Oct  2 2008, 11:35:03) [MSC v.1500 32 bit (Intel)]
adodbapi v2.2.4
adodbapitestconfig.py v 2.2.4
    ...Creating ACCESS db at c:\docume~1\wnvcole\locals~1\temp\test_odbc.mdb
    ...Testing MS-SQL login...
    ...Testing MySql login...
 Unit tests for adodbapi version 2.2.4
Default Date Converter is <adodbapi.pythonDateTimeConverter object at
0x00C54470
...................................................................................................
----------------------------------------------------------------------
Ran 99 tests in 11.125s

OK
Changed dateconverter to <adodbapi.pythonTimeConverter object at 0x01349BB0>
................................................................................
...................
----------------------------------------------------------------------
Ran 99 tests in 4.344s

OK
. . . . . . . . . . . . . . .
...Lib\site-packages\adodbapi\tests>test_adodbapi_dbapi20.py
This module depends on the dbapi20 compliance tests created by Stuart Bishop
(see db-sig mailing list history for info)
Tested with dbapi20 1.10
2.6 (r26:66721, Oct  2 2008, 11:35:03) [MSC v.1500 32 bit (Intel)]
adodbapi v2.2.4
adodbapitestconfig.py v 2.2.4
    ...Creating ACCESS db at c:\docume~1\wnvcole\locals~1\temp\test_odbc.mdb
    ...Testing MS-SQL login...
    ...Testing MySql login...
....................................
----------------------------------------------------------------------
Ran 36 tests in 2.203s

OK
V v v v v v v v v Begin IronPython Tests v v v v v v v v v v v v
...\Lib\site-packages\adodbapi\tests>"C:\Program Files\IronPython
2.0\ipy.exe" adodbapitest.py
2.5.0 (IronPython 2.0 (2.0.0.0) on .NET 2.0.50727.1433)
adodbapi v2.2.4
adodbapitestconfig.py v 2.2.4
    ...Creating ACCESS db at c:\documents and settings\wnvcole\local
settings\temp\test_odbc.mdb
    ...Testing MS-SQL login...
    ...Testing MySql login...
 Unit tests for adodbapi version 2.2.4
Default Date Converter is <pythonDateTimeConverter object at
0x000000000000002B>

............................E......................E...............................................
======================================================================
ERROR: testDataTypeBinary (__main__.TestADOwithAccessDB)
----------------------------------------------------------------------
[...snip...Traceback ...]
  File "C:\Program Files\IronPython
2.0\Lib\site-packages\adodbapi\adodbapi.py", line 685, in _executeHelper
    p.AppendChunk(elem)
TypeError: Specified cast is not valid.
-- on command: "INSERT INTO tblTemp (fldId,fldData) VALUES (?,?)"
-- with parameters: (2, <read-only buffer for 0x000000000000002C, size 4,
offset 0 at 0x000000000000002D>)
======================================================================
ERROR: testDataTypeBinary (__main__.TestADOwithSQLServer)
----------------------------------------------------------------------
[...snip (same error) ...]
----------------------------------------------------------------------
Ran 99 tests in 9.516s

FAILED (errors=2)
Changed dateconverter to <pythonTimeConverter object at 0x0000000000000032>
............................E......................E...............................................
======================================================================
ERROR: testDataTypeBinary (__main__.TestADOwithAccessDB)
----------------------------------------------------------------------
[...snip (same error) ...]
======================================================================
ERROR: testDataTypeBinary (__main__.TestADOwithSQLServer)
----------------------------------------------------------------------
[...snip (same error) ...]
----------------------------------------------------------------------
Ran 99 tests in 2.453s

FAILED (errors=2)
. . . . . . .
...\Lib\site-packages\adodbapi\tests>"C:\Program Files\IronPython
2.0\ipy.exe" test_adodbapi_dbapi20.py
This module depends on the dbapi20 compliance tests created by Stuart Bishop
(see db-sig mailing list history for info)
Tested with dbapi20 1.10
2.5.0 (IronPython 2.0 (2.0.0.0) on .NET 2.0.50727.1433)
adodbapi v2.2.4
adodbapitestconfig.py v 2.2.4
    ...Creating ACCESS db at c:\documents and settings\wnvcole\local
settings\temp\test_odbc.mdb
    ...Testing MS-SQL login...
    ...Testing MySql login...
....................................
----------------------------------------------------------------------
Ran 36 tests in 2.125s

OK
^ ^ ^ ^ ^ ^ end of tests ^ ^ ^ ^ ^ ^
Vernon Cole

On Wed, Nov 26, 2008 at 5:06 PM, Srivatsn Narayanan
<srivatsn at microsoft.com>wrote:

>  [...]
>
> List of bugs fixed for this release:...
> -          19130: One hour error in IPY implementation of time.mktime and
> time.gmtime
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081201/5ca2e444/attachment.html>

From dan.eloff at gmail.com  Mon Dec  1 21:21:33 2008
From: dan.eloff at gmail.com (Dan Eloff)
Date: Mon, 1 Dec 2008 15:21:33 -0500
Subject: [IronPython] __init__ ignored in favor of base class
	constructor? - blocker
In-Reply-To: <350E7D38B6D819428718949920EC2355564A33296B@NA-EXMSG-C102.redmond.corp.microsoft.com>
References: <4817b6fc0811291904v4c284026o3fa7093096a056b@mail.gmail.com>
	<49329565.8030301@voidspace.org.uk>
	<4817b6fc0811300742i210e9a0at834b57803dd2e899@mail.gmail.com>
	<350E7D38B6D819428718949920EC2355564A33296B@NA-EXMSG-C102.redmond.corp.microsoft.com>
Message-ID: <4817b6fc0812011221j6acad46fjf3a0c28a2c71d736@mail.gmail.com>

On Mon, Dec 1, 2008 at 2:43 PM, Dino Viehland <dinov at microsoft.com> wrote:
> Ok, I think the problem here is two-fold.  First there's no public default constructor - only a private one.  So you need to pick between the Stream overload or the StreamResourceInfo, Uri overload.  You're apparently trying to choose the Stream overload.
>
> Which brings me to the 2nd problem.  Quite surprisingly GetResourceStream doesn't seem to return a stream - it returns a StreamResourceInfo.  So ultimately there are no applicable methods for us to call.

That's the second time that method has got me. They should have named
it GetResourceStreamInfo

But the problem still remains:

TypeError: default __new__ does not take parameters

Line 7: class MyTheme(Theme):
Line 8:     def __new__(cls):
Line 9:         return Theme.__new__(cls,
wpf.Application.GetResourceStream(wpf.Uri('theme.xaml',
wpf.UriKind.Relative)).Stream)
Line 10:
Line 11: def demo():

TypeError
at __new__ in app.py, line 9
at demo in app.py, line 12
at app.py in app.py, line 21

Thanks,
-Dan


From dinov at microsoft.com  Mon Dec  1 21:35:42 2008
From: dinov at microsoft.com (Dino Viehland)
Date: Mon, 1 Dec 2008 12:35:42 -0800
Subject: [IronPython] __init__ ignored in favor of base
	class	constructor? - blocker
In-Reply-To: <4817b6fc0812011221j6acad46fjf3a0c28a2c71d736@mail.gmail.com>
References: <4817b6fc0811291904v4c284026o3fa7093096a056b@mail.gmail.com>
	<49329565.8030301@voidspace.org.uk>
	<4817b6fc0811300742i210e9a0at834b57803dd2e899@mail.gmail.com>
	<350E7D38B6D819428718949920EC2355564A33296B@NA-EXMSG-C102.redmond.corp.microsoft.com>
	<4817b6fc0812011221j6acad46fjf3a0c28a2c71d736@mail.gmail.com>
Message-ID: <350E7D38B6D819428718949920EC2355564A709B24@NA-EXMSG-C102.redmond.corp.microsoft.com>

It looks like protected ctor's are completely broken:

import clr
clr.AddReference('test')
from Xyz import Foo
class x(Foo):
    def __new__(cls):
        return Foo.__new__(cls, object())

x()

TypeError: default __new__ does not take parameters

test.cs:

using System;

namespace Xyz {
        public class Foo {
                protected Foo(object x) { }
        }
}



-----Original Message-----
From: users-bounces at lists.ironpython.com [mailto:users-bounces at lists.ironpython.com] On Behalf Of Dan Eloff
Sent: Monday, December 01, 2008 12:22 PM
To: Discussion of IronPython
Subject: Re: [IronPython] __init__ ignored in favor of base class constructor? - blocker

On Mon, Dec 1, 2008 at 2:43 PM, Dino Viehland <dinov at microsoft.com> wrote:
> Ok, I think the problem here is two-fold.  First there's no public default constructor - only a private one.  So you need to pick between the Stream overload or the StreamResourceInfo, Uri overload.  You're apparently trying to choose the Stream overload.
>
> Which brings me to the 2nd problem.  Quite surprisingly GetResourceStream doesn't seem to return a stream - it returns a StreamResourceInfo.  So ultimately there are no applicable methods for us to call.

That's the second time that method has got me. They should have named
it GetResourceStreamInfo

But the problem still remains:

TypeError: default __new__ does not take parameters

Line 7: class MyTheme(Theme):
Line 8:     def __new__(cls):
Line 9:         return Theme.__new__(cls,
wpf.Application.GetResourceStream(wpf.Uri('theme.xaml',
wpf.UriKind.Relative)).Stream)
Line 10:
Line 11: def demo():

TypeError
at __new__ in app.py, line 9
at demo in app.py, line 12
at app.py in app.py, line 21

Thanks,
-Dan
_______________________________________________
Users mailing list
Users at lists.ironpython.com
http://lists.ironpython.com/listinfo.cgi/users-ironpython.com


From fuzzyman at voidspace.org.uk  Mon Dec  1 21:38:02 2008
From: fuzzyman at voidspace.org.uk (Michael Foord)
Date: Mon, 01 Dec 2008 20:38:02 +0000
Subject: [IronPython] __init__ ignored in favor of
 base	class	constructor? - blocker
In-Reply-To: <350E7D38B6D819428718949920EC2355564A709B24@NA-EXMSG-C102.redmond.corp.microsoft.com>
References: <4817b6fc0811291904v4c284026o3fa7093096a056b@mail.gmail.com>	<49329565.8030301@voidspace.org.uk>	<4817b6fc0811300742i210e9a0at834b57803dd2e899@mail.gmail.com>	<350E7D38B6D819428718949920EC2355564A33296B@NA-EXMSG-C102.redmond.corp.microsoft.com>	<4817b6fc0812011221j6acad46fjf3a0c28a2c71d736@mail.gmail.com>
	<350E7D38B6D819428718949920EC2355564A709B24@NA-EXMSG-C102.redmond.corp.microsoft.com>
Message-ID: <49344B2A.9070501@voidspace.org.uk>

Dino Viehland wrote:
> It looks like protected ctor's are completely broken:
>
>   

We've had issues calling other protected methods as well - see the bug 
reports we posted on Friday.

Michael Foord
> import clr
> clr.AddReference('test')
> from Xyz import Foo
> class x(Foo):
>     def __new__(cls):
>         return Foo.__new__(cls, object())
>
> x()
>
> TypeError: default __new__ does not take parameters
>
> test.cs:
>
> using System;
>
> namespace Xyz {
>         public class Foo {
>                 protected Foo(object x) { }
>         }
> }
>
>
>
> -----Original Message-----
> From: users-bounces at lists.ironpython.com [mailto:users-bounces at lists.ironpython.com] On Behalf Of Dan Eloff
> Sent: Monday, December 01, 2008 12:22 PM
> To: Discussion of IronPython
> Subject: Re: [IronPython] __init__ ignored in favor of base class constructor? - blocker
>
> On Mon, Dec 1, 2008 at 2:43 PM, Dino Viehland <dinov at microsoft.com> wrote:
>   
>> Ok, I think the problem here is two-fold.  First there's no public default constructor - only a private one.  So you need to pick between the Stream overload or the StreamResourceInfo, Uri overload.  You're apparently trying to choose the Stream overload.
>>
>> Which brings me to the 2nd problem.  Quite surprisingly GetResourceStream doesn't seem to return a stream - it returns a StreamResourceInfo.  So ultimately there are no applicable methods for us to call.
>>     
>
> That's the second time that method has got me. They should have named
> it GetResourceStreamInfo
>
> But the problem still remains:
>
> TypeError: default __new__ does not take parameters
>
> Line 7: class MyTheme(Theme):
> Line 8:     def __new__(cls):
> Line 9:         return Theme.__new__(cls,
> wpf.Application.GetResourceStream(wpf.Uri('theme.xaml',
> wpf.UriKind.Relative)).Stream)
> Line 10:
> Line 11: def demo():
>
> TypeError
> at __new__ in app.py, line 9
> at demo in app.py, line 12
> at app.py in app.py, line 21
>
> Thanks,
> -Dan
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>   


-- 
http://www.ironpythoninaction.com/
http://www.voidspace.org.uk/blog




From dinov at microsoft.com  Mon Dec  1 21:46:04 2008
From: dinov at microsoft.com (Dino Viehland)
Date: Mon, 1 Dec 2008 12:46:04 -0800
Subject: [IronPython] __init__ ignored in favor of
	base	class	constructor? - blocker
In-Reply-To: <49344B2A.9070501@voidspace.org.uk>
References: <4817b6fc0811291904v4c284026o3fa7093096a056b@mail.gmail.com>
	<49329565.8030301@voidspace.org.uk>
	<4817b6fc0811300742i210e9a0at834b57803dd2e899@mail.gmail.com>
	<350E7D38B6D819428718949920EC2355564A33296B@NA-EXMSG-C102.redmond.corp.microsoft.com>
	<4817b6fc0812011221j6acad46fjf3a0c28a2c71d736@mail.gmail.com>
	<350E7D38B6D819428718949920EC2355564A709B24@NA-EXMSG-C102.redmond.corp.microsoft.com>
	<49344B2A.9070501@voidspace.org.uk>
Message-ID: <350E7D38B6D819428718949920EC2355564A709B34@NA-EXMSG-C102.redmond.corp.microsoft.com>

My guess is that's actually a different bug (possibly the same as 19434 & 19218 which might be dups of each other).  That's actually what I'm trying to look at fixing right now.

-----Original Message-----
From: users-bounces at lists.ironpython.com [mailto:users-bounces at lists.ironpython.com] On Behalf Of Michael Foord
Sent: Monday, December 01, 2008 12:38 PM
To: Discussion of IronPython
Subject: Re: [IronPython] __init__ ignored in favor of base class constructor? - blocker

Dino Viehland wrote:
> It looks like protected ctor's are completely broken:
>
>

We've had issues calling other protected methods as well - see the bug
reports we posted on Friday.

Michael Foord
> import clr
> clr.AddReference('test')
> from Xyz import Foo
> class x(Foo):
>     def __new__(cls):
>         return Foo.__new__(cls, object())
>
> x()
>
> TypeError: default __new__ does not take parameters
>
> test.cs:
>
> using System;
>
> namespace Xyz {
>         public class Foo {
>                 protected Foo(object x) { }
>         }
> }
>
>
>
> -----Original Message-----
> From: users-bounces at lists.ironpython.com [mailto:users-bounces at lists.ironpython.com] On Behalf Of Dan Eloff
> Sent: Monday, December 01, 2008 12:22 PM
> To: Discussion of IronPython
> Subject: Re: [IronPython] __init__ ignored in favor of base class constructor? - blocker
>
> On Mon, Dec 1, 2008 at 2:43 PM, Dino Viehland <dinov at microsoft.com> wrote:
>
>> Ok, I think the problem here is two-fold.  First there's no public default constructor - only a private one.  So you need to pick between the Stream overload or the StreamResourceInfo, Uri overload.  You're apparently trying to choose the Stream overload.
>>
>> Which brings me to the 2nd problem.  Quite surprisingly GetResourceStream doesn't seem to return a stream - it returns a StreamResourceInfo.  So ultimately there are no applicable methods for us to call.
>>
>
> That's the second time that method has got me. They should have named
> it GetResourceStreamInfo
>
> But the problem still remains:
>
> TypeError: default __new__ does not take parameters
>
> Line 7: class MyTheme(Theme):
> Line 8:     def __new__(cls):
> Line 9:         return Theme.__new__(cls,
> wpf.Application.GetResourceStream(wpf.Uri('theme.xaml',
> wpf.UriKind.Relative)).Stream)
> Line 10:
> Line 11: def demo():
>
> TypeError
> at __new__ in app.py, line 9
> at demo in app.py, line 12
> at app.py in app.py, line 21
>
> Thanks,
> -Dan
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>


--
http://www.ironpythoninaction.com/
http://www.voidspace.org.uk/blog


_______________________________________________
Users mailing list
Users at lists.ironpython.com
http://lists.ironpython.com/listinfo.cgi/users-ironpython.com


From Jimmy.Schementi at microsoft.com  Mon Dec  1 23:21:07 2008
From: Jimmy.Schementi at microsoft.com (Jimmy Schementi)
Date: Mon, 1 Dec 2008 14:21:07 -0800
Subject: [IronPython] Some observations about parallel import and
 Silverlight
In-Reply-To: <4817b6fc0811251555y3122a946l480388cd689b6340@mail.gmail.com>
References: <4817b6fc0811251555y3122a946l480388cd689b6340@mail.gmail.com>
Message-ID: <5283CA0A4168DF4FBBD71AE9ECA5A3284BDA9F45E9@NA-EXMSG-C116.redmond.corp.microsoft.com>

This is awesome! Thanks for looking into this.

> -----Original Message-----
> From: users-bounces at lists.ironpython.com [mailto:users-
> bounces at lists.ironpython.com] On Behalf Of Dan Eloff
> Sent: Tuesday, November 25, 2008 3:55 PM
> To: Discussion of IronPython
> Subject: [IronPython] Some observations about parallel import and
> Silverlight
>
> Thanks to the excellent work of Kamil
> (http://github.com/luntain/ipy-parallel-import/tree/master), and the
> help of Jimmy I now have importing being done in parallel in 4 threads
> in Silverlight.
>
> I had to restructure my code a little to reduce circular imports (not
> a bad thing to do anyway) and I had to hand-tweak the dependency graph
> to deal with conditional imports and other difficult to track
> dependencies.
>
> The result was better than I could have hoped for. I'm seeing 43%
> faster loading on my dual-core development pc. I used to have blank
> screen at start that made you wonder if the browser has frozen. But
> now the browser is responsive and I have a nice working progress bar.
> But that's not the best news.
>
> The kicker is that the benefit is even larger on slow single core
> machines with dialup internet. I posted a while back on this list that
> IE does not immediately send an asynchronous request. It waits until
> it's UI thread is idle (which doesn't happen if your importing modules
> like crazy on it.) So the net effect was the application loads (about
> 28 seconds on a P4 2ghz) and then it sends the request, waits for the
> response (which is big in my case), and processes it - all in order.
> So if that takes another 30 seconds, you have nearly 1 minute of load
> time. With the UI thread mostly idle now, this happens in parallel to
> the importing, and the overall load time drops by a whopping 45%.
>
> In the words of my generation - w00t!
>
> -Dan
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com



From Jimmy.Schementi at microsoft.com  Mon Dec  1 23:24:06 2008
From: Jimmy.Schementi at microsoft.com (Jimmy Schementi)
Date: Mon, 1 Dec 2008 14:24:06 -0800
Subject: [IronPython] clr assembly reference error in Ipy /
	Silverlight	(but works in ipy)
In-Reply-To: <20ab7c020811262207u757ab3cfvd8489282c762e79e@mail.gmail.com>
References: <20ab7c020811262135r719b2755j827ee1c821af994d@mail.gmail.com>
	<d2155e360811262137m769c8ac0u3b68c4d6c47342e7@mail.gmail.com>
	<20ab7c020811262144r17b9e8e5sb0b7fe5cf2f78667@mail.gmail.com>
	<d2155e360811262150r7df0a959g3b4f019096d006e8@mail.gmail.com>
	<20ab7c020811262207u757ab3cfvd8489282c762e79e@mail.gmail.com>
Message-ID: <5283CA0A4168DF4FBBD71AE9ECA5A3284BDA9F45F2@NA-EXMSG-C116.redmond.corp.microsoft.com>

You would face this issue even if you used the assembly from C#.

Silverlight has a subset of the .NET framework, so like Curt said, it might be as simple as recompiling, or as hard as changing the code to not depend on missing APIs.


From: users-bounces at lists.ironpython.com [mailto:users-bounces at lists.ironpython.com] On Behalf Of Nummers
Sent: Wednesday, November 26, 2008 10:08 PM
To: Discussion of IronPython
Subject: Re: [IronPython] clr assembly reference error in Ipy / Silverlight (but works in ipy)

ok I think that makes sense, thanks

Just so I understand, this is really an issue of different .NET frameworks between the desktop and Silverlight?  So, I would face this issue even if I did not use DLR and went with C#?
On Thu, Nov 27, 2008 at 12:50 AM, Curt Hagenlocher <curt at hagenlocher.org<mailto:curt at hagenlocher.org>> wrote:
Unfortunately, the Silverlight and desktop CLRs aren't binary-compatible -- you're basically linking against a different set of strongly-named assemblies when you're building a Silverlight version.  You'll almost certainly need to contact the vendor and ask them to provide support for Silverlight in order for it to be usable there.  This may not require much more work than a simple rebuild, but if they've used any APIs that aren't supported under Silverlight (like .NET serialization or the original collection classes) then they'd need to modify the source code as well.

On Wed, Nov 26, 2008 at 9:44 PM, Nummers <danummer at gmail.com<mailto:danummer at gmail.com>> wrote:
Wow , that was fast...

Anyway no I did not / can not, this is a third party dll. From the docs "Technically, it is a .NET 2.0 assembly (DLL library), ..."
On Thu, Nov 27, 2008 at 12:37 AM, Curt Hagenlocher <curt at hagenlocher.org<mailto:curt at hagenlocher.org>> wrote:
You've built two different versions of API.dll -- one for the desktop CLR and one for Silverlight -- yes?
On Wed, Nov 26, 2008 at 9:35 PM, Nummers <danummer at gmail.com<mailto:danummer at gmail.com>> wrote:
Just getting going on Ipy / Silverlight and hit a roadblock.

This works on ipy but gives an error w/ Ipy in Silverlight (IOError: Could not add reference to assembly API.dll).

import clr
clr.AddReference('API.dll')
from OEC.API import *
from OEC.DATA import *

same kind of error with:
clr.AddReference("API, Version=3.2.0.0<http://3.2.0.0/>, Culture=neutral, PublicKeyToken=b6b45f27e2749b17")
and with:
clr.LoadAssemblyByName("API, Version=3.2.0.0<http://3.2.0.0/>, Culture=neutral, PublicKeyToken=b6b45f27e2749b17")

Now, this gives no error:
clr.AddReferenceToFile("API.dll")
However the imports fail (ImportError: No module named OEC.API

Not too much experience loading assemblies but, as I said, all is well in ipy.

Hope I'm just doing something stooopid.

Very much appreciate any help!

Dano
_______________________________________________
Users mailing list
Users at lists.ironpython.com<mailto:Users at lists.ironpython.com>
http://lists.ironpython.com/listinfo.cgi/users-ironpython.com


_______________________________________________
Users mailing list
Users at lists.ironpython.com<mailto:Users at lists.ironpython.com>
http://lists.ironpython.com/listinfo.cgi/users-ironpython.com


_______________________________________________
Users mailing list
Users at lists.ironpython.com<mailto:Users at lists.ironpython.com>
http://lists.ironpython.com/listinfo.cgi/users-ironpython.com


_______________________________________________
Users mailing list
Users at lists.ironpython.com<mailto:Users at lists.ironpython.com>
http://lists.ironpython.com/listinfo.cgi/users-ironpython.com

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081201/44f24e0d/attachment.html>

From Jimmy.Schementi at microsoft.com  Mon Dec  1 23:26:25 2008
From: Jimmy.Schementi at microsoft.com (Jimmy Schementi)
Date: Mon, 1 Dec 2008 14:26:25 -0800
Subject: [IronPython] clr assembly reference error in Ipy /
	Silverlight	(but works in ipy)
In-Reply-To: <26756bf60811272307s4dd27572hb1af1066c348c003@mail.gmail.com>
References: <20ab7c020811262135r719b2755j827ee1c821af994d@mail.gmail.com>
	<26756bf60811272307s4dd27572hb1af1066c348c003@mail.gmail.com>
Message-ID: <5283CA0A4168DF4FBBD71AE9ECA5A3284BDA9F45F7@NA-EXMSG-C116.redmond.corp.microsoft.com>

Correction: the AppManifest.xaml is only for assemblies you want to load when the application starts. Doing clr.AddReference will currently look in the XAP and load the assembly at any time in the program's execution, without needing to have it listed in the AppManifest.xaml.

> -----Original Message-----
> From: users-bounces at lists.ironpython.com [mailto:users-
> bounces at lists.ironpython.com] On Behalf Of Neil(???)
> Sent: Thursday, November 27, 2008 11:08 PM
> To: Discussion of IronPython
> Subject: Re: [IronPython] clr assembly reference error in Ipy /
> Silverlight (but works in ipy)
>
> Well, I guess there is another issue which will possibly cause your
> error message.
>
> With chiron, if you need to reference an additional assembly, you also
> need to modify the AppManifest.xaml file by hand and add an entry to
> it, like this:
>
>   <Deployment.Parts>
>     <!-- Add additional assemblies here -->
>     <AssemblyPart Source="Microsoft.Scripting.Silverlight.dll" />
>     <AssemblyPart Source="Microsoft.Scripting.ExtensionAttribute.dll"
> />
>     <AssemblyPart Source="Microsoft.Scripting.Core.dll" />
>     <AssemblyPart Source="Microsoft.Scripting.dll" />
>     <AssemblyPart Source="IronPython.dll" />
>     <AssemblyPart Source="IronPython.Modules.dll" />
>     <AssemblyPart Source="System.Windows.Controls.Data.dll" />
>   </Deployment.Parts>
>
> Best Regards,
> Neil Chen
>
> On Thu, Nov 27, 2008 at 1:35 PM, Nummers <danummer at gmail.com> wrote:
> > Just getting going on Ipy / Silverlight and hit a roadblock.
> >
> > This works on ipy but gives an error w/ Ipy in Silverlight (IOError:
> Could
> > not add reference to assembly API.dll).
> >
> > import clr
> > clr.AddReference('API.dll')
> > from OEC.API import *
> > from OEC.DATA import *
> >
> > same kind of error with:
> > clr.AddReference("API, Version=3.2.0.0, Culture=neutral,
> > PublicKeyToken=b6b45f27e2749b17")
> > and with:
> > clr.LoadAssemblyByName("API, Version=3.2.0.0, Culture=neutral,
> > PublicKeyToken=b6b45f27e2749b17")
> >
> > Now, this gives no error:
> > clr.AddReferenceToFile("API.dll")
> > However the imports fail (ImportError: No module named OEC.API
> >
> > Not too much experience loading assemblies but, as I said, all is
> well in
> > ipy.
> >
> > Hope I'm just doing something stooopid.
> >
> > Very much appreciate any help!
> >
> > Dano
> > _______________________________________________
> > Users mailing list
> > Users at lists.ironpython.com
> > http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
> >
> >
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com



From jdhardy at gmail.com  Tue Dec  2 03:48:52 2008
From: jdhardy at gmail.com (Jeff Hardy)
Date: Mon, 1 Dec 2008 19:48:52 -0700
Subject: [IronPython] [Hosting] Cannot convert callable class to delegate
In-Reply-To: <350E7D38B6D819428718949920EC2355564A332947@NA-EXMSG-C102.redmond.corp.microsoft.com>
References: <ea89d0e50811221033v9409e07l25868ebccbdcccdc@mail.gmail.com>
	<350E7D38B6D819428718949920EC2355564A332947@NA-EXMSG-C102.redmond.corp.microsoft.com>
Message-ID: <ea89d0e50812011848g36f797b1lb017d12d910dc5a1@mail.gmail.com>

On Mon, Dec 1, 2008 at 12:26 PM, Dino Viehland <dinov at microsoft.com> wrote:
> Thanks for reporting & opening the bug.  The fix for this is likely fairly easy but I don't think we'd spin a new 2.0 build for just this issue because it isn't really blocking anything as you have a work around.
>
> That being said it would certainly be something we'd fix for 2.0.1 and we don't expect the time between 2.0 and 2.0.1 to be particularly long.  But if we were to spin a new 2.0 build for some other reason (which I'm a little worried we might need to do) we might just include this fix.

I agree, no need to hold 2.0 just for this little bug.

-Jeff


From rob.s.oakes at gmail.com  Tue Dec  2 05:01:47 2008
From: rob.s.oakes at gmail.com (Rob Oakes)
Date: Mon, 1 Dec 2008 21:01:47 -0700
Subject: [IronPython] A Multithreaded Question
Message-ID: <4934b348.20038e0a.4384.ffffbfc7@mx.google.com>

Dear Group,

 

For the past several weeks, I have been trying to wrap my head around the
different classes available in System.Threads.  While I feel that I have a
good handle on the "BackgroundWorker" class, I without doubt have a very
long way to go on the others.  With that said, I have a bit of an
architectural question.  I have been working on a multithreaded  download
manager for a small podcast player (written primarily to start teaching
myself Python).  I've written about the details of the download manager at:

 

http://robertsoakes.brinkster.net/blog/?p=222

 

Very briefly, it uses a C# wrapper to interact with the Background
Intelligent Transfer Service (BITS).  Details about the C# wrapper code can
be found at:

 

http://robertsoakes.brinkster.net/blog/?p=141

 

While I have a very rough prototype that works, I'm not sure what the best
way to have it update a progress bar is.  At present, I have created a
BackgroundWorker that sets the ProgressBar.Value whenever something changes.
As this is nearly constant, it seems as though the thread doing the updates
queries the BitsJob hundreds or thousands of times a second.  This seems
unnecessary and kills the responsiveness of the program.  I fixed this
(kindof) by putting the thread doing the query to sleep for 200 ms (via
Thread.Sleep(200)).  While this increases the responsiveness of the program,
it feels like a hack.  Are there any other strategies that I could pursue?

 

Thanks,

 

Rob Oakes



__________ Information from ESET NOD32 Antivirus, version of virus signature
database 3655 (20081201) __________

The message was checked by ESET NOD32 Antivirus.

http://www.eset.com

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081201/52a2b6ea/attachment.html>

From yashgt at gmail.com  Tue Dec  2 09:34:32 2008
From: yashgt at gmail.com (Yash Ganthe)
Date: Tue, 2 Dec 2008 14:04:32 +0530
Subject: [IronPython] Does IronPython 2.0 not recognize packages?
Message-ID: <4f428e7b0812020034o40dbb4a4g8cc39f0f764b976c@mail.gmail.com>

I am using IronPython 2.0 Release Candidate 2


>ipy Program.py
Traceback (most recent call last):
File "Program.py", line 22, in Program.py
File "mscorlib", line unknown, in get_Item
File "mscorlib", line unknown, in ThrowKeyNotFoundException
KeyError: The given key was not present in the dictionary.

Line 22 where it fails has:
from PI import *
I expect all modules from the PI folder to be loaded. The PI folder has
__init__.py which has the following line:
__all__ = ["Authentication"]
Authentication.py is in the PI folder.

I replaced the erring line with :
sys.path.append(r'.\PI')
from Authentication import *

and it worked.

Does IronPython not recognize packages?

Thanks,
Yash
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081202/fd2af29b/attachment.html>

From fuzzyman at voidspace.org.uk  Tue Dec  2 11:47:04 2008
From: fuzzyman at voidspace.org.uk (Michael Foord)
Date: Tue, 02 Dec 2008 10:47:04 +0000
Subject: [IronPython] Does IronPython 2.0 not recognize packages?
In-Reply-To: <4f428e7b0812020034o40dbb4a4g8cc39f0f764b976c@mail.gmail.com>
References: <4f428e7b0812020034o40dbb4a4g8cc39f0f764b976c@mail.gmail.com>
Message-ID: <49351228.10100@voidspace.org.uk>

Yash Ganthe wrote:
> I am using IronPython 2.0 Release Candidate 2
>
>

Packages work fine with IronPython 2.

> >ipy Program.py
> Traceback (most recent call last):
> File "Program.py", line 22, in Program.py
> File "mscorlib", line unknown, in get_Item
> File "mscorlib", line unknown, in ThrowKeyNotFoundException
> KeyError: The given key was not present in the dictionary.
>
> Line 22 where it fails has:
> from PI import *
> I expect all modules from the PI folder to be loaded. The PI folder 
> has __init__.py which has the following line:
> __all__ = ["Authentication"]

This will cause Python to expect the name 'Authentication' to exist in 
the '__init__.py' file itself - which it won't unless that module 
imports it.

> Authentication.py is in the PI folder.
>
> I replaced the erring line with :
> sys.path.append(r'.\PI')
> from Authentication import *
>
This is *very* different to having __all__ = ["Authentication"] in the 
module. Try it without the sys.path manipulation and I think you will 
find it still works.

All the best,

Michael Foord
 

> and it worked.
>
> Does IronPython not recognize packages?
>
> Thanks,
> Yash
>
> ------------------------------------------------------------------------
>
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>   


-- 
http://www.ironpythoninaction.com/



From deepali_abhyankar at persistent.co.in  Tue Dec  2 12:13:03 2008
From: deepali_abhyankar at persistent.co.in (Deepali Abhyankar)
Date: Tue, 2 Dec 2008 16:43:03 +0530
Subject: [IronPython] Logging module error
Message-ID: <002c01c9546e$f1d6c610$d5845230$@co.in>

 

Hi

 

I am using Iron Python 2.0 release candidate 2.

I am using logging module

Code is:

    Import logging

    logger = logging.getLogger("Trial")

    logger.setLevel(logging.INFO)

 

    hdlr = logging.FileHandler("data.txt","w")

    hdlr.setLevel(logging.INFO)

    formatter = logging.Formatter("%(asctime)s-%(name)s-%(message)s")

    hdlr.setFormatter(formatter)

    logger.addHandler(hdlr)   

    logger.info(message)

 

I am facing following error on execution 

 

File "C:\Program Files\IronPython 2.0\Lib\logging\__init__.py", line 985, in
i

nfo

  File "C:\Program Files\IronPython 2.0\Lib\logging\__init__.py", line 1094,
in

_log

  File "C:\Program Files\IronPython 2.0\Lib\logging\__init__.py", line 1063,
in

findCaller

  File "C:\Program Files\IronPython 2.0\Lib\logging\__init__.py", line 71,
in <l

ambda$1>

ValueError: _getframe is not implemented for non-zero depth

 

Please suggest solution to resolve this

 

 

Deepali Abhyankar | Senior Software Engineer| Persistent Systems

 <mailto:monika_pardeshi at persistent.co.in>
deepali_abhyankar at persistent.co.in | Tel: +91 (20) 3023 5040

Innovation in software product design, development and delivery-
<http://www.persistentsys.com/> www.persistentsys.com

 


DISCLAIMER
==========
This e-mail may contain privileged and confidential information which is the property of Persistent Systems Ltd. It is intended only for the use of the individual or entity to which it is addressed. If you are not the intended recipient, you are not authorized to read, retain, copy, print, distribute or use this message. If you have received this communication in error, please notify the sender and delete all copies of this message. Persistent Systems Ltd. does not accept any liability for virus infected mails.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081202/60eda53e/attachment.html>

From fuzzyman at voidspace.org.uk  Tue Dec  2 12:59:35 2008
From: fuzzyman at voidspace.org.uk (Michael Foord)
Date: Tue, 02 Dec 2008 11:59:35 +0000
Subject: [IronPython] Logging module error
In-Reply-To: <002c01c9546e$f1d6c610$d5845230$@co.in>
References: <002c01c9546e$f1d6c610$d5845230$@co.in>
Message-ID: <49352327.4070309@voidspace.org.uk>

Hello Deepali,

sys._getframe is not available in IronPython. If you post the relevant 
section of code from the logging module we may be able to suggest a 
patch / workaround (disabling some functionality may be the way forward 
- but often getframe is used to find a trivial bit of information that 
can be supplied another way).

All the best,

Michael Foord

Deepali Abhyankar wrote:
>
>  
>
> Hi
>
>  
>
> I am using Iron Python 2.0 release candidate 2.
>
> I am using logging module
>
> *Code is*:
>
>     Import logging
>
>     logger = logging.getLogger("Trial")
>
>     logger.setLevel(logging.INFO)
>
>  
>
>     hdlr = logging.FileHandler("data.txt","w")
>
>     hdlr.setLevel(logging.INFO)
>
>     formatter = logging.Formatter("%(asctime)s-%(name)s-%(message)s")
>
>     hdlr.setFormatter(formatter)
>
>     logger.addHandler(hdlr)  
>
>     logger.info(message)
>
>  
>
> *I am facing following error on execution *
>
> * *
>
> File "C:\Program Files\IronPython 2.0\Lib\logging\__init__.py", line 
> 985, in i
>
> nfo
>
>   File "C:\Program Files\IronPython 2.0\Lib\logging\__init__.py", line 
> 1094, in
>
> _log
>
>   File "C:\Program Files\IronPython 2.0\Lib\logging\__init__.py", line 
> 1063, in
>
> findCaller
>
>   File "C:\Program Files\IronPython 2.0\Lib\logging\__init__.py", line 
> 71, in <l
>
> ambda$1>
>
> ValueError: _getframe is not implemented for non-zero depth
>
>  
>
> Please suggest solution to resolve this
>
>  
>
>  
>
> *Deepali Abhyankar **|** Senior Software Engineer**|** Persistent Systems*
>
> *deepali_abhyankar at persistent.co.in* 
> <mailto:monika_pardeshi at persistent.co.in>* **| **Tel: +91 (20) 3023 5040*
>
> *Innovation in software product design, development and delivery- 
> **www.persistentsys.com* <http://www.persistentsys.com/>**
>
>  
>
> DISCLAIMER ========== This e-mail may contain privileged and 
> confidential information which is the property of Persistent Systems 
> Ltd. It is intended only for the use of the individual or entity to 
> which it is addressed. If you are not the intended recipient, you are 
> not authorized to read, retain, copy, print, distribute or use this 
> message. If you have received this communication in error, please 
> notify the sender and delete all copies of this message. Persistent 
> Systems Ltd. does not accept any liability for virus infected mails.
>
> ------------------------------------------------------------------------
>
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>   


-- 
http://www.ironpythoninaction.com/



From fuzzyman at voidspace.org.uk  Tue Dec  2 13:19:48 2008
From: fuzzyman at voidspace.org.uk (Michael Foord)
Date: Tue, 02 Dec 2008 12:19:48 +0000
Subject: [IronPython] Logging module error
In-Reply-To: <000301c95477$4329bce0$c97d36a0$@co.in>
References: <002c01c9546e$f1d6c610$d5845230$@co.in>
	<49352327.4070309@voidspace.org.uk>
	<000301c95477$4329bce0$c97d36a0$@co.in>
Message-ID: <493527E4.1030200@voidspace.org.uk>

Deepali Abhyankar wrote:
> Hi Michael,
>
> This is the only code am writing in ipy.exe
> and getting error
> I am using inbuilt logging module provided by Iron Python.
>   
The logging module is not inbuilt to IronPython - it is part of the
Python standard library. As you can see from the traceback the Python
file the error is occurring in is located at:

"C:\Program Files\IronPython 2.0\Lib\logging\__init__.py"


The traceback also gives the line numbers of the errors.

All the best,

Michael Foord

> Deepali
>
>
>
> -----Original Message-----
> From: Michael Foord [mailto:fuzzyman at voidspace.org.uk] 
> Sent: Tuesday, December 02, 2008 5:30 PM
> To: deepali_abhyankar at persistent.co.in; Discussion of IronPython
> Subject: Re: [IronPython] Logging module error
>
> Hello Deepali,
>
> sys._getframe is not available in IronPython. If you post the relevant 
> section of code from the logging module we may be able to suggest a 
> patch / workaround (disabling some functionality may be the way forward 
> - but often getframe is used to find a trivial bit of information that 
> can be supplied another way).
>
> All the best,
>
> Michael Foord
>
> Deepali Abhyankar wrote:
>   
>>  
>>
>> Hi
>>
>>  
>>
>> I am using Iron Python 2.0 release candidate 2.
>>
>> I am using logging module
>>
>> *Code is*:
>>
>>     Import logging
>>
>>     logger = logging.getLogger("Trial")
>>
>>     logger.setLevel(logging.INFO)
>>
>>  
>>
>>     hdlr = logging.FileHandler("data.txt","w")
>>
>>     hdlr.setLevel(logging.INFO)
>>
>>     formatter = logging.Formatter("%(asctime)s-%(name)s-%(message)s")
>>
>>     hdlr.setFormatter(formatter)
>>
>>     logger.addHandler(hdlr)  
>>
>>     logger.info(message)
>>
>>  
>>
>> *I am facing following error on execution *
>>
>> * *
>>
>> File "C:\Program Files\IronPython 2.0\Lib\logging\__init__.py", line 
>> 985, in i
>>
>> nfo
>>
>>   File "C:\Program Files\IronPython 2.0\Lib\logging\__init__.py", line 
>> 1094, in
>>
>> _log
>>
>>   File "C:\Program Files\IronPython 2.0\Lib\logging\__init__.py", line 
>> 1063, in
>>
>> findCaller
>>
>>   File "C:\Program Files\IronPython 2.0\Lib\logging\__init__.py", line 
>> 71, in <l
>>
>> ambda$1>
>>
>> ValueError: _getframe is not implemented for non-zero depth
>>
>>  
>>
>> Please suggest solution to resolve this
>>
>>  
>>
>>
>> DISCLAIMER ========== This e-mail may contain privileged and 
>> confidential information which is the property of Persistent Systems 
>> Ltd. It is intended only for the use of the individual or entity to 
>> which it is addressed. If you are not the intended recipient, you are 
>> not authorized to read, retain, copy, print, distribute or use this 
>> message. If you have received this communication in error, please 
>> notify the sender and delete all copies of this message. Persistent 
>> Systems Ltd. does not accept any liability for virus infected mails.
>>
>> ------------------------------------------------------------------------
>>
>> _______________________________________________
>> Users mailing list
>> Users at lists.ironpython.com
>> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>>   
>>     
>
>
>   


-- 
http://www.ironpythoninaction.com/




From sanxiyn at gmail.com  Tue Dec  2 13:57:02 2008
From: sanxiyn at gmail.com (Seo Sanghyeon)
Date: Tue, 2 Dec 2008 21:57:02 +0900
Subject: [IronPython] Logging module error
In-Reply-To: <002c01c9546e$f1d6c610$d5845230$@co.in>
References: <002c01c9546e$f1d6c610$d5845230$@co.in>
Message-ID: <5b0248170812020457n42ef8564kc6c00f02aa0b3137@mail.gmail.com>

2008/12/2 Deepali Abhyankar <deepali_abhyankar at persistent.co.in>:
> (logging and _getframe)
> Please suggest solution to resolve this

https://fepy.svn.sourceforge.net/svnroot/fepy/trunk/patches/stdlib/2.4.4/patch-stdlib-logging-getframe

-- 
Seo Sanghyeon


From fuzzyman at voidspace.org.uk  Tue Dec  2 14:01:56 2008
From: fuzzyman at voidspace.org.uk (Michael Foord)
Date: Tue, 02 Dec 2008 13:01:56 +0000
Subject: [IronPython] Logging module error
In-Reply-To: <5b0248170812020457n42ef8564kc6c00f02aa0b3137@mail.gmail.com>
References: <002c01c9546e$f1d6c610$d5845230$@co.in>
	<5b0248170812020457n42ef8564kc6c00f02aa0b3137@mail.gmail.com>
Message-ID: <493531C4.7050505@voidspace.org.uk>

Seo Sanghyeon wrote:
> 2008/12/2 Deepali Abhyankar <deepali_abhyankar at persistent.co.in>:
>   
>> (logging and _getframe)
>> Please suggest solution to resolve this
>>     
>
> https://fepy.svn.sourceforge.net/svnroot/fepy/trunk/patches/stdlib/2.4.4/patch-stdlib-logging-getframe
>
>   
Seo - you're still alive! How about an updated release of FePy? :-)

Michael

-- 
http://www.ironpythoninaction.com/



From muwem at web.de  Tue Dec  2 14:09:14 2008
From: muwem at web.de (Uwe Mueller)
Date: Tue, 02 Dec 2008 14:09:14 +0100
Subject: [IronPython] Status ctypes and IronPython or Alternative?
Message-ID: <1445779584@web.de>

Does anybody know about the status of ctypes for IronPython?

It would be very helpful to have a version that supports function calls to unmanaged code with parameter being a pointer to modifiable byte buffer.

I have tried the version from FEPY on sourceforge. 

It implements only some numeric types and does not have byte or byte buffer types.

Also, it does not run on IPY2.0 because IronPython.Runtime.Calls was removed in IPY2.0

Is there some alternative to ctypes without need to get the feet wet (write non-Python code)?


thx,
Uwe Mueller


____________________________________________________________________
Psssst! Schon vom neuen WEB.DE MultiMessenger geh?rt? 
Der kann`s mit allen: http://www.produkte.web.de/messenger/?did=3123



From oripel at gmail.com  Tue Dec  2 14:16:18 2008
From: oripel at gmail.com (orip)
Date: Tue, 2 Dec 2008 05:16:18 -0800 (PST)
Subject: [IronPython] Does IronPython 2.0 not recognize packages?
In-Reply-To: <4f428e7b0812020034o40dbb4a4g8cc39f0f764b976c@mail.gmail.com>
References: <4f428e7b0812020034o40dbb4a4g8cc39f0f764b976c@mail.gmail.com>
Message-ID: <5eb62ad5-96e1-49d4-89d5-a995526d80c5@n10g2000yqm.googlegroups.com>

If you have the following structure:

 Program.py
 PI
    /__init__.py
    /Authentication.py

an empty __init__.py will allow you to do:

  import PI.Authentication
  from PI import Authentication

If your __init__.py contains:
  import Authentication

Then you can do:
  from PI import *
and
  import PI
  PI.Authentication

'__all__' only works with symbols already defined in __init__.py
itself

On Dec 2, 10:34?am, "Yash Ganthe" <yas... at gmail.com> wrote:
> I am using IronPython 2.0 Release Candidate 2
>
> >ipy Program.py
>
> Traceback (most recent call last):
> File "Program.py", line 22, in Program.py
> File "mscorlib", line unknown, in get_Item
> File "mscorlib", line unknown, in ThrowKeyNotFoundException
> KeyError: The given key was not present in the dictionary.
>
> Line 22 where it fails has:
> from PI import *
> I expect all modules from the PI folder to be loaded. The PI folder has
> __init__.py which has the following line:
> __all__ = ["Authentication"]
> Authentication.py is in the PI folder.
>
> I replaced the erring line with :
> sys.path.append(r'.\PI')
> from Authentication import *
>
> and it worked.
>
> Does IronPython not recognize packages?
>
> Thanks,
> Yash
>
> _______________________________________________
> Users mailing list
> Us... at lists.ironpython.comhttp://lists.ironpython.com/listinfo.cgi/users-ironpython.com


From dan.eloff at gmail.com  Tue Dec  2 14:22:44 2008
From: dan.eloff at gmail.com (Dan Eloff)
Date: Tue, 2 Dec 2008 08:22:44 -0500
Subject: [IronPython] Status ctypes and IronPython or Alternative?
In-Reply-To: <1445779584@web.de>
References: <1445779584@web.de>
Message-ID: <4817b6fc0812020522i4b60df10xb0f8d923868333df@mail.gmail.com>

On Tue, Dec 2, 2008 at 8:09 AM, Uwe Mueller <muwem at web.de> wrote:
> Does anybody know about the status of ctypes for IronPython?

Best we have is in FEPY right now.

> It would be very helpful to have a version that supports function calls to unmanaged code with parameter being a pointer to modifiable byte buffer.
>
> I have tried the version from FEPY on sourceforge.
>
> It implements only some numeric types and does not have byte or byte buffer types.
>
> Also, it does not run on IPY2.0 because IronPython.Runtime.Calls was removed in IPY2.0
>
> Is there some alternative to ctypes without need to get the feet wet (write non-Python code)?
>

No, sorry. PyPy has a pure python ctypes implementation, which would
be the easiest way to get ctypes in IronPython, but it requires
implementing the _rawffi module, which nobody has done yet.

-Dan


From yashgt at gmail.com  Tue Dec  2 14:26:04 2008
From: yashgt at gmail.com (Yash Ganthe)
Date: Tue, 2 Dec 2008 18:56:04 +0530
Subject: [IronPython] Does IronPython 2.0 not recognize packages?
In-Reply-To: <4f428e7b0812020034o40dbb4a4g8cc39f0f764b976c@mail.gmail.com>
References: <4f428e7b0812020034o40dbb4a4g8cc39f0f764b976c@mail.gmail.com>
Message-ID: <4f428e7b0812020526i57123d5ct2e00ef41ccda1974@mail.gmail.com>

>>>>>Michael Foord wrote
This will cause Python to expect the name 'Authentication' to exist in
the '__init__.py' file itself - which it won't unless that module
imports it.
>>>>>

I am pretty new to Python. I put the following in __init__.py
import Authentication
__all__ = ["Authentication"]

I still get the error:
> Traceback (most recent call last):
> File "Program.py", line 22, in Program.py
> File "mscorlib", line unknown, in get_Item
> File "mscorlib", line unknown, in ThrowKeyNotFoundException
> KeyError: The given key was not present in the dictionary.


Thanks,
Yash

-----------------------------------------
Yash Ganthe wrote:
> I am using IronPython 2.0 Release Candidate 2
>
>

Packages work fine with IronPython 2.

> >ipy Program.py
> Traceback (most recent call last):
> File "Program.py", line 22, in Program.py
> File "mscorlib", line unknown, in get_Item
> File "mscorlib", line unknown, in ThrowKeyNotFoundException
> KeyError: The given key was not present in the dictionary.
>
> Line 22 where it fails has:
> from PI import *
> I expect all modules from the PI folder to be loaded. The PI folder
> has __init__.py which has the following line:
> __all__ = ["Authentication"]

This will cause Python to expect the name 'Authentication' to exist in
the '__init__.py' file itself - which it won't unless that module
imports it.

> Authentication.py is in the PI folder.
>
> I replaced the erring line with :
> sys.path.append(r'.\PI')
> from Authentication import *
>
This is *very* different to having __all__ = ["Authentication"] in the
module. Try it without the sys.path manipulation and I think you will
find it still works.

All the best,

Michael Foord
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081202/f033ffc5/attachment.html>

From vernondcole at gmail.com  Tue Dec  2 18:40:56 2008
From: vernondcole at gmail.com (Vernon Cole)
Date: Tue, 2 Dec 2008 10:40:56 -0700
Subject: [IronPython] Logging module error
In-Reply-To: <493531C4.7050505@voidspace.org.uk>
References: <002c01c9546e$f1d6c610$d5845230$@co.in>
	<5b0248170812020457n42ef8564kc6c00f02aa0b3137@mail.gmail.com>
	<493531C4.7050505@voidspace.org.uk>
Message-ID: <b7eb4a700812020940m2fed4784t178fb5e5f9c7c8e9@mail.gmail.com>

Seo:
   Yes, please, make a new release as soon as you can. I have uploaded the
IPy-compatible version of adodbapi to the trunk, and it needs 2.0 to work
correctly. The 2.0 installer from MS is nice, but the "batteries included"
version from FePy will be nicer.
--
Vernon

On Tue, Dec 2, 2008 at 6:01 AM, Michael Foord <fuzzyman at voidspace.org.uk>wrote:

> Seo Sanghyeon wrote:
>
>>
>>  Seo - you're still alive! How about an updated release of FePy? :-)
>
> Michael
>
> --
> http://www.ironpythoninaction.com/
>
>
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081202/3d59b83c/attachment.html>

From Marty.Nelson at symyx.com  Tue Dec  2 20:11:13 2008
From: Marty.Nelson at symyx.com (Marty Nelson)
Date: Tue, 2 Dec 2008 11:11:13 -0800
Subject: [IronPython] Event unhooking and IPy2
In-Reply-To: <93fc29d60811291315r17da962avc82789baabc75029@mail.gmail.com>
References: <2679f6510811280336t2aaca6b7pcdce1b206aa8f4a9@mail.gmail.com><492FD877.2040108@voidspace.org.uk>
	<93fc29d60811291315r17da962avc82789baabc75029@mail.gmail.com>
Message-ID: <515335F32AA04440B3DE6FEA1993E9AD03269FB3@srv-be-101.Symyx-IC.symyx.com>


We've had probably with memory leaks on Windows Form subscribed menu
item clicks, just with plain vanilla C#.

Check out this link http://www.bobpowell.net/eventsubscribers.htm (There
are some additional steps for System.Windows.Forms.Component), and
here's some sample code I wrote in C# that you could leverage:

    [TestClass]
    public class UnhookEventsTests
    {
        private class Item
        {
            public event EventHandler Clicked;

            public void FireEvent()
            {
                if(Clicked != null)
                {
                    Clicked(this, EventArgs.Empty);
                }
            }
        }

        [TestMethod]
        public void UnhookEvents()
        {
            Item item = new Item();
            item.Clicked += Clicked;
            item.Clicked += Clicked2;

            item.FireEvent();

            Assert.AreEqual(2, callCount_);

            FieldInfo field = typeof (Item).GetField("Clicked",
BindingFlags.NonPublic | BindingFlags.Instance);
            Delegate del = (Delegate) field.GetValue(item);
            Delegate[] list = del.GetInvocationList();

            Assert.AreEqual(2, list.Length);

            //This works:
            //foreach (Delegate d in list)
            //{
            //    item.Clicked -= (EventHandler) d;
            //}

            //This is more generic and avoids the cast to the specific
handler type
            EventInfo eventInfo = typeof (Item).GetEvent("Clicked");
            foreach (Delegate d in list)
            {
                eventInfo.RemoveEventHandler(item, d);
            }

            item.FireEvent();

            Assert.AreEqual(2, callCount_);
        }
        private int callCount_;
        private void Clicked(object sender, EventArgs e)
        {
            callCount_++;
        }        
        
        private void Clicked2(object sender, EventArgs e)
        {
            callCount_++;
        }
    }

-----Original Message-----
From: users-bounces at lists.ironpython.com
[mailto:users-bounces at lists.ironpython.com] On Behalf Of Kamil
Dworakowski
Sent: Saturday, November 29, 2008 1:16 PM
To: Discussion of IronPython
Subject: Re: [IronPython] Event unhooking and IPy2

On Fri, Nov 28, 2008 at 11:39 AM, Michael Foord
<fuzzyman at voidspace.org.uk> wrote:
> Just as a follow up to this - the reason for the hack below is that
> subscribing to events from IronPython is still causing our UI objects
(and
> their whole object graphs) to not be garbage collected.

To be fair, we have no reason to belive it is IronPython's fault. The
bug that event handlers would prevent garbage collection was fixed,
and we were able to remove *some* of the event nuking.
_______________________________________________
Users mailing list
Users at lists.ironpython.com
http://lists.ironpython.com/listinfo.cgi/users-ironpython.com

=======
Notice: This e-mail message, together with any attachments, contains
information of Symyx Technologies, Inc. or any of its affiliates or
subsidiaries that may be confidential, proprietary, copyrighted,
privileged and/or protected work product, and is meant solely for
the intended recipient. If you are not the intended recipient, and
have received this message in error, please contact the sender
immediately, permanently delete the original and any copies of this
email and any attachments thereto.


From fuzzyman at voidspace.org.uk  Tue Dec  2 20:48:52 2008
From: fuzzyman at voidspace.org.uk (Michael Foord)
Date: Tue, 02 Dec 2008 19:48:52 +0000
Subject: [IronPython] Event unhooking and IPy2
In-Reply-To: <515335F32AA04440B3DE6FEA1993E9AD03269FB3@srv-be-101.Symyx-IC.symyx.com>
References: <2679f6510811280336t2aaca6b7pcdce1b206aa8f4a9@mail.gmail.com><492FD877.2040108@voidspace.org.uk>	<93fc29d60811291315r17da962avc82789baabc75029@mail.gmail.com>
	<515335F32AA04440B3DE6FEA1993E9AD03269FB3@srv-be-101.Symyx-IC.symyx.com>
Message-ID: <49359124.90808@voidspace.org.uk>

Marty Nelson wrote:
> We've had probably with memory leaks on Windows Form subscribed menu
> item clicks, just with plain vanilla C#.
>
>   
Ok - it may well not be the fault of IronPython. (Certainly some event 
handlers don't cause us problems whilst others do). On the other hand, 
as IronPython *is* keeping track of all the handlers we add, an easy 
interface to unhook them all would be extremely useful.

Michael Foord

> Check out this link http://www.bobpowell.net/eventsubscribers.htm (There
> are some additional steps for System.Windows.Forms.Component), and
> here's some sample code I wrote in C# that you could leverage:
>
>     [TestClass]
>     public class UnhookEventsTests
>     {
>         private class Item
>         {
>             public event EventHandler Clicked;
>
>             public void FireEvent()
>             {
>                 if(Clicked != null)
>                 {
>                     Clicked(this, EventArgs.Empty);
>                 }
>             }
>         }
>
>         [TestMethod]
>         public void UnhookEvents()
>         {
>             Item item = new Item();
>             item.Clicked += Clicked;
>             item.Clicked += Clicked2;
>
>             item.FireEvent();
>
>             Assert.AreEqual(2, callCount_);
>
>             FieldInfo field = typeof (Item).GetField("Clicked",
> BindingFlags.NonPublic | BindingFlags.Instance);
>             Delegate del = (Delegate) field.GetValue(item);
>             Delegate[] list = del.GetInvocationList();
>
>             Assert.AreEqual(2, list.Length);
>
>             //This works:
>             //foreach (Delegate d in list)
>             //{
>             //    item.Clicked -= (EventHandler) d;
>             //}
>
>             //This is more generic and avoids the cast to the specific
> handler type
>             EventInfo eventInfo = typeof (Item).GetEvent("Clicked");
>             foreach (Delegate d in list)
>             {
>                 eventInfo.RemoveEventHandler(item, d);
>             }
>
>             item.FireEvent();
>
>             Assert.AreEqual(2, callCount_);
>         }
>         private int callCount_;
>         private void Clicked(object sender, EventArgs e)
>         {
>             callCount_++;
>         }        
>         
>         private void Clicked2(object sender, EventArgs e)
>         {
>             callCount_++;
>         }
>     }
>
> -----Original Message-----
> From: users-bounces at lists.ironpython.com
> [mailto:users-bounces at lists.ironpython.com] On Behalf Of Kamil
> Dworakowski
> Sent: Saturday, November 29, 2008 1:16 PM
> To: Discussion of IronPython
> Subject: Re: [IronPython] Event unhooking and IPy2
>
> On Fri, Nov 28, 2008 at 11:39 AM, Michael Foord
> <fuzzyman at voidspace.org.uk> wrote:
>   
>> Just as a follow up to this - the reason for the hack below is that
>> subscribing to events from IronPython is still causing our UI objects
>>     
> (and
>   
>> their whole object graphs) to not be garbage collected.
>>     
>
> To be fair, we have no reason to belive it is IronPython's fault. The
> bug that event handlers would prevent garbage collection was fixed,
> and we were able to remove *some* of the event nuking.
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>
> =======
> Notice: This e-mail message, together with any attachments, contains
> information of Symyx Technologies, Inc. or any of its affiliates or
> subsidiaries that may be confidential, proprietary, copyrighted,
> privileged and/or protected work product, and is meant solely for
> the intended recipient. If you are not the intended recipient, and
> have received this message in error, please contact the sender
> immediately, permanently delete the original and any copies of this
> email and any attachments thereto.
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>   


-- 
http://www.ironpythoninaction.com/
http://www.voidspace.org.uk/blog




From dinov at microsoft.com  Tue Dec  2 21:01:29 2008
From: dinov at microsoft.com (Dino Viehland)
Date: Tue, 2 Dec 2008 12:01:29 -0800
Subject: [IronPython] Event unhooking and IPy2
In-Reply-To: <49359124.90808@voidspace.org.uk>
References: <2679f6510811280336t2aaca6b7pcdce1b206aa8f4a9@mail.gmail.com><492FD877.2040108@voidspace.org.uk>
	<93fc29d60811291315r17da962avc82789baabc75029@mail.gmail.com>
	<515335F32AA04440B3DE6FEA1993E9AD03269FB3@srv-be-101.Symyx-IC.symyx.com>
	<49359124.90808@voidspace.org.uk>
Message-ID: <350E7D38B6D819428718949920EC2355564A70A140@NA-EXMSG-C102.redmond.corp.microsoft.com>

We don't always remember what we've wired up though.  If you give us a delegate which matches the type exactly we'll just go ahead and hook it up (to avoid adding an extra level of indirection).  Only if you give us some random callable object will we remember it.  It'd be easy enough for us to remember but I wonder if this is really a good idea.

What if someone else has subscribed to the event?  If you have the ability to clear the events then you could cause someone elses subscription to go away.  Maybe you "know" that no one else will have subscribed but it does seem like this breaks the encapsulation of events.

I do think we have the event leaks fixed (we hold no strong references at all now).  To trackl down leaks here I'd suggest using windbg & SOS to track down the remaining references.  You can !DumpHeap to list the entire heap, find the object type you think is keeping everything alive, and then do !GCRoot on its address and see who's keeping it alive.

Alternately you could use the weak event pattern.

> -----Original Message-----
> From: users-bounces at lists.ironpython.com [mailto:users-
> bounces at lists.ironpython.com] On Behalf Of Michael Foord
> Sent: Tuesday, December 02, 2008 11:49 AM
> To: Discussion of IronPython
> Subject: Re: [IronPython] Event unhooking and IPy2
>
> Marty Nelson wrote:
> > We've had probably with memory leaks on Windows Form subscribed menu
> > item clicks, just with plain vanilla C#.
> >
> >
> Ok - it may well not be the fault of IronPython. (Certainly some event
> handlers don't cause us problems whilst others do). On the other hand,
> as IronPython *is* keeping track of all the handlers we add, an easy
> interface to unhook them all would be extremely useful.
>
> Michael Foord
>
> > Check out this link http://www.bobpowell.net/eventsubscribers.htm
> (There
> > are some additional steps for System.Windows.Forms.Component), and
> > here's some sample code I wrote in C# that you could leverage:
> >
> >     [TestClass]
> >     public class UnhookEventsTests
> >     {
> >         private class Item
> >         {
> >             public event EventHandler Clicked;
> >
> >             public void FireEvent()
> >             {
> >                 if(Clicked != null)
> >                 {
> >                     Clicked(this, EventArgs.Empty);
> >                 }
> >             }
> >         }
> >
> >         [TestMethod]
> >         public void UnhookEvents()
> >         {
> >             Item item = new Item();
> >             item.Clicked += Clicked;
> >             item.Clicked += Clicked2;
> >
> >             item.FireEvent();
> >
> >             Assert.AreEqual(2, callCount_);
> >
> >             FieldInfo field = typeof (Item).GetField("Clicked",
> > BindingFlags.NonPublic | BindingFlags.Instance);
> >             Delegate del = (Delegate) field.GetValue(item);
> >             Delegate[] list = del.GetInvocationList();
> >
> >             Assert.AreEqual(2, list.Length);
> >
> >             //This works:
> >             //foreach (Delegate d in list)
> >             //{
> >             //    item.Clicked -= (EventHandler) d;
> >             //}
> >
> >             //This is more generic and avoids the cast to the
> specific
> > handler type
> >             EventInfo eventInfo = typeof (Item).GetEvent("Clicked");
> >             foreach (Delegate d in list)
> >             {
> >                 eventInfo.RemoveEventHandler(item, d);
> >             }
> >
> >             item.FireEvent();
> >
> >             Assert.AreEqual(2, callCount_);
> >         }
> >         private int callCount_;
> >         private void Clicked(object sender, EventArgs e)
> >         {
> >             callCount_++;
> >         }
> >
> >         private void Clicked2(object sender, EventArgs e)
> >         {
> >             callCount_++;
> >         }
> >     }
> >
> > -----Original Message-----
> > From: users-bounces at lists.ironpython.com
> > [mailto:users-bounces at lists.ironpython.com] On Behalf Of Kamil
> > Dworakowski
> > Sent: Saturday, November 29, 2008 1:16 PM
> > To: Discussion of IronPython
> > Subject: Re: [IronPython] Event unhooking and IPy2
> >
> > On Fri, Nov 28, 2008 at 11:39 AM, Michael Foord
> > <fuzzyman at voidspace.org.uk> wrote:
> >
> >> Just as a follow up to this - the reason for the hack below is that
> >> subscribing to events from IronPython is still causing our UI
> objects
> >>
> > (and
> >
> >> their whole object graphs) to not be garbage collected.
> >>
> >
> > To be fair, we have no reason to belive it is IronPython's fault. The
> > bug that event handlers would prevent garbage collection was fixed,
> > and we were able to remove *some* of the event nuking.
> > _______________________________________________
> > Users mailing list
> > Users at lists.ironpython.com
> > http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
> >
> > =======
> > Notice: This e-mail message, together with any attachments, contains
> > information of Symyx Technologies, Inc. or any of its affiliates or
> > subsidiaries that may be confidential, proprietary, copyrighted,
> > privileged and/or protected work product, and is meant solely for
> > the intended recipient. If you are not the intended recipient, and
> > have received this message in error, please contact the sender
> > immediately, permanently delete the original and any copies of this
> > email and any attachments thereto.
> > _______________________________________________
> > Users mailing list
> > Users at lists.ironpython.com
> > http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
> >
>
>
> --
> http://www.ironpythoninaction.com/
> http://www.voidspace.org.uk/blog
>
>
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com



From fuzzyman at voidspace.org.uk  Tue Dec  2 21:14:29 2008
From: fuzzyman at voidspace.org.uk (Michael Foord)
Date: Tue, 02 Dec 2008 20:14:29 +0000
Subject: [IronPython] Event unhooking and IPy2
In-Reply-To: <350E7D38B6D819428718949920EC2355564A70A140@NA-EXMSG-C102.redmond.corp.microsoft.com>
References: <2679f6510811280336t2aaca6b7pcdce1b206aa8f4a9@mail.gmail.com><492FD877.2040108@voidspace.org.uk>	<93fc29d60811291315r17da962avc82789baabc75029@mail.gmail.com>	<515335F32AA04440B3DE6FEA1993E9AD03269FB3@srv-be-101.Symyx-IC.symyx.com>	<49359124.90808@voidspace.org.uk>
	<350E7D38B6D819428718949920EC2355564A70A140@NA-EXMSG-C102.redmond.corp.microsoft.com>
Message-ID: <49359725.40605@voidspace.org.uk>

Dino Viehland wrote:
> We don't always remember what we've wired up though.  If you give us a delegate which matches the type exactly we'll just go ahead and hook it up (to avoid adding an extra level of indirection).  Only if you give us some random callable object will we remember it.  It'd be easy enough for us to remember but I wonder if this is really a good idea.
>
>   
That sounds fair enough.

> What if someone else has subscribed to the event?  If you have the ability to clear the events then you could cause someone elses subscription to go away.  Maybe you "know" that no one else will have subscribed but it does seem like this breaks the encapsulation of events.
>
>   

Well, in our case it is when we want to dispose of UI elements - so 
anything listening *must* unsubscribe.

> I do think we have the event leaks fixed (we hold no strong references at all now).  To trackl down leaks here I'd suggest using windbg & SOS to track down the remaining references.  You can !DumpHeap to list the entire heap, find the object type you think is keeping everything alive, and then do !GCRoot on its address and see who's keeping it alive.
>
>   

This is what we do - and great fun it is too. :-)

Michael

> Alternately you could use the weak event pattern.
>
>   
>> -----Original Message-----
>> From: users-bounces at lists.ironpython.com [mailto:users-
>> bounces at lists.ironpython.com] On Behalf Of Michael Foord
>> Sent: Tuesday, December 02, 2008 11:49 AM
>> To: Discussion of IronPython
>> Subject: Re: [IronPython] Event unhooking and IPy2
>>
>> Marty Nelson wrote:
>>     
>>> We've had probably with memory leaks on Windows Form subscribed menu
>>> item clicks, just with plain vanilla C#.
>>>
>>>
>>>       
>> Ok - it may well not be the fault of IronPython. (Certainly some event
>> handlers don't cause us problems whilst others do). On the other hand,
>> as IronPython *is* keeping track of all the handlers we add, an easy
>> interface to unhook them all would be extremely useful.
>>
>> Michael Foord
>>
>>     
>>> Check out this link http://www.bobpowell.net/eventsubscribers.htm
>>>       
>> (There
>>     
>>> are some additional steps for System.Windows.Forms.Component), and
>>> here's some sample code I wrote in C# that you could leverage:
>>>
>>>     [TestClass]
>>>     public class UnhookEventsTests
>>>     {
>>>         private class Item
>>>         {
>>>             public event EventHandler Clicked;
>>>
>>>             public void FireEvent()
>>>             {
>>>                 if(Clicked != null)
>>>                 {
>>>                     Clicked(this, EventArgs.Empty);
>>>                 }
>>>             }
>>>         }
>>>
>>>         [TestMethod]
>>>         public void UnhookEvents()
>>>         {
>>>             Item item = new Item();
>>>             item.Clicked += Clicked;
>>>             item.Clicked += Clicked2;
>>>
>>>             item.FireEvent();
>>>
>>>             Assert.AreEqual(2, callCount_);
>>>
>>>             FieldInfo field = typeof (Item).GetField("Clicked",
>>> BindingFlags.NonPublic | BindingFlags.Instance);
>>>             Delegate del = (Delegate) field.GetValue(item);
>>>             Delegate[] list = del.GetInvocationList();
>>>
>>>             Assert.AreEqual(2, list.Length);
>>>
>>>             //This works:
>>>             //foreach (Delegate d in list)
>>>             //{
>>>             //    item.Clicked -= (EventHandler) d;
>>>             //}
>>>
>>>             //This is more generic and avoids the cast to the
>>>       
>> specific
>>     
>>> handler type
>>>             EventInfo eventInfo = typeof (Item).GetEvent("Clicked");
>>>             foreach (Delegate d in list)
>>>             {
>>>                 eventInfo.RemoveEventHandler(item, d);
>>>             }
>>>
>>>             item.FireEvent();
>>>
>>>             Assert.AreEqual(2, callCount_);
>>>         }
>>>         private int callCount_;
>>>         private void Clicked(object sender, EventArgs e)
>>>         {
>>>             callCount_++;
>>>         }
>>>
>>>         private void Clicked2(object sender, EventArgs e)
>>>         {
>>>             callCount_++;
>>>         }
>>>     }
>>>
>>> -----Original Message-----
>>> From: users-bounces at lists.ironpython.com
>>> [mailto:users-bounces at lists.ironpython.com] On Behalf Of Kamil
>>> Dworakowski
>>> Sent: Saturday, November 29, 2008 1:16 PM
>>> To: Discussion of IronPython
>>> Subject: Re: [IronPython] Event unhooking and IPy2
>>>
>>> On Fri, Nov 28, 2008 at 11:39 AM, Michael Foord
>>> <fuzzyman at voidspace.org.uk> wrote:
>>>
>>>       
>>>> Just as a follow up to this - the reason for the hack below is that
>>>> subscribing to events from IronPython is still causing our UI
>>>>         
>> objects
>>     
>>> (and
>>>
>>>       
>>>> their whole object graphs) to not be garbage collected.
>>>>
>>>>         
>>> To be fair, we have no reason to belive it is IronPython's fault. The
>>> bug that event handlers would prevent garbage collection was fixed,
>>> and we were able to remove *some* of the event nuking.
>>> _______________________________________________
>>> Users mailing list
>>> Users at lists.ironpython.com
>>> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>>>
>>> =======
>>> Notice: This e-mail message, together with any attachments, contains
>>> information of Symyx Technologies, Inc. or any of its affiliates or
>>> subsidiaries that may be confidential, proprietary, copyrighted,
>>> privileged and/or protected work product, and is meant solely for
>>> the intended recipient. If you are not the intended recipient, and
>>> have received this message in error, please contact the sender
>>> immediately, permanently delete the original and any copies of this
>>> email and any attachments thereto.
>>> _______________________________________________
>>> Users mailing list
>>> Users at lists.ironpython.com
>>> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>>>
>>>       
>> --
>> http://www.ironpythoninaction.com/
>> http://www.voidspace.org.uk/blog
>>
>>
>> _______________________________________________
>> Users mailing list
>> Users at lists.ironpython.com
>> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>>     
>
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>   


-- 
http://www.ironpythoninaction.com/
http://www.voidspace.org.uk/blog




From dan.eloff at gmail.com  Wed Dec  3 00:09:48 2008
From: dan.eloff at gmail.com (Dan Eloff)
Date: Tue, 2 Dec 2008 18:09:48 -0500
Subject: [IronPython] Does IronPython 2.0 not recognize packages?
In-Reply-To: <4f428e7b0812020526i57123d5ct2e00ef41ccda1974@mail.gmail.com>
References: <4f428e7b0812020034o40dbb4a4g8cc39f0f764b976c@mail.gmail.com>
	<4f428e7b0812020526i57123d5ct2e00ef41ccda1974@mail.gmail.com>
Message-ID: <4817b6fc0812021509g181c6177o81678ee1fc79c71e@mail.gmail.com>

On Tue, Dec 2, 2008 at 8:26 AM, Yash Ganthe <yashgt at gmail.com> wrote:
> I still get the error:
>> Traceback (most recent call last):
>> File "Program.py", line 22, in Program.py
>> File "mscorlib", line unknown, in get_Item
>> File "mscorlib", line unknown, in ThrowKeyNotFoundException
>> KeyError: The given key was not present in the dictionary.
>

That's not an ImportError. What's happening on line 22 of Program.py ?

-Dan


From dan.eloff at gmail.com  Wed Dec  3 02:35:00 2008
From: dan.eloff at gmail.com (Dan Eloff)
Date: Tue, 2 Dec 2008 20:35:00 -0500
Subject: [IronPython] Data Binding with DLR Languages
Message-ID: <4817b6fc0812021735r18f57a0as3b516fe0ec7d8b69@mail.gmail.com>

I'm posting this here, for lack of a better place, but anyone using
dynamic languages on silverlight, really feels the lack of support for
ICustomTypeDescriptor in data binding (i.e. there isn't any) Just
trying to raise awareness of this *cough* Jimmy *cough*

-Dan

P.S. Harry Pierson wrote a great blog entry on data binding with
IronPython here:
http://devhawk.net/2008/11/18/IronPython+And+WPF+Part+3+Data+Binding.aspx


From deepali_abhyankar at persistent.co.in  Wed Dec  3 05:12:48 2008
From: deepali_abhyankar at persistent.co.in (Deepali Abhyankar)
Date: Wed, 3 Dec 2008 09:42:48 +0530
Subject: [IronPython] Logging module error
In-Reply-To: <5b0248170812020457n42ef8564kc6c00f02aa0b3137@mail.gmail.com>
References: <002c01c9546e$f1d6c610$d5845230$@co.in>
	<5b0248170812020457n42ef8564kc6c00f02aa0b3137@mail.gmail.com>
Message-ID: <002c01c954fd$6718d6f0$354a84d0$@co.in>

Thanks Seo
For Logging Module to work, I added the code in '__Init__.py', suggested at URL, given by you
Code added in __Init__.py file of Logging module 

if not hasattr(sys, "_getframe"):
   _srcfile = None
if sys.platform == 'cli':
   _srcfile = None

Thanks and Regards
Deepali

-----Original Message-----
From: Seo Sanghyeon [mailto:sanxiyn at gmail.com] 
Sent: Tuesday, December 02, 2008 6:27 PM
To: deepali_abhyankar at persistent.co.in; Discussion of IronPython
Subject: Re: [IronPython] Logging module error

2008/12/2 Deepali Abhyankar <deepali_abhyankar at persistent.co.in>:
> (logging and _getframe)
> Please suggest solution to resolve this

https://fepy.svn.sourceforge.net/svnroot/fepy/trunk/patches/stdlib/2.4.4/patch-stdlib-logging-getframe

-- 
Seo Sanghyeon


DISCLAIMER
==========
This e-mail may contain privileged and confidential information which is the property of Persistent Systems Ltd. It is intended only for the use of the individual or entity to which it is addressed. If you are not the intended recipient, you are not authorized to read, retain, copy, print, distribute or use this message. If you have received this communication in error, please notify the sender and delete all copies of this message. Persistent Systems Ltd. does not accept any liability for virus infected mails.


From Jimmy.Schementi at microsoft.com  Wed Dec  3 20:27:40 2008
From: Jimmy.Schementi at microsoft.com (Jimmy Schementi)
Date: Wed, 3 Dec 2008 11:27:40 -0800
Subject: [IronPython] Data Binding with DLR Languages
In-Reply-To: <4817b6fc0812021735r18f57a0as3b516fe0ec7d8b69@mail.gmail.com>
References: <4817b6fc0812021735r18f57a0as3b516fe0ec7d8b69@mail.gmail.com>
Message-ID: <5283CA0A4168DF4FBBD71AE9ECA5A3284BDAE88C2E@NA-EXMSG-C116.redmond.corp.microsoft.com>

http://www.codeplex.com/sdlsdk/WorkItem/View.aspx?WorkItemId=11844

Agreed that it needs to be done. Feel free to make the first couple stabs at it, but it is a hard problem, so not sure if you have time for it. I'm organizing the project a bit more to have a set of tangible projects people can start working on, and I'll make sure that data binding is on that list, and make sure someone is tackling it.

~js

> -----Original Message-----
> From: users-bounces at lists.ironpython.com [mailto:users-
> bounces at lists.ironpython.com] On Behalf Of Dan Eloff
> Sent: Tuesday, December 02, 2008 5:35 PM
> To: Discussion of IronPython
> Subject: [IronPython] Data Binding with DLR Languages
>
> I'm posting this here, for lack of a better place, but anyone using
> dynamic languages on silverlight, really feels the lack of support for
> ICustomTypeDescriptor in data binding (i.e. there isn't any) Just
> trying to raise awareness of this *cough* Jimmy *cough*
>
> -Dan
>
> P.S. Harry Pierson wrote a great blog entry on data binding with
> IronPython here:
> http://devhawk.net/2008/11/18/IronPython+And+WPF+Part+3+Data+Binding.as
> px
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com



From dan.eloff at gmail.com  Wed Dec  3 21:24:41 2008
From: dan.eloff at gmail.com (Dan Eloff)
Date: Wed, 3 Dec 2008 15:24:41 -0500
Subject: [IronPython] Data Binding with DLR Languages
In-Reply-To: <5283CA0A4168DF4FBBD71AE9ECA5A3284BDAE88C2E@NA-EXMSG-C116.redmond.corp.microsoft.com>
References: <4817b6fc0812021735r18f57a0as3b516fe0ec7d8b69@mail.gmail.com>
	<5283CA0A4168DF4FBBD71AE9ECA5A3284BDAE88C2E@NA-EXMSG-C116.redmond.corp.microsoft.com>
Message-ID: <4817b6fc0812031224y12326d39nd9c70fe271b11c47@mail.gmail.com>

On Wed, Dec 3, 2008 at 2:27 PM, Jimmy Schementi
<Jimmy.Schementi at microsoft.com> wrote:
> http://www.codeplex.com/sdlsdk/WorkItem/View.aspx?WorkItemId=11844
>
> Agreed that it needs to be done. Feel free to make the first couple stabs at it, but it is a hard problem, so not sure if you have time for it. I'm organizing the project a bit more to have a set of tangible projects people can start working on, and I'll make sure that data binding is on that list, and make sure someone is tackling it.
>

I wish I did, but I'm unable to keep up with my current commitments.
By the time that changes, this will hopefully no longer be an issue.
But I'm sure there will be plenty of other things in dynamic
silverlight for me to contribute to. I am encouraged by your response,
thank you for taking this issue seriously.

-Dan


From steve at mcneel.com  Thu Dec  4 01:46:07 2008
From: steve at mcneel.com (Steve Baer)
Date: Wed, 3 Dec 2008 16:46:07 -0800
Subject: [IronPython] Parsing scripts
Message-ID: <1ACE609827CC4A139000F3923F9846F2@mcneel.com>

I would like to be able to add some sort of "intellisense" to a text editor for IronPython scripts. Something along the lines of reporting the available functions in a module after hitting a period.

import clr
clr. <- (show something here)

Are there utilities available in IronPython for analyzing a partial script to figure this out and if so could you point me in the right direction?

Thanks,
-Steve
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081203/749f86fd/attachment.html>

From dinov at microsoft.com  Thu Dec  4 01:49:55 2008
From: dinov at microsoft.com (Dino Viehland)
Date: Wed, 3 Dec 2008 16:49:55 -0800
Subject: [IronPython] Parsing scripts
In-Reply-To: <1ACE609827CC4A139000F3923F9846F2@mcneel.com>
References: <1ACE609827CC4A139000F3923F9846F2@mcneel.com>
Message-ID: <350E7D38B6D819428718949920EC2355564A70AA07@NA-EXMSG-C102.redmond.corp.microsoft.com>

The closest we have to this currently is what the console does w/ live objects.  I think you could also find some examples of this in IronPython Studio (http://www.codeplex.com/IronPythonStudio) but there's nothing that's too sophisticated right now.

From: users-bounces at lists.ironpython.com [mailto:users-bounces at lists.ironpython.com] On Behalf Of Steve Baer
Sent: Wednesday, December 03, 2008 4:46 PM
To: Discussion of IronPython
Subject: [IronPython] Parsing scripts

I would like to be able to add some sort of "intellisense" to a text editor for IronPython scripts. Something along the lines of reporting the available functions in a module after hitting a period.

import clr
clr. <- (show something here)

Are there utilities available in IronPython for analyzing a partial script to figure this out and if so could you point me in the right direction?

Thanks,
-Steve
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081203/21c1c0dc/attachment.html>

From dfugate at microsoft.com  Thu Dec  4 02:12:13 2008
From: dfugate at microsoft.com (Dave Fugate)
Date: Wed, 3 Dec 2008 17:12:13 -0800
Subject: [IronPython] Blocker: ProcessDialogKey on TextBox subclass
In-Reply-To: <2679f6510811280500s55133e6fv7156b5f04aaac8af@mail.gmail.com>
References: <2679f6510811280500s55133e6fv7156b5f04aaac8af@mail.gmail.com>
Message-ID: <EECFB552375FDE49A23A13CE85EEC2FD1E97FC6B75@NA-EXMSG-C112.redmond.corp.microsoft.com>

I think the following might be a workaround for this:
import clr
clr.AddReference('System.Windows.Forms')
clr.AddReference('IronPython')

from System.Windows.Forms import Form, Application, TextBox

class MyTextBox(TextBox):
    ProcessDialogKeyCopy = TextBox.ProcessDialogKey

def ProcessDialogKey(s, key):
    print key
    return MyTextBox.ProcessDialogKeyCopy(s, key)

MyTextBox.ProcessDialogKey = ProcessDialogKey

class ThisApp(Form):
    def __init__(self):
        Form.__init__(self)
        tb = MyTextBox()
        self.Controls.Add(tb)
        Application.Run(self)


def main():
    f = ThisApp()


if __name__ == '__main__':
    main()

From: users-bounces at lists.ironpython.com [mailto:users-bounces at lists.ironpython.com] On Behalf Of Glenn Jones
Sent: Friday, November 28, 2008 5:01 AM
To: Discussion of IronPython
Subject: [IronPython] Blocker: ProcessDialogKey on TextBox subclass

Hi All,

Here is a little app that illustrates an issue that we've just discovered:
---------------------------------------------------------------------------------------------------
---------------------------------------------------------------------------------------------------
import clr
clr.AddReference('System.Windows.Forms')
clr.AddReference('IronPython')

from System.Windows.Forms import Form, Application, TextBox

class MyTextBox(TextBox):
    def ProcessDialogKey(self, key):
        print key
        #super(MyTextBox, self).ProcessDialogKey(key) # Fails with a stack overflow
        #TextBox.ProcessDialogKey(self, key) # Fails with: Microsoft.Scripting.ArgumentTypeException: cannot access protected member ProcessDialogKey without a python subclass of TextBoxBase

class ThisApp(Form):

    def __init__(self):
        Form.__init__(self)
        tb = MyTextBox()
        self.Controls.Add(tb)
        Application.Run(self)


def main():
    f = ThisApp()


if __name__ == '__main__':
    main()
---------------------------------------------------------------------------------------------------
---------------------------------------------------------------------------------------------------

Uncommenting either of the two commented lines produces the results on that line when a key is pressed in the text box. The second method of calling ProcessDialogKey worked in IPy 1.

This is likely to block our progress on porting Resolver One to IPy2 unless there is a different way of accomplishing the same thing.

Thanks
Glenn & Will
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081203/b73585c2/attachment.html>

From mtraudt at alum.mit.edu  Thu Dec  4 02:48:50 2008
From: mtraudt at alum.mit.edu (Mark Traudt)
Date: Wed, 3 Dec 2008 17:48:50 -0800 (PST)
Subject: [IronPython] Problem inheriting from C# abstract base class with
 overloaded constructors
Message-ID: <20804327.post@talk.nabble.com>


If I create an abstract base class in C# that has overloaded constructors,
and then subclass from this in IronPython, then I am unable to call the
non-default base class constructor from the __new__ method in the subclass. 
If I try, I get the result shown after the code sample.

Interestingly, if I change the base class to not be abstract and to have
public constructors (both changes are required) then my IronPython script
works fine.  

Is this a bug?  If not, then am I doing something wrong, or is this not
supported for some reason?

I reproduced this with IronPython-1.1.2 and IronPython-2.0B5.


*** OneOff.cs ***

namespace Test
{
             public abstract class Foo
	{
		protected Foo()	
		{}

		protected Foo(int a)
		{ 
			a_ = a; 
		} 

		public override string ToString()	
		{	
			return "Foo: " + a_;
		}

		private int a_;
	}
}


*** OneOff.py ***

import clr
clr.AddReferenceToFile("OneOff.dll")
from Test import *

class Bar(Foo):
   def __new__ (cls, a=0):
        return Foo.__new__(cls, a)

print Bar()
print Bar(42)


*** Results ***

Traceback (most recent call last):
  File C:\dev\OneOff\IronPython\OneOff.py, line 10, in
  File C:\dev\OneOff\IronPython\OneOff.py, line 8, in
  File , line 0, in DefaultNew##15
TypeError: default __new__ does not take parameters


-- 
View this message in context: http://www.nabble.com/Problem-inheriting-from-C--abstract-base-class-with-overloaded-constructors-tp20804327p20804327.html
Sent from the IronPython mailing list archive at Nabble.com.



From dan.eloff at gmail.com  Thu Dec  4 03:07:36 2008
From: dan.eloff at gmail.com (Dan Eloff)
Date: Wed, 3 Dec 2008 21:07:36 -0500
Subject: [IronPython] SilverShell 0.6 Released
Message-ID: <4817b6fc0812031807k2e75ba82i3fcec27dc7231137@mail.gmail.com>

A blatant plug for my own software here, but I just uploaded
SilverShell 0.6.0: http://code.google.com/p/silvershell/

New in this release is the ability to run on the desktop with WPF,
execution of code in background threads, and a scratchpad canvas for
playing with UI controls.

There are still some bugs lurking in the intellisense code, and I'll
be following soon with a bugfix release. But it works, and you can try
it out. I find it more user friendly than the command prompt.

-Dan


From dinov at microsoft.com  Thu Dec  4 03:19:38 2008
From: dinov at microsoft.com (Dino Viehland)
Date: Wed, 3 Dec 2008 18:19:38 -0800
Subject: [IronPython] Problem inheriting from C# abstract base class
 with overloaded constructors
In-Reply-To: <20804327.post@talk.nabble.com>
References: <20804327.post@talk.nabble.com>
Message-ID: <350E7D38B6D819428718949920EC2355564A70AA3E@NA-EXMSG-C102.redmond.corp.microsoft.com>

This is bug #20021 - http://www.codeplex.com/IronPython/WorkItem/View.aspx?WorkItemId=20021

You should be able to declare the ctor public.  It shouldn't be an error to do that but you'll get an FxCop warning if you run FxCop.

I checked a fix for this in today into the Main branch (which is 2.1).  I think that should show up on CodePlex tonight or tomorrow if you're willing to build IronPython from source.  If not we'll definitely fix this in 2.0.1 if we don't spin a new build for 2.0 which would also likely include the fix.

> -----Original Message-----
> From: users-bounces at lists.ironpython.com [mailto:users-
> bounces at lists.ironpython.com] On Behalf Of Mark Traudt
> Sent: Wednesday, December 03, 2008 5:49 PM
> To: users at lists.ironpython.com
> Subject: [IronPython] Problem inheriting from C# abstract base class
> with overloaded constructors
>
>
> If I create an abstract base class in C# that has overloaded
> constructors,
> and then subclass from this in IronPython, then I am unable to call the
> non-default base class constructor from the __new__ method in the
> subclass.
> If I try, I get the result shown after the code sample.
>
> Interestingly, if I change the base class to not be abstract and to
> have
> public constructors (both changes are required) then my IronPython
> script
> works fine.
>
> Is this a bug?  If not, then am I doing something wrong, or is this not
> supported for some reason?
>
> I reproduced this with IronPython-1.1.2 and IronPython-2.0B5.
>
>
> *** OneOff.cs ***
>
> namespace Test
> {
>              public abstract class Foo
>         {
>                 protected Foo()
>                 {}
>
>                 protected Foo(int a)
>                 {
>                         a_ = a;
>                 }
>
>                 public override string ToString()
>                 {
>                         return "Foo: " + a_;
>                 }
>
>                 private int a_;
>         }
> }
>
>
> *** OneOff.py ***
>
> import clr
> clr.AddReferenceToFile("OneOff.dll")
> from Test import *
>
> class Bar(Foo):
>    def __new__ (cls, a=0):
>         return Foo.__new__(cls, a)
>
> print Bar()
> print Bar(42)
>
>
> *** Results ***
>
> Traceback (most recent call last):
>   File C:\dev\OneOff\IronPython\OneOff.py, line 10, in
>   File C:\dev\OneOff\IronPython\OneOff.py, line 8, in
>   File , line 0, in DefaultNew##15
> TypeError: default __new__ does not take parameters
>
>
> --
> View this message in context: http://www.nabble.com/Problem-inheriting-
> from-C--abstract-base-class-with-overloaded-constructors-
> tp20804327p20804327.html
> Sent from the IronPython mailing list archive at Nabble.com.
>
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com



From dan.eloff at gmail.com  Thu Dec  4 04:00:01 2008
From: dan.eloff at gmail.com (Dan Eloff)
Date: Wed, 3 Dec 2008 22:00:01 -0500
Subject: [IronPython] Problem inheriting from C# abstract base class
	with overloaded constructors
In-Reply-To: <350E7D38B6D819428718949920EC2355564A70AA3E@NA-EXMSG-C102.redmond.corp.microsoft.com>
References: <20804327.post@talk.nabble.com>
	<350E7D38B6D819428718949920EC2355564A70AA3E@NA-EXMSG-C102.redmond.corp.microsoft.com>
Message-ID: <4817b6fc0812031900o41935f9akb34a2d52752e4417@mail.gmail.com>

On Wed, Dec 3, 2008 at 9:19 PM, Dino Viehland <dinov at microsoft.com> wrote:
> This is bug #20021 - http://www.codeplex.com/IronPython/WorkItem/View.aspx?WorkItemId=20021
>
> I checked a fix for this in today into the Main branch (which is 2.1).  I think that should show up on CodePlex tonight or tomorrow if you're willing to build IronPython from source.

Sweet. Will definitely grab those bits. Thanks for fixing it.

-Dan


From kfarmer at thuban.org  Thu Dec  4 05:17:43 2008
From: kfarmer at thuban.org (Keith J. Farmer)
Date: Wed, 3 Dec 2008 20:17:43 -0800
Subject: [IronPython] SilverShell 0.6 Released
In-Reply-To: <4817b6fc0812031807k2e75ba82i3fcec27dc7231137@mail.gmail.com>
References: <4817b6fc0812031807k2e75ba82i3fcec27dc7231137@mail.gmail.com>
Message-ID: <EEDCD954DD4A0D488BBDBDDCC260754466FB6A@meta-exchange.circle-cross.home>

Neat..

I wonder how difficult it would be to be able to host IP, PS, etc, much like the Silverlight/DLR demo from last year.  I imagine for DLR languages that's probably not terribly difficult, but I don't know how hard it would be to wrap a wrapper around the Powershell interpreter to make it behave similarly.

-----Original Message-----
From: users-bounces at lists.ironpython.com [mailto:users-bounces at lists.ironpython.com] On Behalf Of Dan Eloff
Sent: Wednesday, December 03, 2008 6:08 PM
To: Discussion of IronPython
Subject: [IronPython] SilverShell 0.6 Released

A blatant plug for my own software here, but I just uploaded
SilverShell 0.6.0: http://code.google.com/p/silvershell/

New in this release is the ability to run on the desktop with WPF,
execution of code in background threads, and a scratchpad canvas for
playing with UI controls.

There are still some bugs lurking in the intellisense code, and I'll
be following soon with a bugfix release. But it works, and you can try
it out. I find it more user friendly than the command prompt.

-Dan
_______________________________________________
Users mailing list
Users at lists.ironpython.com
http://lists.ironpython.com/listinfo.cgi/users-ironpython.com

From oripel at gmail.com  Thu Dec  4 07:57:18 2008
From: oripel at gmail.com (orip)
Date: Wed, 3 Dec 2008 22:57:18 -0800 (PST)
Subject: [IronPython] Parsing scripts
In-Reply-To: <350E7D38B6D819428718949920EC2355564A70AA07@NA-EXMSG-C102.redmond.corp.microsoft.com>
References: <1ACE609827CC4A139000F3923F9846F2@mcneel.com>
	<350E7D38B6D819428718949920EC2355564A70AA07@NA-EXMSG-C102.redmond.corp.microsoft.com>
Message-ID: <94036fb2-94ca-4256-bf93-7531fbfdcaeb@3g2000yqs.googlegroups.com>

PySmell by Orestis Markou does completion through static analysis of
Python code.
I haven't tried it with IronPython but it could work, and from what I
gather Orestis is happy to help.
http://code.google.com/p/pysmell/

On Dec 4, 2:49?am, Dino Viehland <di... at microsoft.com> wrote:
> The closest we have to this currently is what the console does w/ live objects. ?I think you could also find some examples of this in IronPython Studio (http://www.codeplex.com/IronPythonStudio) but there's nothing that's too sophisticated right now.
>
> From: users-boun... at lists.ironpython.com [mailto:users-boun... at lists.ironpython.com] On Behalf Of Steve Baer
> Sent: Wednesday, December 03, 2008 4:46 PM
> To: Discussion of IronPython
> Subject: [IronPython] Parsing scripts
>
> I would like to be able to add some sort of "intellisense" to a text editor for IronPython scripts. Something along the lines of reporting the available functions in a module after hitting a period.
>
> import clr
> clr. <- (show something here)
>
> Are there utilities available in IronPython for analyzing a partial script to figure this out and if so could you point me in the right direction?
>
> Thanks,
> -Steve
>
> _______________________________________________
> Users mailing list
> Us... at lists.ironpython.comhttp://lists.ironpython.com/listinfo.cgi/users-ironpython.com


From orestis at resolversystems.com  Thu Dec  4 11:45:17 2008
From: orestis at resolversystems.com (Orestis Markou)
Date: Thu, 04 Dec 2008 10:45:17 +0000
Subject: [IronPython] Parsing scripts
In-Reply-To: <350E7D38B6D819428718949920EC2355564A70AA07@NA-EXMSG-C102.redmond.corp.microsoft.com>
References: <1ACE609827CC4A139000F3923F9846F2@mcneel.com>
	<350E7D38B6D819428718949920EC2355564A70AA07@NA-EXMSG-C102.redmond.corp.microsoft.com>
Message-ID: <4937B4BD.9080301@resolversystems.com>

I've written PySmell, but it does static analysis rather than dynamic. 
That means that it won't handle code that lives in dlls, but it will 
handle any scripts of your own.

I'll add support for dynamic analysis soon, though - it's easy to do.

Have a look at http://pypi.python.org/pypi/pysmell

Dino Viehland wrote:
> The closest we have to this currently is what the console does w/ live 
> objects.  I think you could also find some examples of this in 
> IronPython Studio (http://www.codeplex.com/IronPythonStudio) but there?s 
> nothing that?s too sophisticated right now.
> 
>  
> 
> *From:* users-bounces at lists.ironpython.com 
> [mailto:users-bounces at lists.ironpython.com] *On Behalf Of *Steve Baer
> *Sent:* Wednesday, December 03, 2008 4:46 PM
> *To:* Discussion of IronPython
> *Subject:* [IronPython] Parsing scripts
> 
>  
> 
> I would like to be able to add some sort of "intellisense" to a text 
> editor for IronPython scripts. Something along the lines of reporting 
> the available functions in a module after hitting a period.
> 
>  
> 
> import clr
> 
> clr. <- (show something here)
> 
>  
> 
> Are there utilities available in IronPython for analyzing a partial 
> script to figure this out and if so could you point me in the right 
> direction?
> 
>  
> 
> Thanks,
> 
> -Steve
> 
> 
> ------------------------------------------------------------------------
> 
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com

-- 
Orestis Markou
Software Engineer,
Resolver Systems Ltd.
orestis at resolversystems.com
+44 (0) 20 7253 6372



From fuzzyman at voidspace.org.uk  Thu Dec  4 12:36:37 2008
From: fuzzyman at voidspace.org.uk (Michael Foord)
Date: Thu, 04 Dec 2008 11:36:37 +0000
Subject: [IronPython] SilverShell 0.6 Released
In-Reply-To: <4817b6fc0812031807k2e75ba82i3fcec27dc7231137@mail.gmail.com>
References: <4817b6fc0812031807k2e75ba82i3fcec27dc7231137@mail.gmail.com>
Message-ID: <4937C0C5.50801@voidspace.org.uk>

Dan Eloff wrote:
> A blatant plug for my own software here, but I just uploaded
> SilverShell 0.6.0: http://code.google.com/p/silvershell/
>
> New in this release is the ability to run on the desktop with WPF,
> execution of code in background threads, and a scratchpad canvas for
> playing with UI controls.
>
> There are still some bugs lurking in the intellisense code, and I'll
> be following soon with a bugfix release. But it works, and you can try
> it out. I find it more user friendly than the command prompt.
>   

Very nice - congratulations.

Michael

> -Dan
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>   


-- 
http://www.ironpythoninaction.com/



From sh at defuze.org  Thu Dec  4 20:17:53 2008
From: sh at defuze.org (Sylvain Hellegouarch)
Date: Thu, 04 Dec 2008 20:17:53 +0100
Subject: [IronPython] Don't let the IronPython IRC channel die
Message-ID: <49382CE1.4080208@defuze.org>

Hi folks,

There's been a IRC channel about IronPython for some time now but it's 
usually really quiet. I've seen friendly people and interesting 
discussions but too sparse to make them a lively channel.

Maybe you're not aware of that channel but everyone is welcome. I think 
it'd be quite important if IronPython developers could popup sometimes 
too. IRC channels can be a great way to build a community as well as a 
good resource for casual developers.

So please come and say hi at:

Network: freenode
Channel: #ironpython

Thanks,
- Sylvain

More information on IRC if you need: http://en.wikipedia.org/wiki/IRC



From jslutter at reactorzero.com  Thu Dec  4 22:13:45 2008
From: jslutter at reactorzero.com (Jeff Slutter)
Date: Thu, 04 Dec 2008 16:13:45 -0500
Subject: [IronPython] Calling functions in IronPython that don't really exist
Message-ID: <49384809.6030709@reactorzero.com>

I'm writing an application that I want to provide runtime scripting via
IronPython for. This application loads "plug-ins" from assemblies to add
functionality.

Each plug-in is a class with, for the sake of simplicity, a "Do" function.

What I'm trying to figure out is that if I have a class like:

class AddNumbers
{
    public string Do( params object[] args )
    {
        ..check args..
        ..add the two arguments..
        ..return the result as a string..
    }
}


That I could call it from IronPython like:

AddNumbers( 3, 4 )

When that happens a new instance of the AddNumbers class is created, the
arguments are then passed to the Do member function, the result is
returned back to Python.

This may seem like a weird way for scripting. But, doing it this way I
can do some other important things, such as keeping a history of the
commands. If, along with Do, the commands have Undo and Redo functions,
then I can keep the instances of the commands stored and should the user
need to undo their command, I can internally run the Undo function.

Any ideas on how I should approach this? I don't mind having to do a
good chunk of backend work so that to add new command plugins, I just
have to write a class with a Do function and possibly provide some
Attributes if necessary. If I have to make some concessions that is
fine, but I need to stick with the concept of creating an instance of a
class and executing a function to perform the operation.

I assume I'm either going to have to generate some sort of thunk
function (I have no idea where to start) or provide some sort of
'command runner' that really gets called.

Any help or guidance is mighty appreciated,
Jeff


From nytrokiss at gmail.com  Thu Dec  4 22:17:23 2008
From: nytrokiss at gmail.com (James Matthews)
Date: Thu, 4 Dec 2008 23:17:23 +0200
Subject: [IronPython] Calling functions in IronPython that don't really
	exist
In-Reply-To: <49384809.6030709@reactorzero.com>
References: <49384809.6030709@reactorzero.com>
Message-ID: <8a6b8e350812041317m59120aa6mb9de91ac27d671f7@mail.gmail.com>

You can just create the function name and do nothing with it. In Cpython i
would use meta classes but idk if they are in IronPython

On Thu, Dec 4, 2008 at 11:13 PM, Jeff Slutter <jslutter at reactorzero.com>wrote:

> I'm writing an application that I want to provide runtime scripting via
> IronPython for. This application loads "plug-ins" from assemblies to add
> functionality.
>
> Each plug-in is a class with, for the sake of simplicity, a "Do" function.
>
> What I'm trying to figure out is that if I have a class like:
>
> class AddNumbers
> {
>    public string Do( params object[] args )
>    {
>        ..check args..
>        ..add the two arguments..
>        ..return the result as a string..
>    }
> }
>
>
> That I could call it from IronPython like:
>
> AddNumbers( 3, 4 )
>
> When that happens a new instance of the AddNumbers class is created, the
> arguments are then passed to the Do member function, the result is
> returned back to Python.
>
> This may seem like a weird way for scripting. But, doing it this way I
> can do some other important things, such as keeping a history of the
> commands. If, along with Do, the commands have Undo and Redo functions,
> then I can keep the instances of the commands stored and should the user
> need to undo their command, I can internally run the Undo function.
>
> Any ideas on how I should approach this? I don't mind having to do a
> good chunk of backend work so that to add new command plugins, I just
> have to write a class with a Do function and possibly provide some
> Attributes if necessary. If I have to make some concessions that is
> fine, but I need to stick with the concept of creating an instance of a
> class and executing a function to perform the operation.
>
> I assume I'm either going to have to generate some sort of thunk
> function (I have no idea where to start) or provide some sort of
> 'command runner' that really gets called.
>
> Any help or guidance is mighty appreciated,
> Jeff
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>



-- 
http://www.astorandblack.com/

http://www.thewatcherys.com/
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081204/023dde94/attachment.html>

From dinov at microsoft.com  Thu Dec  4 22:20:07 2008
From: dinov at microsoft.com (Dino Viehland)
Date: Thu, 4 Dec 2008 13:20:07 -0800
Subject: [IronPython] Calling functions in IronPython that don't really
 exist
In-Reply-To: <49384809.6030709@reactorzero.com>
References: <49384809.6030709@reactorzero.com>
Message-ID: <350E7D38B6D819428718949920EC2355564A7FAEA8@NA-EXMSG-C102.redmond.corp.microsoft.com>

Do you mean you'd call it like "AddNumbers().Do(3, 4)"?

This is really easy.  Make AddNumbers public, then do:

import clr
clr.AddReference('MyAssembly')
import AddNumbers
AddNumbers().Do(3, 4)

If you really want to do AddNumbers(3, 4) then you'd just write it as:

public class AddNumbers
{
    public AddNumbers(params object[] args) {
         Do(args);
    }
    public string Do( params object[] args )
    {
        ..check args..
        ..add the two arguments..
        ..return the result as a string..
    }
}

And do the same:

import clr
clr.AddReference('MyAssembly')
import AddNumbers
AddNumbers(3, 4)

> -----Original Message-----
> From: users-bounces at lists.ironpython.com [mailto:users-
> bounces at lists.ironpython.com] On Behalf Of Jeff Slutter
> Sent: Thursday, December 04, 2008 1:14 PM
> To: Discussion of IronPython
> Subject: [IronPython] Calling functions in IronPython that don't really
> exist
>
> I'm writing an application that I want to provide runtime scripting via
> IronPython for. This application loads "plug-ins" from assemblies to
> add
> functionality.
>
> Each plug-in is a class with, for the sake of simplicity, a "Do"
> function.
>
> What I'm trying to figure out is that if I have a class like:
>
> class AddNumbers
> {
>     public string Do( params object[] args )
>     {
>         ..check args..
>         ..add the two arguments..
>         ..return the result as a string..
>     }
> }
>
>
> That I could call it from IronPython like:
>
> AddNumbers( 3, 4 )
>
> When that happens a new instance of the AddNumbers class is created,
> the
> arguments are then passed to the Do member function, the result is
> returned back to Python.
>
> This may seem like a weird way for scripting. But, doing it this way I
> can do some other important things, such as keeping a history of the
> commands. If, along with Do, the commands have Undo and Redo functions,
> then I can keep the instances of the commands stored and should the
> user
> need to undo their command, I can internally run the Undo function.
>
> Any ideas on how I should approach this? I don't mind having to do a
> good chunk of backend work so that to add new command plugins, I just
> have to write a class with a Do function and possibly provide some
> Attributes if necessary. If I have to make some concessions that is
> fine, but I need to stick with the concept of creating an instance of a
> class and executing a function to perform the operation.
>
> I assume I'm either going to have to generate some sort of thunk
> function (I have no idea where to start) or provide some sort of
> 'command runner' that really gets called.
>
> Any help or guidance is mighty appreciated,
> Jeff
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com


From jslutter at reactorzero.com  Thu Dec  4 22:30:24 2008
From: jslutter at reactorzero.com (Jeff Slutter)
Date: Thu, 04 Dec 2008 16:30:24 -0500
Subject: [IronPython] Calling functions in IronPython that don't really
 exist
In-Reply-To: <350E7D38B6D819428718949920EC2355564A7FAEA8@NA-EXMSG-C102.redmond.corp.microsoft.com>
References: <49384809.6030709@reactorzero.com>
	<350E7D38B6D819428718949920EC2355564A7FAEA8@NA-EXMSG-C102.redmond.corp.microsoft.com>
Message-ID: <49384BF0.4080001@reactorzero.com>

Ok, now I feel a little stupid. Thanks :)

I think it would be wise to drop the "params object[] args" and go with
just actually specifying the parameters to the constructor so I can take
advantage of type checking, etc. from Python.

In the constructor I'll have to save the instance of the class created
to my internal History list (for the undo, redo).

Thanks again!

Dino Viehland wrote:
> Do you mean you'd call it like "AddNumbers().Do(3, 4)"?
>
> This is really easy.  Make AddNumbers public, then do:
>
> import clr
> clr.AddReference('MyAssembly')
> import AddNumbers
> AddNumbers().Do(3, 4)
>
> If you really want to do AddNumbers(3, 4) then you'd just write it as:
>
> public class AddNumbers
> {
>     public AddNumbers(params object[] args) {
>          Do(args);
>     }
>     public string Do( params object[] args )
>     {
>         ..check args..
>         ..add the two arguments..
>         ..return the result as a string..
>     }
> }
>
> And do the same:
>
> import clr
> clr.AddReference('MyAssembly')
> import AddNumbers
> AddNumbers(3, 4)
>
>   
>


From jslutter at reactorzero.com  Thu Dec  4 22:59:24 2008
From: jslutter at reactorzero.com (Jeff Slutter)
Date: Thu, 04 Dec 2008 16:59:24 -0500
Subject: [IronPython] Calling functions in IronPython that don't really
 exist
In-Reply-To: <350E7D38B6D819428718949920EC2355564A7FAEA8@NA-EXMSG-C102.redmond.corp.microsoft.com>
References: <49384809.6030709@reactorzero.com>
	<350E7D38B6D819428718949920EC2355564A7FAEA8@NA-EXMSG-C102.redmond.corp.microsoft.com>
Message-ID: <493852BC.5030705@reactorzero.com>

Actually this doesn't exactly work. If I want the command to be able to
return a value, I can't go with the constructor route:

res = AddNumbers( 3, 5 );

you want res equal to "8", but really it is an instance of my AddNumbers
class. I don't want the script writers to be able to save this class
instance, and it would be 'ugly' if they had to call ".GetReturn()" to
get the return value of the command.

And I don't want the call to "Do" to be explicit.

I sound really picky...


Dino Viehland wrote:
> Do you mean you'd call it like "AddNumbers().Do(3, 4)"?
>
> This is really easy.  Make AddNumbers public, then do:
>
> import clr
> clr.AddReference('MyAssembly')
> import AddNumbers
> AddNumbers().Do(3, 4)
>
> If you really want to do AddNumbers(3, 4) then you'd just write it as:
>
> public class AddNumbers
> {
>     public AddNumbers(params object[] args) {
>          Do(args);
>     }
>     public string Do( params object[] args )
>     {
>         ..check args..
>         ..add the two arguments..
>         ..return the result as a string..
>     }
> }
>
> And do the same:
>
> import clr
> clr.AddReference('MyAssembly')
> import AddNumbers
> AddNumbers(3, 4)
>


From dinov at microsoft.com  Thu Dec  4 23:12:38 2008
From: dinov at microsoft.com (Dino Viehland)
Date: Thu, 4 Dec 2008 14:12:38 -0800
Subject: [IronPython] Calling functions in IronPython that don't really
 exist
In-Reply-To: <493852BC.5030705@reactorzero.com>
References: <49384809.6030709@reactorzero.com>
	<350E7D38B6D819428718949920EC2355564A7FAEA8@NA-EXMSG-C102.redmond.corp.microsoft.com>
	<493852BC.5030705@reactorzero.com>
Message-ID: <350E7D38B6D819428718949920EC2355564A7FAF2C@NA-EXMSG-C102.redmond.corp.microsoft.com>

You can just import static functions instead.  Something like:

public static class ScriptHelpers {
        public static object AddNumbers(params object[] args) {
                return 42;
        }
}

Add the reference and then:

from ScriptHelpers import *

If you really need to construct an instance for some reason you could do it inside of AddNumbers and do the .GetReturn there as well.

> -----Original Message-----
> From: users-bounces at lists.ironpython.com [mailto:users-
> bounces at lists.ironpython.com] On Behalf Of Jeff Slutter
> Sent: Thursday, December 04, 2008 1:59 PM
> To: Discussion of IronPython
> Subject: Re: [IronPython] Calling functions in IronPython that don't
> really exist
>
> Actually this doesn't exactly work. If I want the command to be able to
> return a value, I can't go with the constructor route:
>
> res = AddNumbers( 3, 5 );
>
> you want res equal to "8", but really it is an instance of my
> AddNumbers
> class. I don't want the script writers to be able to save this class
> instance, and it would be 'ugly' if they had to call ".GetReturn()" to
> get the return value of the command.
>
> And I don't want the call to "Do" to be explicit.
>
> I sound really picky...
>
>
> Dino Viehland wrote:
> > Do you mean you'd call it like "AddNumbers().Do(3, 4)"?
> >
> > This is really easy.  Make AddNumbers public, then do:
> >
> > import clr
> > clr.AddReference('MyAssembly')
> > import AddNumbers
> > AddNumbers().Do(3, 4)
> >
> > If you really want to do AddNumbers(3, 4) then you'd just write it
> as:
> >
> > public class AddNumbers
> > {
> >     public AddNumbers(params object[] args) {
> >          Do(args);
> >     }
> >     public string Do( params object[] args )
> >     {
> >         ..check args..
> >         ..add the two arguments..
> >         ..return the result as a string..
> >     }
> > }
> >
> > And do the same:
> >
> > import clr
> > clr.AddReference('MyAssembly')
> > import AddNumbers
> > AddNumbers(3, 4)
> >
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com


From jdhardy at gmail.com  Fri Dec  5 02:57:12 2008
From: jdhardy at gmail.com (Jeff Hardy)
Date: Thu, 4 Dec 2008 18:57:12 -0700
Subject: [IronPython] NWSGI 0.7 released
Message-ID: <ea89d0e50812041757o4d5833c7qf647fd38d37b26d6@mail.gmail.com>

NWSGI 0.7 is now available. This version adds support for ASP.NET
sessions, fixes Unicode handling, and adds support for wildcard
mappings. It is available at
http://www.codeplex.com/NWSGI/Release/ProjectReleases.aspx?ReleaseId=20189.

What is NWSGI?
NWSGI is an ASP.NET HttpHandler that implements the Python WSGI spec
(PEP 333 - http://www.python.org/dev/peps/pep-0333/) for interfacing
web servers and web frameworks using IronPython. There are still some
IronPYthon bugs that prevent everything from working, but some things
do.

For more details on this release, see
http://jdhardy.blogspot.com/2008/12/nwsgi-07-released.html.

- Jeff


From dr.addn at gmail.com  Fri Dec  5 04:26:36 2008
From: dr.addn at gmail.com (Ph.T)
Date: Thu, 4 Dec 2008 20:26:36 -0700
Subject: [IronPython] SilverShell 0.6 Released
In-Reply-To: <4817b6fc0812031807k2e75ba82i3fcec27dc7231137@mail.gmail.com>
References: <4817b6fc0812031807k2e75ba82i3fcec27dc7231137@mail.gmail.com>
Message-ID: <8fd4a2fe0812041926w6aa3c735g3d01a2c7c836c88c@mail.gmail.com>

. I was disappointed to find on code.google
something that could be holding off on submitting source code
to some time in the indefinite future . gmail put this in the spam bucket .
. dll's and other microsoft magic should be confined to vmware or some other
virtual mach .

On Wed, Dec 3, 2008 at 7:07 PM, Dan Eloff <dan.eloff at gmail.com> wrote:

> A blatant plug for my own software here, but I just uploaded
> SilverShell 0.6.0: http://code.google.com/p/silvershell/
>

-- 
Americium Dream Documents
"(real opportunity starts with real documentation)
http://amerdreamdocs.tripod.com/ http://www.angelfire.com/psy/dr.addn/
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081204/3f94ddbb/attachment.html>

From dan.eloff at gmail.com  Fri Dec  5 05:37:52 2008
From: dan.eloff at gmail.com (Dan Eloff)
Date: Thu, 4 Dec 2008 23:37:52 -0500
Subject: [IronPython] SilverShell 0.6 Released
In-Reply-To: <8fd4a2fe0812041926w6aa3c735g3d01a2c7c836c88c@mail.gmail.com>
References: <4817b6fc0812031807k2e75ba82i3fcec27dc7231137@mail.gmail.com>
	<8fd4a2fe0812041926w6aa3c735g3d01a2c7c836c88c@mail.gmail.com>
Message-ID: <4817b6fc0812042037h137d5badmbdb4596cb6e6b565@mail.gmail.com>

On Thu, Dec 4, 2008 at 10:26 PM, Ph. T <dr.addn at gmail.com> wrote:
> . I was disappointed to find on code.google
> something that could be holding off on submitting source code
> to some time in the indefinite future . gmail put this in the spam bucket .
> . dll's and other microsoft magic should be confined to vmware or some other
> virtual mach .
>

I'm sorry for that, I just wanted to get the release out there, it
only occurred to me later that the .dll will restrict people from
running on versions of IronPython different from the one I packaged
with it. I will include the source for the _wpf.dll in the 0.6.1 bug
fix release which will be available within a week.

-Dan


From jslutter at reactorzero.com  Fri Dec  5 06:32:20 2008
From: jslutter at reactorzero.com (Jeff Slutter)
Date: Fri, 05 Dec 2008 00:32:20 -0500
Subject: [IronPython] Calling functions in IronPython that don't really
 exist
In-Reply-To: <350E7D38B6D819428718949920EC2355564A7FAF2C@NA-EXMSG-C102.redmond.corp.microsoft.com>
References: <49384809.6030709@reactorzero.com>	<350E7D38B6D819428718949920EC2355564A7FAEA8@NA-EXMSG-C102.redmond.corp.microsoft.com>	<493852BC.5030705@reactorzero.com>
	<350E7D38B6D819428718949920EC2355564A7FAF2C@NA-EXMSG-C102.redmond.corp.microsoft.com>
Message-ID: <4938BCE4.7090705@reactorzero.com>

I have found another way that better suits what I'm trying to do (and
control).

In a nutshell, I just add the function to the builtins. Here is the code:

// some member function, etc.
delegate int DoDelegate();
private int DoFunc()
{
  return 10;
}

//... in the IronPython initialization, after the scripting engine has
been created ...

ScriptScope builtins = Python.GetBuiltinModule(m_engine);
builtins.SetVariable("SomeTestFunction", new DoDelegate(DoFunc));




now, in the Python scripts I can call:

SomeTestFunction();


Simple problem and solution, but I just wanted a way to dynamically add
new functions to Python with names that I control.



From deepali_abhyankar at persistent.co.in  Fri Dec  5 07:08:17 2008
From: deepali_abhyankar at persistent.co.in (Deepali Abhyankar)
Date: Fri, 5 Dec 2008 11:38:17 +0530
Subject: [IronPython] About Python TextTestRunner
In-Reply-To: <ea89d0e50812041757o4d5833c7qf647fd38d37b26d6@mail.gmail.com>
References: <ea89d0e50812041757o4d5833c7qf647fd38d37b26d6@mail.gmail.com>
Message-ID: <002201c9569f$ddff61a0$99fe24e0$@co.in>

Hi

I used following code to write result on different media.

unittest.TextTestRunner(verbosity=2).run(suite) : Writes result to StdOutput
unittest.TextTestRunner(file,verbosity=2).run(suite) : Writes result to file


But what should I do to write test result on file as well as StdOutput, on
single go?
Is there any other method to do this?

Regards
Deepali


-----Original Message-----
From: users-bounces at lists.ironpython.com
[mailto:users-bounces at lists.ironpython.com] On Behalf Of Jeff Hardy
Sent: Friday, December 05, 2008 7:27 AM
To: Discussion of IronPython; web-sig at python.org
Subject: [IronPython] NWSGI 0.7 released

NWSGI 0.7 is now available. This version adds support for ASP.NET
sessions, fixes Unicode handling, and adds support for wildcard
mappings. It is available at
http://www.codeplex.com/NWSGI/Release/ProjectReleases.aspx?ReleaseId=20189.

What is NWSGI?
NWSGI is an ASP.NET HttpHandler that implements the Python WSGI spec
(PEP 333 - http://www.python.org/dev/peps/pep-0333/) for interfacing
web servers and web frameworks using IronPython. There are still some
IronPYthon bugs that prevent everything from working, but some things
do.

For more details on this release, see
http://jdhardy.blogspot.com/2008/12/nwsgi-07-released.html.

- Jeff
_______________________________________________
Users mailing list
Users at lists.ironpython.com
http://lists.ironpython.com/listinfo.cgi/users-ironpython.com


DISCLAIMER
==========
This e-mail may contain privileged and confidential information which is the property of Persistent Systems Ltd. It is intended only for the use of the individual or entity to which it is addressed. If you are not the intended recipient, you are not authorized to read, retain, copy, print, distribute or use this message. If you have received this communication in error, please notify the sender and delete all copies of this message. Persistent Systems Ltd. does not accept any liability for virus infected mails.


From miha.valencic at gmail.com  Fri Dec  5 09:11:24 2008
From: miha.valencic at gmail.com (Miha Valencic)
Date: Fri, 5 Dec 2008 09:11:24 +0100
Subject: [IronPython] NWSGI 0.7 released
In-Reply-To: <ea89d0e50812041757o4d5833c7qf647fd38d37b26d6@mail.gmail.com>
References: <ea89d0e50812041757o4d5833c7qf647fd38d37b26d6@mail.gmail.com>
Message-ID: <233dfa1d0812050011v1ac776a8q72f82853f18ea353@mail.gmail.com>

Jeff, do you have any info on whether Django works with this setup?

Thanks,
 Miha.

2008/12/5 Jeff Hardy <jdhardy at gmail.com>

> NWSGI 0.7 is now available. This version adds support for ASP.NET
> sessions, fixes Unicode handling, and adds support for wildcard
> mappings. It is available at
> http://www.codeplex.com/NWSGI/Release/ProjectReleases.aspx?ReleaseId=20189
> .
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081205/04ea8008/attachment.html>

From orestis at resolversystems.com  Fri Dec  5 14:22:51 2008
From: orestis at resolversystems.com (Orestis Markou)
Date: Fri, 05 Dec 2008 13:22:51 +0000
Subject: [IronPython] About Python TextTestRunner
In-Reply-To: <002201c9569f$ddff61a0$99fe24e0$@co.in>
References: <ea89d0e50812041757o4d5833c7qf647fd38d37b26d6@mail.gmail.com>
	<002201c9569f$ddff61a0$99fe24e0$@co.in>
Message-ID: <49392B2B.3070107@resolversystems.com>

You can either use pipe the stdout through tee to save it in a file too,
or in a more pythonic way define a file-like object that prints to 
stdout as you write to it:

import sys
class PrintStream(object):
     def __init__(self, file):
         self.out = file
     def write(self, b):
         self.out.write(b)
         sys.stdout.write(b)


unittest.TextTestRunner(PrintStream(file),verbosity=2).run(suite)

Hope that helped,
Orestis


Deepali Abhyankar wrote:
> Hi
> 
> I used following code to write result on different media.
> 
> unittest.TextTestRunner(verbosity=2).run(suite) : Writes result to StdOutput
> unittest.TextTestRunner(file,verbosity=2).run(suite) : Writes result to file
> 
> 
> But what should I do to write test result on file as well as StdOutput, on
> single go?
> Is there any other method to do this?
> 
> Regards
> Deepali
> 
> 
> -----Original Message-----
> From: users-bounces at lists.ironpython.com
> [mailto:users-bounces at lists.ironpython.com] On Behalf Of Jeff Hardy
> Sent: Friday, December 05, 2008 7:27 AM
> To: Discussion of IronPython; web-sig at python.org
> Subject: [IronPython] NWSGI 0.7 released
> 
> NWSGI 0.7 is now available. This version adds support for ASP.NET
> sessions, fixes Unicode handling, and adds support for wildcard
> mappings. It is available at
> http://www.codeplex.com/NWSGI/Release/ProjectReleases.aspx?ReleaseId=20189.
> 
> What is NWSGI?
> NWSGI is an ASP.NET HttpHandler that implements the Python WSGI spec
> (PEP 333 - http://www.python.org/dev/peps/pep-0333/) for interfacing
> web servers and web frameworks using IronPython. There are still some
> IronPYthon bugs that prevent everything from working, but some things
> do.
> 
> For more details on this release, see
> http://jdhardy.blogspot.com/2008/12/nwsgi-07-released.html.
> 
> - Jeff
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
> 
> 
> DISCLAIMER
> ==========
> This e-mail may contain privileged and confidential information which is the property of Persistent Systems Ltd. It is intended only for the use of the individual or entity to which it is addressed. If you are not the intended recipient, you are not authorized to read, retain, copy, print, distribute or use this message. If you have received this communication in error, please notify the sender and delete all copies of this message. Persistent Systems Ltd. does not accept any liability for virus infected mails.
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com

-- 
Orestis Markou
Software Engineer,
Resolver Systems Ltd.
orestis at resolversystems.com
+44 (0) 20 7253 6372



From giulio.petrucci at gmail.com  Fri Dec  5 15:11:55 2008
From: giulio.petrucci at gmail.com (Giulio Petrucci)
Date: Fri, 5 Dec 2008 15:11:55 +0100
Subject: [IronPython] Newbie Questions
Message-ID: <c138ac70812050611i7d96d027vcd0c03c79896e12@mail.gmail.com>

Hi there,

I've just started playing with IronPython and I have two (maybe)
simple questions:
1) I have a .NET library with an enum E and a class C. The class C
expose a method accepting E as an argument. How can I do this (I mean:
using enums) in IronPython?
2) I have a .NET library which reads some settings from the .config
file. I'd like to test it interactively using the IronPython
interactive console. How can I "load" the right .config file?

Thanks in advance and... see you here or on the #ironpython freenode channel!!!
-- 
Ciao,
Giulio

-- 
OnAir:
http://www.giuliopetrucci.it
http://www.fujikomonamour.com


From dinov at microsoft.com  Fri Dec  5 18:47:26 2008
From: dinov at microsoft.com (Dino Viehland)
Date: Fri, 5 Dec 2008 09:47:26 -0800
Subject: [IronPython] Newbie Questions
In-Reply-To: <c138ac70812050611i7d96d027vcd0c03c79896e12@mail.gmail.com>
References: <c138ac70812050611i7d96d027vcd0c03c79896e12@mail.gmail.com>
Message-ID: <350E7D38B6D819428718949920EC2355564A7FB258@NA-EXMSG-C102.redmond.corp.microsoft.com>

You can add a reference to the assembly and then import both the class and the enum.  For example:

import clr
clr.AddReference('MyLibrary')
from MyLibrary import MyClass, MyEnum
MyClass().SomeMethod(MyEnum.SomeValue)

The 2nd question... is this an app.exe.config file that you're reading?  If so this might be helpful:

http://www.devnewsgroups.net/group/microsoft.public.dotnet.framework.windowsforms/topic64732.aspx


-----Original Message-----
From: users-bounces at lists.ironpython.com [mailto:users-bounces at lists.ironpython.com] On Behalf Of Giulio Petrucci
Sent: Friday, December 05, 2008 6:12 AM
To: users at lists.ironpython.com
Subject: [IronPython] Newbie Questions

Hi there,

I've just started playing with IronPython and I have two (maybe)
simple questions:
1) I have a .NET library with an enum E and a class C. The class C
expose a method accepting E as an argument. How can I do this (I mean:
using enums) in IronPython?
2) I have a .NET library which reads some settings from the .config
file. I'd like to test it interactively using the IronPython
interactive console. How can I "load" the right .config file?

Thanks in advance and... see you here or on the #ironpython freenode channel!!!
--
Ciao,
Giulio

--
OnAir:
http://www.giuliopetrucci.it
http://www.fujikomonamour.com
_______________________________________________
Users mailing list
Users at lists.ironpython.com
http://lists.ironpython.com/listinfo.cgi/users-ironpython.com


From dan.eloff at gmail.com  Fri Dec  5 20:53:49 2008
From: dan.eloff at gmail.com (Dan Eloff)
Date: Fri, 5 Dec 2008 14:53:49 -0500
Subject: [IronPython] NWSGI 0.7 released
In-Reply-To: <233dfa1d0812050011v1ac776a8q72f82853f18ea353@mail.gmail.com>
References: <ea89d0e50812041757o4d5833c7qf647fd38d37b26d6@mail.gmail.com>
	<233dfa1d0812050011v1ac776a8q72f82853f18ea353@mail.gmail.com>
Message-ID: <4817b6fc0812051153r2d607e86k4941db8e87f3892f@mail.gmail.com>

Django depends on unicode and str being two different creatures, so I
imagine not so well on IronPython. However once we have bytes objects
and Django is updated for python 3, then I think the story will
improve.

-Dan

On Fri, Dec 5, 2008 at 3:11 AM, Miha Valencic <miha.valencic at gmail.com> wrote:
> Jeff, do you have any info on whether Django works with this setup?
>
> Thanks,
>  Miha.
>
> 2008/12/5 Jeff Hardy <jdhardy at gmail.com>
>>
>> NWSGI 0.7 is now available. This version adds support for ASP.NET
>> sessions, fixes Unicode handling, and adds support for wildcard
>> mappings. It is available at
>>
>> http://www.codeplex.com/NWSGI/Release/ProjectReleases.aspx?ReleaseId=20189.
>>
>
>
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>
>


From carolj at microsoft.com  Sat Dec  6 00:02:10 2008
From: carolj at microsoft.com (Carolyn Johnston (MSNAR))
Date: Fri, 5 Dec 2008 15:02:10 -0800
Subject: [IronPython] can't import namespace from assembly
Message-ID: <387DA5F5806B6A49A69468E4EA24DB2A7225C75918@NA-EXMSG-C125.redmond.corp.microsoft.com>

I have a dll that I've built in VS2008, C# (and I even strongly named it): Normalizer.dll. It contains a namespace (StringUtilities) which contains a single class (MaxNormalizer).

I can't import the StringUtilities namespace:

>>> ass = Assembly.Load("Normalizer.dll")
>>> ass
<Assembly Normalizer, Version=1.0.0.0, Culture=neutral, PublicKeyToken=0707cddf2bb66849>
>>> clr.AddReference(ass)
>>> import StringUtilities
Traceback (most recent call last):
  File , line 0, in <stdin>##229
  File , line 0, in __import__##7
ImportError: No module named StringUtilities

However you can see that "ass" does contain the StringUtilities namespace (I've included the listNamespaces code at bottom of this email):

>>> ass.GetTypes()
System.Type[](<System.RuntimeType object at 0x000000000000002E [StringUtilities.MaxNormalizer]>)

I also found a help page from Dino here (http://lists.ironpython.com/pipermail/users-ironpython.com/2007-April/004792.html) that implied StringUtilities ought to be an attribute of the assembly itself, so I tried that too:

>>> ass.StringUtilities
Traceback (most recent call last):
  File , line 0, in <stdin>##231
AttributeError: assembly Normalizer has no type StringUtilities

And then I made one last desperate shot at it:

>>> from Normalizer import StringUtilities
Traceback (most recent call last):
  File , line 0, in <stdin>##232
  File , line 0, in __import__##7
ImportError: No module named Normalizer
>>>

Any clues as to why this namespace can't be imported? I'm stumped.

Thanks,
Carolyn
---
Carolyn Johnston (carolj)
Lead Researcher, Local Search, Virtual Earth
Bldg 116/2276 | MIcrosoft Corporation
425-706-2153


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081205/3c9eff15/attachment.html>

From fuzzyman at voidspace.org.uk  Sat Dec  6 00:05:00 2008
From: fuzzyman at voidspace.org.uk (Michael Foord)
Date: Fri, 05 Dec 2008 23:05:00 +0000
Subject: [IronPython] can't import namespace from assembly
In-Reply-To: <387DA5F5806B6A49A69468E4EA24DB2A7225C75918@NA-EXMSG-C125.redmond.corp.microsoft.com>
References: <387DA5F5806B6A49A69468E4EA24DB2A7225C75918@NA-EXMSG-C125.redmond.corp.microsoft.com>
Message-ID: <4939B39C.6070105@voidspace.org.uk>

In the past I've had this error when I've forgotten to make the class 
public. Possibly not the problem, but you never know. :-)

Michael Foord

Carolyn Johnston (MSNAR) wrote:
>
> I have a dll that I?ve built in VS2008, C# (and I even strongly named 
> it): Normalizer.dll. It contains a namespace (StringUtilities) which 
> contains a single class (MaxNormalizer).
>
> I can?t import the StringUtilities namespace:
>
> >>> ass = Assembly.Load("Normalizer.dll")
>
> >>> ass
>
> <Assembly Normalizer, Version=1.0.0.0, Culture=neutral, 
> PublicKeyToken=0707cddf2bb66849>
>
> >>> clr.AddReference(ass)
>
> >>> import StringUtilities
>
> Traceback (most recent call last):
>
> File , line 0, in <stdin>##229
>
> File , line 0, in __import__##7
>
> ImportError: No module named StringUtilities
>
> However you can see that ?ass? does contain the StringUtilities 
> namespace (I?ve included the listNamespaces code at bottom of this email):
>
> >>> ass.GetTypes()
>
> System.Type[](<System.RuntimeType object at 0x000000000000002E 
> [StringUtilities.MaxNormalizer]>)
>
> I also found a help page from Dino here 
> (http://lists.ironpython.com/pipermail/users-ironpython.com/2007-April/004792.html) 
> that implied StringUtilities ought to be an attribute of the assembly 
> itself, so I tried that too:
>
> >>> ass.StringUtilities
>
> Traceback (most recent call last):
>
> File , line 0, in <stdin>##231
>
> AttributeError: assembly Normalizer has no type StringUtilities
>
> And then I made one last desperate shot at it:
>
> >>> from Normalizer import StringUtilities
>
> Traceback (most recent call last):
>
> File , line 0, in <stdin>##232
>
> File , line 0, in __import__##7
>
> ImportError: No module named Normalizer
>
> >>>
>
> Any clues as to why this namespace can?t be imported? I?m stumped.
>
> Thanks,
>
> Carolyn
>
> *---
> Carolyn Johnston (carolj)*
> Lead Researcher, Local Search, Virtual Earth
> Bldg 116/2276 | MIcrosoft Corporation
> 425-706-2153**
>
> ------------------------------------------------------------------------
>
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>   


-- 
http://www.ironpythoninaction.com/
http://www.voidspace.org.uk/blog




From Rodney.Howeedy at Sun.COM  Sat Dec  6 00:11:03 2008
From: Rodney.Howeedy at Sun.COM (Rodney Howeedy)
Date: Fri, 05 Dec 2008 16:11:03 -0700
Subject: [IronPython] 'DispPropertyGet' error returned from read-write COM
	interop property
Message-ID: <4939B507.4050104@sun.com>

Hi everyone,

I encounter an error with 2.0 RC2 that didn't occur in the same 
IronPython code on 2.0 Beta 3. 

I use a factory object from a COM interop assembly that has a read-write 
property named "Field".  I get a TypeError when I assign a value to this 
property (read-write):
 >>> mytestRun.Field['RN_USER_01'] = 'abc'
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
TypeError: 'DispPropertyGet' object is unsubscriptable

 >>> print mytestRun.Field['RN_USER_01']
None

I imported the same COM interop assembly that worked with 2.0 Beta3.  
The interop was converted with tlbimp.exe from a COM API .dll.  Here's 
the 2.0 Beta3 output:
 >>> mytestRun.Field['RN_USER_01'] = 'abc'
 >>> print mytestRun.Field['RN_USER_01']
abc

My best guess is that IronPython 2.0 RC2 interprets the property as 
read-only even though the COM interop assembly defines the property 
read-write. The error occurs in IronPython 2.0 RC2 with every COM 
interop object with the read-write "Field" property.

I can read the default assigned value but cannot assign anything.  Any 
ideas?  I'm fresh out of them after trying many permutations of the syntax.

Thank you,
Rodney Howeedy
Staff Engineer, Sun Microsystems


From carolj at microsoft.com  Sat Dec  6 00:11:06 2008
From: carolj at microsoft.com (Carolyn Johnston (MSNAR))
Date: Fri, 5 Dec 2008 15:11:06 -0800
Subject: [IronPython] can't import namespace from assembly
In-Reply-To: <4939B39C.6070105@voidspace.org.uk>
References: <387DA5F5806B6A49A69468E4EA24DB2A7225C75918@NA-EXMSG-C125.redmond.corp.microsoft.com>
	<4939B39C.6070105@voidspace.org.uk>
Message-ID: <387DA5F5806B6A49A69468E4EA24DB2A7225C75934@NA-EXMSG-C125.redmond.corp.microsoft.com>

Thank you, that was the problem.

:) Carolyn

-----Original Message-----
From: users-bounces at lists.ironpython.com [mailto:users-bounces at lists.ironpython.com] On Behalf Of Michael Foord
Sent: Friday, December 05, 2008 3:05 PM
To: Discussion of IronPython
Subject: Re: [IronPython] can't import namespace from assembly

In the past I've had this error when I've forgotten to make the class
public. Possibly not the problem, but you never know. :-)

Michael Foord

Carolyn Johnston (MSNAR) wrote:
>
> I have a dll that I've built in VS2008, C# (and I even strongly named
> it): Normalizer.dll. It contains a namespace (StringUtilities) which
> contains a single class (MaxNormalizer).
>
> I can't import the StringUtilities namespace:
>
> >>> ass = Assembly.Load("Normalizer.dll")
>
> >>> ass
>
> <Assembly Normalizer, Version=1.0.0.0, Culture=neutral,
> PublicKeyToken=0707cddf2bb66849>
>
> >>> clr.AddReference(ass)
>
> >>> import StringUtilities
>
> Traceback (most recent call last):
>
> File , line 0, in <stdin>##229
>
> File , line 0, in __import__##7
>
> ImportError: No module named StringUtilities
>
> However you can see that "ass" does contain the StringUtilities
> namespace (I've included the listNamespaces code at bottom of this email):
>
> >>> ass.GetTypes()
>
> System.Type[](<System.RuntimeType object at 0x000000000000002E
> [StringUtilities.MaxNormalizer]>)
>
> I also found a help page from Dino here
> (http://lists.ironpython.com/pipermail/users-ironpython.com/2007-April/004792.html)
> that implied StringUtilities ought to be an attribute of the assembly
> itself, so I tried that too:
>
> >>> ass.StringUtilities
>
> Traceback (most recent call last):
>
> File , line 0, in <stdin>##231
>
> AttributeError: assembly Normalizer has no type StringUtilities
>
> And then I made one last desperate shot at it:
>
> >>> from Normalizer import StringUtilities
>
> Traceback (most recent call last):
>
> File , line 0, in <stdin>##232
>
> File , line 0, in __import__##7
>
> ImportError: No module named Normalizer
>
> >>>
>
> Any clues as to why this namespace can't be imported? I'm stumped.
>
> Thanks,
>
> Carolyn
>
> *---
> Carolyn Johnston (carolj)*
> Lead Researcher, Local Search, Virtual Earth
> Bldg 116/2276 | MIcrosoft Corporation
> 425-706-2153**
>
> ------------------------------------------------------------------------
>
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>


--
http://www.ironpythoninaction.com/
http://www.voidspace.org.uk/blog


_______________________________________________
Users mailing list
Users at lists.ironpython.com
http://lists.ironpython.com/listinfo.cgi/users-ironpython.com



From dinov at microsoft.com  Sat Dec  6 00:12:12 2008
From: dinov at microsoft.com (Dino Viehland)
Date: Fri, 5 Dec 2008 15:12:12 -0800
Subject: [IronPython] 'DispPropertyGet' error returned from read-write
 COM	interop property
In-Reply-To: <4939B507.4050104@sun.com>
References: <4939B507.4050104@sun.com>
Message-ID: <350E7D38B6D819428718949920EC2355564A7FB4CC@NA-EXMSG-C102.redmond.corp.microsoft.com>

Does:

mytestRun.Field.Item['RN_USER_01'] = 'abc'

work?

-----Original Message-----
From: users-bounces at lists.ironpython.com [mailto:users-bounces at lists.ironpython.com] On Behalf Of Rodney Howeedy
Sent: Friday, December 05, 2008 3:11 PM
To: users at lists.ironpython.com
Subject: [IronPython] 'DispPropertyGet' error returned from read-write COM interop property

Hi everyone,

I encounter an error with 2.0 RC2 that didn't occur in the same
IronPython code on 2.0 Beta 3.

I use a factory object from a COM interop assembly that has a read-write
property named "Field".  I get a TypeError when I assign a value to this
property (read-write):
 >>> mytestRun.Field['RN_USER_01'] = 'abc'
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
TypeError: 'DispPropertyGet' object is unsubscriptable

 >>> print mytestRun.Field['RN_USER_01']
None

I imported the same COM interop assembly that worked with 2.0 Beta3.
The interop was converted with tlbimp.exe from a COM API .dll.  Here's
the 2.0 Beta3 output:
 >>> mytestRun.Field['RN_USER_01'] = 'abc'
 >>> print mytestRun.Field['RN_USER_01']
abc

My best guess is that IronPython 2.0 RC2 interprets the property as
read-only even though the COM interop assembly defines the property
read-write. The error occurs in IronPython 2.0 RC2 with every COM
interop object with the read-write "Field" property.

I can read the default assigned value but cannot assign anything.  Any
ideas?  I'm fresh out of them after trying many permutations of the syntax.

Thank you,
Rodney Howeedy
Staff Engineer, Sun Microsystems
_______________________________________________
Users mailing list
Users at lists.ironpython.com
http://lists.ironpython.com/listinfo.cgi/users-ironpython.com


From Rodney.Howeedy at Sun.COM  Sat Dec  6 00:21:32 2008
From: Rodney.Howeedy at Sun.COM (Rodney Howeedy)
Date: Fri, 05 Dec 2008 16:21:32 -0700
Subject: [IronPython] 'DispPropertyGet' error returned from read-write
	COM	interop property
In-Reply-To: <350E7D38B6D819428718949920EC2355564A7FB4CC@NA-EXMSG-C102.redmond.corp.microsoft.com>
References: <4939B507.4050104@sun.com>
	<350E7D38B6D819428718949920EC2355564A7FB4CC@NA-EXMSG-C102.redmond.corp.microsoft.com>
Message-ID: <4939B77C.2060603@sun.com>

I found that one and tried before.  I didn't have any luck:
 >>> mytestRun.Field.Item['RN_USER_01'] = 'abc'
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
SystemError: Object reference not set to an instance of an object.

 >>> print mytestRun.Field.Item['RN_USER_01']
None

I couldn't decipher the error so I couldn't figure out what I'm doing 
wrong. 
I suspect the folks on this list can. =)

Dino Viehland wrote:
> Does:
>
> mytestRun.Field.Item['RN_USER_01'] = 'abc'
>
> work?
>
> -----Original Message-----
> From: users-bounces at lists.ironpython.com [mailto:users-bounces at lists.ironpython.com] On Behalf Of Rodney Howeedy
> Sent: Friday, December 05, 2008 3:11 PM
> To: users at lists.ironpython.com
> Subject: [IronPython] 'DispPropertyGet' error returned from read-write COM interop property
>
> Hi everyone,
>
> I encounter an error with 2.0 RC2 that didn't occur in the same
> IronPython code on 2.0 Beta 3.
>
> I use a factory object from a COM interop assembly that has a read-write
> property named "Field".  I get a TypeError when I assign a value to this
> property (read-write):
>  >>> mytestRun.Field['RN_USER_01'] = 'abc'
> Traceback (most recent call last):
>   File "<stdin>", line 1, in <module>
> TypeError: 'DispPropertyGet' object is unsubscriptable
>
>  >>> print mytestRun.Field['RN_USER_01']
> None
>
> I imported the same COM interop assembly that worked with 2.0 Beta3.
> The interop was converted with tlbimp.exe from a COM API .dll.  Here's
> the 2.0 Beta3 output:
>  >>> mytestRun.Field['RN_USER_01'] = 'abc'
>  >>> print mytestRun.Field['RN_USER_01']
> abc
>
> My best guess is that IronPython 2.0 RC2 interprets the property as
> read-only even though the COM interop assembly defines the property
> read-write. The error occurs in IronPython 2.0 RC2 with every COM
> interop object with the read-write "Field" property.
>
> I can read the default assigned value but cannot assign anything.  Any
> ideas?  I'm fresh out of them after trying many permutations of the syntax.
>
> Thank you,
> Rodney Howeedy
> Staff Engineer, Sun Microsystems
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>   


From dinov at microsoft.com  Sat Dec  6 00:28:49 2008
From: dinov at microsoft.com (Dino Viehland)
Date: Fri, 5 Dec 2008 15:28:49 -0800
Subject: [IronPython] 'DispPropertyGet' error returned from
	read-write	COM	interop property
In-Reply-To: <4939B77C.2060603@sun.com>
References: <4939B507.4050104@sun.com>
	<350E7D38B6D819428718949920EC2355564A7FB4CC@NA-EXMSG-C102.redmond.corp.microsoft.com>
	<4939B77C.2060603@sun.com>
Message-ID: <350E7D38B6D819428718949920EC2355564A7FB4DD@NA-EXMSG-C102.redmond.corp.microsoft.com>

What's the stack trace for the object reference exception if you run with the -X:ExceptionDetail option?

-----Original Message-----
From: users-bounces at lists.ironpython.com [mailto:users-bounces at lists.ironpython.com] On Behalf Of Rodney Howeedy
Sent: Friday, December 05, 2008 3:22 PM
To: Discussion of IronPython
Subject: Re: [IronPython] 'DispPropertyGet' error returned from read-write COM interop property

I found that one and tried before.  I didn't have any luck:
 >>> mytestRun.Field.Item['RN_USER_01'] = 'abc'
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
SystemError: Object reference not set to an instance of an object.

 >>> print mytestRun.Field.Item['RN_USER_01']
None

I couldn't decipher the error so I couldn't figure out what I'm doing
wrong.
I suspect the folks on this list can. =)

Dino Viehland wrote:
> Does:
>
> mytestRun.Field.Item['RN_USER_01'] = 'abc'
>
> work?
>
> -----Original Message-----
> From: users-bounces at lists.ironpython.com [mailto:users-bounces at lists.ironpython.com] On Behalf Of Rodney Howeedy
> Sent: Friday, December 05, 2008 3:11 PM
> To: users at lists.ironpython.com
> Subject: [IronPython] 'DispPropertyGet' error returned from read-write COM interop property
>
> Hi everyone,
>
> I encounter an error with 2.0 RC2 that didn't occur in the same
> IronPython code on 2.0 Beta 3.
>
> I use a factory object from a COM interop assembly that has a read-write
> property named "Field".  I get a TypeError when I assign a value to this
> property (read-write):
>  >>> mytestRun.Field['RN_USER_01'] = 'abc'
> Traceback (most recent call last):
>   File "<stdin>", line 1, in <module>
> TypeError: 'DispPropertyGet' object is unsubscriptable
>
>  >>> print mytestRun.Field['RN_USER_01']
> None
>
> I imported the same COM interop assembly that worked with 2.0 Beta3.
> The interop was converted with tlbimp.exe from a COM API .dll.  Here's
> the 2.0 Beta3 output:
>  >>> mytestRun.Field['RN_USER_01'] = 'abc'
>  >>> print mytestRun.Field['RN_USER_01']
> abc
>
> My best guess is that IronPython 2.0 RC2 interprets the property as
> read-only even though the COM interop assembly defines the property
> read-write. The error occurs in IronPython 2.0 RC2 with every COM
> interop object with the read-write "Field" property.
>
> I can read the default assigned value but cannot assign anything.  Any
> ideas?  I'm fresh out of them after trying many permutations of the syntax.
>
> Thank you,
> Rodney Howeedy
> Staff Engineer, Sun Microsystems
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>
_______________________________________________
Users mailing list
Users at lists.ironpython.com
http://lists.ironpython.com/listinfo.cgi/users-ironpython.com



From Rodney.Howeedy at Sun.COM  Sat Dec  6 00:34:51 2008
From: Rodney.Howeedy at Sun.COM (Rodney Howeedy)
Date: Fri, 05 Dec 2008 16:34:51 -0700
Subject: [IronPython] 'DispPropertyGet' error returned from	read-write
	COM	interop property
In-Reply-To: <350E7D38B6D819428718949920EC2355564A7FB4DD@NA-EXMSG-C102.redmond.corp.microsoft.com>
References: <4939B507.4050104@sun.com>
	<350E7D38B6D819428718949920EC2355564A7FB4CC@NA-EXMSG-C102.redmond.corp.microsoft.com>
	<4939B77C.2060603@sun.com>
	<350E7D38B6D819428718949920EC2355564A7FB4DD@NA-EXMSG-C102.redmond.corp.microsoft.com>
Message-ID: <4939BA9B.3070701@sun.com>

Stack trace for the statement that raises the SytemError:
 >>> mytestRun.Field.Item['RN_USER_01'] = 'abc'
Object reference not set to an instance of an object.
   at 
IronPython.Runtime.Types.ReflectedGetterSetter.CallTarget(CodeContext 
context, SiteLocalStorage`1 storage, MethodInfo[] targets, Object 
instance, Object[] args)
   at 
IronPython.Runtime.Types.ReflectedGetterSetter.CallSetter(CodeContext 
context, SiteLocalStorage`1 storage, Objectinstance, Object[] args, 
Object value)
   at IronPython.Runtime.Types.ReflectedIndexer.SetValue(CodeContext 
context, SiteLocalStorage`1 storage, Object[] keys, Object value)
   at 
IronPython.Runtime.Types.ReflectedIndexer.set_Item(SiteLocalStorage`1 
storage, Object[] key, Object value)
   at _stub_$1209##669(Closure , CallSite , Object , String , String )
   at 
Microsoft.Scripting.Actions.MatchCaller.Call3[T0,T1,T2,TRet](Func`5 
target, CallSite site, Object[] args)
   at Microsoft.Scripting.Actions.CallSite`1.UpdateAndExecute(Object[] args)
   at 
Microsoft.Scripting.Actions.UpdateDelegates.Update3[T,T0,T1,T2,TRet](CallSite 
site, T0 arg0, T1 arg1, T2 arg2)
   at <module>$1203##663(Closure , Scope , LanguageContext )
   at 
Microsoft.Scripting.Runtime.OptimizedScriptCode.InvokeTarget(LambdaExpression 
code, Scope scope)
   at Microsoft.Scripting.ScriptCode.Run(Scope scope)
   at 
IronPython.Hosting.PythonCommandLine.<>c__DisplayClass1.<RunOneInteraction>b__0()
SystemError: Object reference not set to an instance of an object.

Dino Viehland wrote:
> What's the stack trace for the object reference exception if you run with the -X:ExceptionDetail option?
>
> -----Original Message-----
> From: users-bounces at lists.ironpython.com [mailto:users-bounces at lists.ironpython.com] On Behalf Of Rodney Howeedy
> Sent: Friday, December 05, 2008 3:22 PM
> To: Discussion of IronPython
> Subject: Re: [IronPython] 'DispPropertyGet' error returned from read-write COM interop property
>
> I found that one and tried before.  I didn't have any luck:
>  >>> mytestRun.Field.Item['RN_USER_01'] = 'abc'
> Traceback (most recent call last):
>   File "<stdin>", line 1, in <module>
> SystemError: Object reference not set to an instance of an object.
>
>  >>> print mytestRun.Field.Item['RN_USER_01']
> None
>
> I couldn't decipher the error so I couldn't figure out what I'm doing
> wrong.
> I suspect the folks on this list can. =)
>
> Dino Viehland wrote:
>   
>> Does:
>>
>> mytestRun.Field.Item['RN_USER_01'] = 'abc'
>>
>> work?
>>
>> -----Original Message-----
>> From: users-bounces at lists.ironpython.com [mailto:users-bounces at lists.ironpython.com] On Behalf Of Rodney Howeedy
>> Sent: Friday, December 05, 2008 3:11 PM
>> To: users at lists.ironpython.com
>> Subject: [IronPython] 'DispPropertyGet' error returned from read-write COM interop property
>>
>> Hi everyone,
>>
>> I encounter an error with 2.0 RC2 that didn't occur in the same
>> IronPython code on 2.0 Beta 3.
>>
>> I use a factory object from a COM interop assembly that has a read-write
>> property named "Field".  I get a TypeError when I assign a value to this
>> property (read-write):
>>  >>> mytestRun.Field['RN_USER_01'] = 'abc'
>> Traceback (most recent call last):
>>   File "<stdin>", line 1, in <module>
>> TypeError: 'DispPropertyGet' object is unsubscriptable
>>
>>  >>> print mytestRun.Field['RN_USER_01']
>> None
>>
>> I imported the same COM interop assembly that worked with 2.0 Beta3.
>> The interop was converted with tlbimp.exe from a COM API .dll.  Here's
>> the 2.0 Beta3 output:
>>  >>> mytestRun.Field['RN_USER_01'] = 'abc'
>>  >>> print mytestRun.Field['RN_USER_01']
>> abc
>>
>> My best guess is that IronPython 2.0 RC2 interprets the property as
>> read-only even though the COM interop assembly defines the property
>> read-write. The error occurs in IronPython 2.0 RC2 with every COM
>> interop object with the read-write "Field" property.
>>
>> I can read the default assigned value but cannot assign anything.  Any
>> ideas?  I'm fresh out of them after trying many permutations of the syntax.
>>
>> Thank you,
>> Rodney Howeedy
>> Staff Engineer, Sun Microsystems
>> _______________________________________________
>> Users mailing list
>> Users at lists.ironpython.com
>> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>> _______________________________________________
>> Users mailing list
>> Users at lists.ironpython.com
>> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>>
>>     
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>   


From dinov at microsoft.com  Sat Dec  6 01:05:05 2008
From: dinov at microsoft.com (Dino Viehland)
Date: Fri, 5 Dec 2008 16:05:05 -0800
Subject: [IronPython] 'DispPropertyGet' error returned
	from	read-write	COM	interop property
In-Reply-To: <4939BA9B.3070701@sun.com>
References: <4939B507.4050104@sun.com>
	<350E7D38B6D819428718949920EC2355564A7FB4CC@NA-EXMSG-C102.redmond.corp.microsoft.com>
	<4939B77C.2060603@sun.com>
	<350E7D38B6D819428718949920EC2355564A7FB4DD@NA-EXMSG-C102.redmond.corp.microsoft.com>
	<4939BA9B.3070701@sun.com>
Message-ID: <350E7D38B6D819428718949920EC2355564A7FB506@NA-EXMSG-C102.redmond.corp.microsoft.com>

This might be broken but the output of dir(mytestRun) might be helpful as well to make sure (e.g. maybe there'll be some set_Field method or something).

But I suspect you might want to pass the -X:PreferComInteropAssembly command line option and I think that'll restore the beta 3 behavior for you.

-----Original Message-----
From: users-bounces at lists.ironpython.com [mailto:users-bounces at lists.ironpython.com] On Behalf Of Rodney Howeedy
Sent: Friday, December 05, 2008 3:35 PM
To: Discussion of IronPython
Subject: Re: [IronPython] 'DispPropertyGet' error returned from read-write COM interop property

Stack trace for the statement that raises the SytemError:
 >>> mytestRun.Field.Item['RN_USER_01'] = 'abc'
Object reference not set to an instance of an object.
   at
IronPython.Runtime.Types.ReflectedGetterSetter.CallTarget(CodeContext
context, SiteLocalStorage`1 storage, MethodInfo[] targets, Object
instance, Object[] args)
   at
IronPython.Runtime.Types.ReflectedGetterSetter.CallSetter(CodeContext
context, SiteLocalStorage`1 storage, Objectinstance, Object[] args,
Object value)
   at IronPython.Runtime.Types.ReflectedIndexer.SetValue(CodeContext
context, SiteLocalStorage`1 storage, Object[] keys, Object value)
   at
IronPython.Runtime.Types.ReflectedIndexer.set_Item(SiteLocalStorage`1
storage, Object[] key, Object value)
   at _stub_$1209##669(Closure , CallSite , Object , String , String )
   at
Microsoft.Scripting.Actions.MatchCaller.Call3[T0,T1,T2,TRet](Func`5
target, CallSite site, Object[] args)
   at Microsoft.Scripting.Actions.CallSite`1.UpdateAndExecute(Object[] args)
   at
Microsoft.Scripting.Actions.UpdateDelegates.Update3[T,T0,T1,T2,TRet](CallSite
site, T0 arg0, T1 arg1, T2 arg2)
   at <module>$1203##663(Closure , Scope , LanguageContext )
   at
Microsoft.Scripting.Runtime.OptimizedScriptCode.InvokeTarget(LambdaExpression
code, Scope scope)
   at Microsoft.Scripting.ScriptCode.Run(Scope scope)
   at
IronPython.Hosting.PythonCommandLine.<>c__DisplayClass1.<RunOneInteraction>b__0()
SystemError: Object reference not set to an instance of an object.

Dino Viehland wrote:
> What's the stack trace for the object reference exception if you run with the -X:ExceptionDetail option?
>
> -----Original Message-----
> From: users-bounces at lists.ironpython.com [mailto:users-bounces at lists.ironpython.com] On Behalf Of Rodney Howeedy
> Sent: Friday, December 05, 2008 3:22 PM
> To: Discussion of IronPython
> Subject: Re: [IronPython] 'DispPropertyGet' error returned from read-write COM interop property
>
> I found that one and tried before.  I didn't have any luck:
>  >>> mytestRun.Field.Item['RN_USER_01'] = 'abc'
> Traceback (most recent call last):
>   File "<stdin>", line 1, in <module>
> SystemError: Object reference not set to an instance of an object.
>
>  >>> print mytestRun.Field.Item['RN_USER_01']
> None
>
> I couldn't decipher the error so I couldn't figure out what I'm doing
> wrong.
> I suspect the folks on this list can. =)
>
> Dino Viehland wrote:
>
>> Does:
>>
>> mytestRun.Field.Item['RN_USER_01'] = 'abc'
>>
>> work?
>>
>> -----Original Message-----
>> From: users-bounces at lists.ironpython.com [mailto:users-bounces at lists.ironpython.com] On Behalf Of Rodney Howeedy
>> Sent: Friday, December 05, 2008 3:11 PM
>> To: users at lists.ironpython.com
>> Subject: [IronPython] 'DispPropertyGet' error returned from read-write COM interop property
>>
>> Hi everyone,
>>
>> I encounter an error with 2.0 RC2 that didn't occur in the same
>> IronPython code on 2.0 Beta 3.
>>
>> I use a factory object from a COM interop assembly that has a read-write
>> property named "Field".  I get a TypeError when I assign a value to this
>> property (read-write):
>>  >>> mytestRun.Field['RN_USER_01'] = 'abc'
>> Traceback (most recent call last):
>>   File "<stdin>", line 1, in <module>
>> TypeError: 'DispPropertyGet' object is unsubscriptable
>>
>>  >>> print mytestRun.Field['RN_USER_01']
>> None
>>
>> I imported the same COM interop assembly that worked with 2.0 Beta3.
>> The interop was converted with tlbimp.exe from a COM API .dll.  Here's
>> the 2.0 Beta3 output:
>>  >>> mytestRun.Field['RN_USER_01'] = 'abc'
>>  >>> print mytestRun.Field['RN_USER_01']
>> abc
>>
>> My best guess is that IronPython 2.0 RC2 interprets the property as
>> read-only even though the COM interop assembly defines the property
>> read-write. The error occurs in IronPython 2.0 RC2 with every COM
>> interop object with the read-write "Field" property.
>>
>> I can read the default assigned value but cannot assign anything.  Any
>> ideas?  I'm fresh out of them after trying many permutations of the syntax.
>>
>> Thank you,
>> Rodney Howeedy
>> Staff Engineer, Sun Microsystems
>> _______________________________________________
>> Users mailing list
>> Users at lists.ironpython.com
>> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>> _______________________________________________
>> Users mailing list
>> Users at lists.ironpython.com
>> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>>
>>
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>
_______________________________________________
Users mailing list
Users at lists.ironpython.com
http://lists.ironpython.com/listinfo.cgi/users-ironpython.com


From Shri.Borde at microsoft.com  Sat Dec  6 01:06:30 2008
From: Shri.Borde at microsoft.com (Shri Borde)
Date: Fri, 5 Dec 2008 16:06:30 -0800
Subject: [IronPython] Welcome David DiCato
Message-ID: <710DF26F214D2B4BB94287123FFE980A2DCBD82311@NA-EXMSG-C104.redmond.corp.microsoft.com>

Please welcome David DiCato to the IronPython team. He has his first shelveset in the Snap queue (queue.__init__ calls clear) already. You should hopefully hear more from him in the mailing list. Here is his intro in his own words.

Hi everyone, I'm David DiCato, the new Dev on the IronPython team. I grew up in Surf City (i.e. Huntington Beach, CA) and moved just 50 miles north to go to school in Pasadena. At Caltech, I earned my BS in Computer Science, applied to Microsoft after a short break, and the rest is history. Although I have never touched a surf board, I like to dabble in snowboarding, as well as piano, guitar and biking, although I'm good at none of these things. I'm also a big fan of music, incl. classic rock, punk and electronic, and film, incl. Kubrick, Hitchcock and Scorsese.

Thanks,
Shri

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081205/c85ffd46/attachment.html>

From Rodney.Howeedy at Sun.COM  Sat Dec  6 01:38:57 2008
From: Rodney.Howeedy at Sun.COM (Rodney Howeedy)
Date: Fri, 05 Dec 2008 17:38:57 -0700
Subject: [IronPython] 'DispPropertyGet' error returned	from	read-write
	COM	interop property
In-Reply-To: <350E7D38B6D819428718949920EC2355564A7FB506@NA-EXMSG-C102.redmond.corp.microsoft.com>
References: <4939B507.4050104@sun.com>
	<350E7D38B6D819428718949920EC2355564A7FB4CC@NA-EXMSG-C102.redmond.corp.microsoft.com>
	<4939B77C.2060603@sun.com>
	<350E7D38B6D819428718949920EC2355564A7FB4DD@NA-EXMSG-C102.redmond.corp.microsoft.com>
	<4939BA9B.3070701@sun.com>
	<350E7D38B6D819428718949920EC2355564A7FB506@NA-EXMSG-C102.redmond.corp.microsoft.com>
Message-ID: <4939C9A1.9080909@sun.com>

Thank you for that work around... now I can start the weekend.  :-D

There used to be a mytestRun.Field.SetValue in the dir() listing in Beta3:
 >>> dir(mytestRun.Field)
['Equals', 'GetHashCode', 'GetType', 'GetValue', 'Item', 
'ReferenceEquals', 'SetValue', 'ToString', '__class__', '__delattr__', 
'__delete__', '__doc__', '__get_
_', '__getattribute__', '__getitem__', '__hash__', '__init__', 
'__name__', '__new__', '__objclass__', '__reduce__', '__reduce_ex__', 
'__repr__', '__setattr__',
'__setitem__', '__str__']

But Field.SetValue disappeared in RC2:
 >>> dir(mytestRun.Field)
['ComMethodDesc', 'DispatchObject', 'Equals', 'GetHashCode', 
'GetMetaObject', 'GetType', 'Item', 'MemberwiseClone', 'Ref
erenceEquals', 'ToString', '__class__', '__delattr__', '__doc__', 
'__eq__', '__getattribute__', '__getitem__', '__hash__
', '__init__', '__ne__', '__new__', '__reduce__', '__reduce_ex__', 
'__repr__', '__setattr__', '__str__']

I didn't know what the Dev team had in mind.  Some things became more 
straightforward but I couldn't get this critical piece to work.

Additional dir() output on RC2:
 >>> dir(mytestRun.Field.Item)
['Equals', 'GetHashCode', 'GetType', 'GetValue', 'Item', 
'MemberwiseClone', 'ReferenceEquals', 'SetValue', 'ToString', '
__class__', '__delattr__', '__delete__', '__doc__', '__get__', 
'__getattribute__', '__getitem__', '__hash__', '__init__'
, '__name__', '__new__', '__objclass__', '__reduce__', '__reduce_ex__', 
'__repr__', '__setattr__', '__setitem__', '__str
__']

I already found and tried Item's SetValue but encountered the same 
SystemError.  I tried to find the expected inputs:
 >>> print mytestRun.Field.Item.SetValue.__doc__
bool SetValue(self, CodeContext context, 
SiteLocalStorage[CallSite[Func[CallSite, CodeContext, object, 
Array[object], ob
ject]]] storage, Array[object] keys, object value)

I may have gotten close, but no cigar:
 >>> mytestRun.Field.Item.SetValue(Array[object](["RN_DURATION"]),"10")
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
SystemError: Object reference not set to an instance of an object.

The requested 2.0 RC2 output:
 >>> dir(mytestRun)
['Attachments', 'AutoPost', 'BPStepParamFactory', 'CancelRun', 
'CopyDesignSteps', 'CopyStepsToTest', 'CreateObjRef', 'Eq
uals', 'ExtendedStorage', 'Field', 'GetHashCode', 'GetLifetimeService', 
'GetType', 'HasAttachment', 'History', 'ID', 'In
itializeLifetimeService', 'IsLocked', 'LockObject', 'MemberwiseClone', 
'Modified', 'Name', 'Params', 'Post', 'ReferenceE
quals', 'Refresh', 'ResolveStepsParameters', 'ResultLocation', 'Status', 
'StepFactory', 'TestId', 'TestInstance', 'TestI
nstanceID', 'TestSetID', 'ToString', 'UnLockObject', 'Undo', 'Virtual', 
'__class__', '__delattr__', '__doc__', '__getatt
ribute__', '__hash__', '__init__', '__new__', '__reduce__', 
'__reduce_ex__', '__repr__', '__setattr__', '__str__']

Thank you,
Rodney Howeedy
Staff Engineer, Sun Microsystems


Dino Viehland wrote:
> This might be broken but the output of dir(mytestRun) might be helpful as well to make sure (e.g. maybe there'll be some set_Field method or something).
>
> But I suspect you might want to pass the -X:PreferComInteropAssembly command line option and I think that'll restore the beta 3 behavior for you.
>
> -----Original Message-----
> From: users-bounces at lists.ironpython.com [mailto:users-bounces at lists.ironpython.com] On Behalf Of Rodney Howeedy
> Sent: Friday, December 05, 2008 3:35 PM
> To: Discussion of IronPython
> Subject: Re: [IronPython] 'DispPropertyGet' error returned from read-write COM interop property
>
> Stack trace for the statement that raises the SytemError:
>  >>> mytestRun.Field.Item['RN_USER_01'] = 'abc'
> Object reference not set to an instance of an object.
>    at
> IronPython.Runtime.Types.ReflectedGetterSetter.CallTarget(CodeContext
> context, SiteLocalStorage`1 storage, MethodInfo[] targets, Object
> instance, Object[] args)
>    at
> IronPython.Runtime.Types.ReflectedGetterSetter.CallSetter(CodeContext
> context, SiteLocalStorage`1 storage, Objectinstance, Object[] args,
> Object value)
>    at IronPython.Runtime.Types.ReflectedIndexer.SetValue(CodeContext
> context, SiteLocalStorage`1 storage, Object[] keys, Object value)
>    at
> IronPython.Runtime.Types.ReflectedIndexer.set_Item(SiteLocalStorage`1
> storage, Object[] key, Object value)
>    at _stub_$1209##669(Closure , CallSite , Object , String , String )
>    at
> Microsoft.Scripting.Actions.MatchCaller.Call3[T0,T1,T2,TRet](Func`5
> target, CallSite site, Object[] args)
>    at Microsoft.Scripting.Actions.CallSite`1.UpdateAndExecute(Object[] args)
>    at
> Microsoft.Scripting.Actions.UpdateDelegates.Update3[T,T0,T1,T2,TRet](CallSite
> site, T0 arg0, T1 arg1, T2 arg2)
>    at <module>$1203##663(Closure , Scope , LanguageContext )
>    at
> Microsoft.Scripting.Runtime.OptimizedScriptCode.InvokeTarget(LambdaExpression
> code, Scope scope)
>    at Microsoft.Scripting.ScriptCode.Run(Scope scope)
>    at
> IronPython.Hosting.PythonCommandLine.<>c__DisplayClass1.<RunOneInteraction>b__0()
> SystemError: Object reference not set to an instance of an object.
>
> Dino Viehland wrote:
>   
>> What's the stack trace for the object reference exception if you run with the -X:ExceptionDetail option?
>>
>> -----Original Message-----
>> From: users-bounces at lists.ironpython.com [mailto:users-bounces at lists.ironpython.com] On Behalf Of Rodney Howeedy
>> Sent: Friday, December 05, 2008 3:22 PM
>> To: Discussion of IronPython
>> Subject: Re: [IronPython] 'DispPropertyGet' error returned from read-write COM interop property
>>
>> I found that one and tried before.  I didn't have any luck:
>>  >>> mytestRun.Field.Item['RN_USER_01'] = 'abc'
>> Traceback (most recent call last):
>>   File "<stdin>", line 1, in <module>
>> SystemError: Object reference not set to an instance of an object.
>>
>>  >>> print mytestRun.Field.Item['RN_USER_01']
>> None
>>
>> I couldn't decipher the error so I couldn't figure out what I'm doing
>> wrong.
>> I suspect the folks on this list can. =)
>>
>> Dino Viehland wrote:
>>
>>     
>>> Does:
>>>
>>> mytestRun.Field.Item['RN_USER_01'] = 'abc'
>>>
>>> work?
>>>
>>> -----Original Message-----
>>> From: users-bounces at lists.ironpython.com [mailto:users-bounces at lists.ironpython.com] On Behalf Of Rodney Howeedy
>>> Sent: Friday, December 05, 2008 3:11 PM
>>> To: users at lists.ironpython.com
>>> Subject: [IronPython] 'DispPropertyGet' error returned from read-write COM interop property
>>>
>>> Hi everyone,
>>>
>>> I encounter an error with 2.0 RC2 that didn't occur in the same
>>> IronPython code on 2.0 Beta 3.
>>>
>>> I use a factory object from a COM interop assembly that has a read-write
>>> property named "Field".  I get a TypeError when I assign a value to this
>>> property (read-write):
>>>  >>> mytestRun.Field['RN_USER_01'] = 'abc'
>>> Traceback (most recent call last):
>>>   File "<stdin>", line 1, in <module>
>>> TypeError: 'DispPropertyGet' object is unsubscriptable
>>>
>>>  >>> print mytestRun.Field['RN_USER_01']
>>> None
>>>
>>> I imported the same COM interop assembly that worked with 2.0 Beta3.
>>> The interop was converted with tlbimp.exe from a COM API .dll.  Here's
>>> the 2.0 Beta3 output:
>>>  >>> mytestRun.Field['RN_USER_01'] = 'abc'
>>>  >>> print mytestRun.Field['RN_USER_01']
>>> abc
>>>
>>> My best guess is that IronPython 2.0 RC2 interprets the property as
>>> read-only even though the COM interop assembly defines the property
>>> read-write. The error occurs in IronPython 2.0 RC2 with every COM
>>> interop object with the read-write "Field" property.
>>>
>>> I can read the default assigned value but cannot assign anything.  Any
>>> ideas?  I'm fresh out of them after trying many permutations of the syntax.
>>>
>>> Thank you,
>>> Rodney Howeedy
>>> Staff Engineer, Sun Microsystems
>>> _______________________________________________
>>> Users mailing list
>>> Users at lists.ironpython.com
>>> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>>> _______________________________________________
>>> Users mailing list
>>> Users at lists.ironpython.com
>>> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>>>
>>>
>>>       
>> _______________________________________________
>> Users mailing list
>> Users at lists.ironpython.com
>> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>>
>> _______________________________________________
>> Users mailing list
>> Users at lists.ironpython.com
>> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>>
>>     
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>   


From fuzzyman at voidspace.org.uk  Sat Dec  6 02:06:32 2008
From: fuzzyman at voidspace.org.uk (Michael Foord)
Date: Sat, 06 Dec 2008 01:06:32 +0000
Subject: [IronPython] Welcome David DiCato
In-Reply-To: <710DF26F214D2B4BB94287123FFE980A2DCBD82311@NA-EXMSG-C104.redmond.corp.microsoft.com>
References: <710DF26F214D2B4BB94287123FFE980A2DCBD82311@NA-EXMSG-C104.redmond.corp.microsoft.com>
Message-ID: <4939D018.6020103@voidspace.org.uk>

Welcome David - and congratulations. Of course you realise that having a 
blog is a compulsory part of being on the dynamic languages team...

Michael Foord

Shri Borde wrote:
>
> Please welcome David DiCato to the IronPython team. He has his first 
> shelveset in the Snap queue (queue.__init__ calls clear) already. You 
> should hopefully hear more from him in the mailing list. Here is his 
> intro in his own words.
>
> Hi everyone, I?m David DiCato, the new Dev on the IronPython team. I 
> grew up in Surf City (i.e. Huntington Beach, CA) and moved just 50 
> miles north to go to school in Pasadena. At Caltech, I earned my BS in 
> Computer Science, applied to Microsoft after a short break, and the 
> rest is history. Although I have never touched a surf board, I like to 
> dabble in snowboarding, as well as piano, guitar and biking, although 
> I?m good at none of these things. I?m also a big fan of music, incl. 
> classic rock, punk and electronic, and film, incl. Kubrick, Hitchcock 
> and Scorsese.
>
> Thanks,
>
> Shri
>
> ------------------------------------------------------------------------
>
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>   


-- 
http://www.ironpythoninaction.com/
http://www.voidspace.org.uk/blog




From dan.eloff at gmail.com  Sat Dec  6 03:42:20 2008
From: dan.eloff at gmail.com (Dan Eloff)
Date: Fri, 5 Dec 2008 21:42:20 -0500
Subject: [IronPython] Welcome David DiCato
In-Reply-To: <710DF26F214D2B4BB94287123FFE980A2DCBD82311@NA-EXMSG-C104.redmond.corp.microsoft.com>
References: <710DF26F214D2B4BB94287123FFE980A2DCBD82311@NA-EXMSG-C104.redmond.corp.microsoft.com>
Message-ID: <4817b6fc0812051842reff208dk7312c0ec4f6131fa@mail.gmail.com>

Welcome David. It is heartening to see that even in these difficult
economic times Microsoft is allocating more resources to dynamic
languages.

Best wishes to you in your new job.

-Dan


From dan.eloff at gmail.com  Sat Dec  6 14:38:59 2008
From: dan.eloff at gmail.com (Dan Eloff)
Date: Sat, 6 Dec 2008 08:38:59 -0500
Subject: [IronPython] Last source drop was Nov 30. Is something broken?
Message-ID: <4817b6fc0812060538u4c89864apb4a945da5fbd1695@mail.gmail.com>

Just eager to get my hands on the latest code.

-Dan


From jdhardy at gmail.com  Sat Dec  6 17:07:51 2008
From: jdhardy at gmail.com (Jeff Hardy)
Date: Sat, 6 Dec 2008 09:07:51 -0700
Subject: [IronPython] NWSGI 0.7 released
In-Reply-To: <233dfa1d0812050011v1ac776a8q72f82853f18ea353@mail.gmail.com>
References: <ea89d0e50812041757o4d5833c7qf647fd38d37b26d6@mail.gmail.com>
	<233dfa1d0812050011v1ac776a8q72f82853f18ea353@mail.gmail.com>
Message-ID: <ea89d0e50812060807u79d45ceen7ea410acede0dfe@mail.gmail.com>

Hi Miha,
As Dan said, there are issues with str == unicode (although there is a
Jython branch of Django that I should look into). There are also some
other issues that seem to shift - trying to hit two moving targets
makes tracking them difficult. For very simple stuff, Django might
work, but I wouldn't expect it to do too much.

It is on my list of things to get working (along with Trac, which will
be much harder), but it's that pesky time thing, as always :).

- Jeff

On Fri, Dec 5, 2008 at 1:11 AM, Miha Valencic <miha.valencic at gmail.com> wrote:
> Jeff, do you have any info on whether Django works with this setup?
>
> Thanks,
>  Miha.
>
> 2008/12/5 Jeff Hardy <jdhardy at gmail.com>
>>
>> NWSGI 0.7 is now available. This version adds support for ASP.NET
>> sessions, fixes Unicode handling, and adds support for wildcard
>> mappings. It is available at
>>
>> http://www.codeplex.com/NWSGI/Release/ProjectReleases.aspx?ReleaseId=20189.
>>
>
>
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>
>


From ctrachte at gmail.com  Sat Dec  6 17:09:51 2008
From: ctrachte at gmail.com (Carl Trachte)
Date: Sat, 6 Dec 2008 09:09:51 -0700
Subject: [IronPython] Welcome David DiCato
In-Reply-To: <4817b6fc0812051842reff208dk7312c0ec4f6131fa@mail.gmail.com>
References: <710DF26F214D2B4BB94287123FFE980A2DCBD82311@NA-EXMSG-C104.redmond.corp.microsoft.com>
	<4817b6fc0812051842reff208dk7312c0ec4f6131fa@mail.gmail.com>
Message-ID: <426ada670812060809n7899cb84jb7c8ee2ef4a46f5d@mail.gmail.com>

Samo Arigato, David DiCato (and Dino and Curt and Jim H. too)

On Fri, Dec 5, 2008 at 7:42 PM, Dan Eloff <dan.eloff at gmail.com> wrote:
> Welcome David. It is heartening to see that even in these difficult
> economic times Microsoft is allocating more resources to dynamic
> languages.
>
> Best wishes to you in your new job.
>
> -Dan
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>


From fuzzyman at voidspace.org.uk  Sat Dec  6 18:20:48 2008
From: fuzzyman at voidspace.org.uk (Michael Foord)
Date: Sat, 06 Dec 2008 17:20:48 +0000
Subject: [IronPython] NWSGI 0.7 released
In-Reply-To: <ea89d0e50812060807u79d45ceen7ea410acede0dfe@mail.gmail.com>
References: <ea89d0e50812041757o4d5833c7qf647fd38d37b26d6@mail.gmail.com>	<233dfa1d0812050011v1ac776a8q72f82853f18ea353@mail.gmail.com>
	<ea89d0e50812060807u79d45ceen7ea410acede0dfe@mail.gmail.com>
Message-ID: <493AB470.4010705@voidspace.org.uk>

Jeff Hardy wrote:
> Hi Miha,
> As Dan said, there are issues with str == unicode (although there is a
> Jython branch of Django that I should look into). There are also some
> other issues that seem to shift - trying to hit two moving targets
> makes tracking them difficult. For very simple stuff, Django might
> work, but I wouldn't expect it to do too much.
>   

The Django guys are very interested in getting it to work on IronPython 
and would be open to discussion as to ways round problem. For example - 
how to resolve the str / unicode issue (perhaps patching in a compatible 
'bytes' like implementation rather than fixing on str as a way of 
deciding to serve binary data).

All the best,

Michael Foord

> It is on my list of things to get working (along with Trac, which will
> be much harder), but it's that pesky time thing, as always :).
>
> - Jeff
>
> On Fri, Dec 5, 2008 at 1:11 AM, Miha Valencic <miha.valencic at gmail.com> wrote:
>   
>> Jeff, do you have any info on whether Django works with this setup?
>>
>> Thanks,
>>  Miha.
>>
>> 2008/12/5 Jeff Hardy <jdhardy at gmail.com>
>>     
>>> NWSGI 0.7 is now available. This version adds support for ASP.NET
>>> sessions, fixes Unicode handling, and adds support for wildcard
>>> mappings. It is available at
>>>
>>> http://www.codeplex.com/NWSGI/Release/ProjectReleases.aspx?ReleaseId=20189.
>>>
>>>       
>> _______________________________________________
>> Users mailing list
>> Users at lists.ironpython.com
>> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>>
>>
>>     
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>   


-- 
http://www.ironpythoninaction.com/
http://www.voidspace.org.uk/blog




From jdhardy at gmail.com  Sat Dec  6 21:52:40 2008
From: jdhardy at gmail.com (Jeff Hardy)
Date: Sat, 6 Dec 2008 13:52:40 -0700
Subject: [IronPython] NWSGI 0.7 released
In-Reply-To: <493AB470.4010705@voidspace.org.uk>
References: <ea89d0e50812041757o4d5833c7qf647fd38d37b26d6@mail.gmail.com>
	<233dfa1d0812050011v1ac776a8q72f82853f18ea353@mail.gmail.com>
	<ea89d0e50812060807u79d45ceen7ea410acede0dfe@mail.gmail.com>
	<493AB470.4010705@voidspace.org.uk>
Message-ID: <ea89d0e50812061252w2c8941f4qf31a26bc765fdc54@mail.gmail.com>

On Sat, Dec 6, 2008 at 10:20 AM, Michael Foord
<fuzzyman at voidspace.org.uk> wrote:
> The Django guys are very interested in getting it to work on IronPython and
> would be open to discussion as to ways round problem. For example - how to
> resolve the str / unicode issue (perhaps patching in a compatible 'bytes'
> like implementation rather than fixing on str as a way of deciding to serve
> binary data).

While that's great to hear, there's a long way to go before that
becomes the biggest issue. There are still some annoying IPy bugs that
break more fundamental things (i.e. MultiValueDict is currently
broken). I haven't even run into the str/unicode problem yet, but it's
certainly good to keep in mind.

- Jeff


>
> All the best,
>
> Michael Foord
>
>> It is on my list of things to get working (along with Trac, which will
>> be much harder), but it's that pesky time thing, as always :).
>>
>> - Jeff
>>
>> On Fri, Dec 5, 2008 at 1:11 AM, Miha Valencic <miha.valencic at gmail.com>
>> wrote:
>>
>>>
>>> Jeff, do you have any info on whether Django works with this setup?
>>>
>>> Thanks,
>>>  Miha.
>>>
>>> 2008/12/5 Jeff Hardy <jdhardy at gmail.com>
>>>
>>>>
>>>> NWSGI 0.7 is now available. This version adds support for ASP.NET
>>>> sessions, fixes Unicode handling, and adds support for wildcard
>>>> mappings. It is available at
>>>>
>>>>
>>>> http://www.codeplex.com/NWSGI/Release/ProjectReleases.aspx?ReleaseId=20189.
>>>>
>>>>
>>>
>>> _______________________________________________
>>> Users mailing list
>>> Users at lists.ironpython.com
>>> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>>>
>>>
>>>
>>
>> _______________________________________________
>> Users mailing list
>> Users at lists.ironpython.com
>> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>>
>
>
> --
> http://www.ironpythoninaction.com/
> http://www.voidspace.org.uk/blog
>
>
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>


From jdhardy at gmail.com  Sat Dec  6 21:57:28 2008
From: jdhardy at gmail.com (Jeff Hardy)
Date: Sat, 6 Dec 2008 13:57:28 -0700
Subject: [IronPython] CP #20099: Stack Overflow using super() in dict
	subclass
Message-ID: <ea89d0e50812061257j2d3f495bn3a633d5658b76a2f@mail.gmail.com>

Hi,
Django's MultiValueDict is a key piece of its URL routing, and it
appears to be broken on IronPython 2.0RC2 (and RC1) when using a key
of None. See http://www.codeplex.com/IronPython/WorkItem/View.aspx?WorkItemId=20099
for details.

I hate to be needy with a release so close, but is there any chance of
fixing this for 2.0?

- Jeff


From fuzzyman at voidspace.org.uk  Sat Dec  6 22:01:21 2008
From: fuzzyman at voidspace.org.uk (Michael Foord)
Date: Sat, 06 Dec 2008 21:01:21 +0000
Subject: [IronPython] NWSGI 0.7 released
In-Reply-To: <ea89d0e50812061252w2c8941f4qf31a26bc765fdc54@mail.gmail.com>
References: <ea89d0e50812041757o4d5833c7qf647fd38d37b26d6@mail.gmail.com>	<233dfa1d0812050011v1ac776a8q72f82853f18ea353@mail.gmail.com>	<ea89d0e50812060807u79d45ceen7ea410acede0dfe@mail.gmail.com>	<493AB470.4010705@voidspace.org.uk>
	<ea89d0e50812061252w2c8941f4qf31a26bc765fdc54@mail.gmail.com>
Message-ID: <493AE821.5080402@voidspace.org.uk>

Jeff Hardy wrote:
> On Sat, Dec 6, 2008 at 10:20 AM, Michael Foord
> <fuzzyman at voidspace.org.uk> wrote:
>   
>> The Django guys are very interested in getting it to work on IronPython and
>> would be open to discussion as to ways round problem. For example - how to
>> resolve the str / unicode issue (perhaps patching in a compatible 'bytes'
>> like implementation rather than fixing on str as a way of deciding to serve
>> binary data).
>>     
>
> While that's great to hear, there's a long way to go before that
> becomes the biggest issue. There are still some annoying IPy bugs that
> break more fundamental things (i.e. MultiValueDict is currently
> broken). I haven't even run into the str/unicode problem yet, but it's
> certainly good to keep in mind.
>   

Cool. Would be great to have a reprise of what the current blockers are.

Michael

> - Jeff
>
>
>   
>> All the best,
>>
>> Michael Foord
>>
>>     
>>> It is on my list of things to get working (along with Trac, which will
>>> be much harder), but it's that pesky time thing, as always :).
>>>
>>> - Jeff
>>>
>>> On Fri, Dec 5, 2008 at 1:11 AM, Miha Valencic <miha.valencic at gmail.com>
>>> wrote:
>>>
>>>       
>>>> Jeff, do you have any info on whether Django works with this setup?
>>>>
>>>> Thanks,
>>>>  Miha.
>>>>
>>>> 2008/12/5 Jeff Hardy <jdhardy at gmail.com>
>>>>
>>>>         
>>>>> NWSGI 0.7 is now available. This version adds support for ASP.NET
>>>>> sessions, fixes Unicode handling, and adds support for wildcard
>>>>> mappings. It is available at
>>>>>
>>>>>
>>>>> http://www.codeplex.com/NWSGI/Release/ProjectReleases.aspx?ReleaseId=20189.
>>>>>
>>>>>
>>>>>           
>>>> _______________________________________________
>>>> Users mailing list
>>>> Users at lists.ironpython.com
>>>> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>>>>
>>>>
>>>>
>>>>         
>>> _______________________________________________
>>> Users mailing list
>>> Users at lists.ironpython.com
>>> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>>>
>>>       
>> --
>> http://www.ironpythoninaction.com/
>> http://www.voidspace.org.uk/blog
>>
>>
>> _______________________________________________
>> Users mailing list
>> Users at lists.ironpython.com
>> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>>
>>     
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>   


-- 
http://www.ironpythoninaction.com/
http://www.voidspace.org.uk/blog




From dinov at microsoft.com  Sat Dec  6 22:16:42 2008
From: dinov at microsoft.com (Dino Viehland)
Date: Sat, 6 Dec 2008 13:16:42 -0800
Subject: [IronPython] CP #20099: Stack Overflow using super() in
	dict	subclass
In-Reply-To: <ea89d0e50812061257j2d3f495bn3a633d5658b76a2f@mail.gmail.com>
References: <ea89d0e50812061257j2d3f495bn3a633d5658b76a2f@mail.gmail.com>
Message-ID: <350E7D38B6D819428718949920EC2355564B31B088@NA-EXMSG-C102.redmond.corp.microsoft.com>

The immediate question would be does fixing this unblock all other Django issues?  If there's going to be a long tail of other bugs then it'd be better to just fix them all for 2.0.1.

If you want to try this out the fix is easy, in PythonDictionary.cs replace the line:

                    this[(object)null] = value;

with:

                    _storage.Add(null, value);


You'll probably find that alone isn't sufficient - the getter probably needs a similar change where it does the same thing as this[object key].


> -----Original Message-----
> From: users-bounces at lists.ironpython.com [mailto:users-
> bounces at lists.ironpython.com] On Behalf Of Jeff Hardy
> Sent: Saturday, December 06, 2008 12:57 PM
> To: Discussion of IronPython
> Subject: [IronPython] CP #20099: Stack Overflow using super() in dict
> subclass
>
> Hi,
> Django's MultiValueDict is a key piece of its URL routing, and it
> appears to be broken on IronPython 2.0RC2 (and RC1) when using a key
> of None. See
> http://www.codeplex.com/IronPython/WorkItem/View.aspx?WorkItemId=20099
> for details.
>
> I hate to be needy with a release so close, but is there any chance of
> fixing this for 2.0?
>
> - Jeff
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com


From mbw.05 at multiparadigm.com  Sat Dec  6 23:55:04 2008
From: mbw.05 at multiparadigm.com (M. Whitener)
Date: Sat, 06 Dec 2008 15:55:04 -0700
Subject: [IronPython] =?utf-8?q?Examples_for_calling_IronPython_from_?=
	=?utf-8?b?wqBDIw==?=
Message-ID: <20081206155504.0e377ce7397666338d3aeed6efde40f2.7e1cc2525a.wbe@email.secureserver.net>

Is there a good source of documentation on how to use functions and
classes in an IronPython program from C#?
I need info on how to do this using C# 3.0 or lower - in other words
_not_ using C# 4.0 dynamic types.
The only example I have found is a very incomplete one from Alex Turner
given at PDC 2008.
Thanks in advance.




From seshapv at microsoft.com  Sun Dec  7 00:02:09 2008
From: seshapv at microsoft.com (Seshadri Pillailokam Vijayaraghavan)
Date: Sat, 6 Dec 2008 15:02:09 -0800
Subject: [IronPython]
 =?iso-8859-1?q?Examples_for_calling_IronPython_from_?=
 =?iso-8859-1?q?=A0C=23?=
In-Reply-To: <20081206155504.0e377ce7397666338d3aeed6efde40f2.7e1cc2525a.wbe@email.secureserver.net>
References: <20081206155504.0e377ce7397666338d3aeed6efde40f2.7e1cc2525a.wbe@email.secureserver.net>
Message-ID: <4F06AAB8EC03C349A5D6D0A797C4881C697E968274@NA-EXMSG-C112.redmond.corp.microsoft.com>

See if this helps - http://blogs.msdn.com/seshadripv/archive/2008/06/30/how-to-invoke-a-python-function-from-c-using-the-dlr-hosting-api.aspx

However, the sample will not compile as some of the runtime initialization APIs have changed since that sample was posted. To fix this replace the 'ScriptRuntime.Create()' in this line line - 'ScriptEngine pyEng = ScriptRuntime.Create().GetEngine("python");' with

'IronPython.Hosting.Python.CreateRuntime();'. The new line should look like ' ScriptEngine pyEng = IronPython.Hosting.Python.CreateRuntime().GetEngine("python");'

Let me know if you have any questions or need some information on this.

Thanks
Sesh
-----Original Message-----
From: users-bounces at lists.ironpython.com [mailto:users-bounces at lists.ironpython.com] On Behalf Of M. Whitener
Sent: Saturday, December 06, 2008 2:55 PM
To: Users at lists.ironpython.com
Subject: [IronPython] Examples for calling IronPython from  C#

Is there a good source of documentation on how to use functions and
classes in an IronPython program from C#?
I need info on how to do this using C# 3.0 or lower - in other words
_not_ using C# 4.0 dynamic types.
The only example I have found is a very incomplete one from Alex Turner
given at PDC 2008.
Thanks in advance.


_______________________________________________
Users mailing list
Users at lists.ironpython.com
http://lists.ironpython.com/listinfo.cgi/users-ironpython.com



From jslutter at reactorzero.com  Sun Dec  7 00:33:09 2008
From: jslutter at reactorzero.com (Jeff Slutter)
Date: Sat, 06 Dec 2008 18:33:09 -0500
Subject: [IronPython] SourceCodeKind.Statements versus
	SourceCodeKind.InteractiveCode
Message-ID: <493B0BB5.9040408@reactorzero.com>

I'm working on getting an interactive script console into my
application. Information on how to do this is a bit hard to find but I
came across an app.py by Jim Hugunin showing how to do it with Silverlight.

I based my C# app around that and have something working pretty well.
I'm always using SourceCodeKind.InteractiveCode when compiling the
source snippets and it works great. But there is a situation where it
doesn't do what I expect.

If I pass in the following string:

a = 10 + 3;
print(a);

(note: two statements, separated with a newline)

It will give me an error saying that the 'print' is an unexpected token.

If I pass in the following string:

a = 10 + 3; print(a);

It works as expected (printing out 13).

It works fine if there are multiple lines for a statement like:

if( a > 10 ):
 print(a);

If I use SourceCodeKind.Statements then all of the above works just
fine, but I don't get the nice things like the automatic print of the
returned value of the statement or the "_" variable.

Is there a reason why InteractiveCode does things different like that? I
want to give a consistent interface to my users and I think they would
expect that they can give two statements at once (especially if they can
do it on the same line).

I can work around this to get what I want by building up my buffer one
line at a time and testing to see if it is a "complete" statement and
executing that, then continuing to feed in the next line, etc. But, if I
don't have to, I don't want to do that.

Any ideas?

Thanks
Jeff



From dinov at microsoft.com  Sun Dec  7 00:44:53 2008
From: dinov at microsoft.com (Dino Viehland)
Date: Sat, 6 Dec 2008 15:44:53 -0800
Subject: [IronPython] SourceCodeKind.Statements
	versus	SourceCodeKind.InteractiveCode
In-Reply-To: <493B0BB5.9040408@reactorzero.com>
References: <493B0BB5.9040408@reactorzero.com>
Message-ID: <350E7D38B6D819428718949920EC2355564B31B08A@NA-EXMSG-C102.redmond.corp.microsoft.com>

This behavior is actually defined by CPython which specs interactive input as being:

interactive_input ::=  [stmt_list] NEWLINE | compound_stmt NEWLINE

The stmt_list allows the semi-colon delinated lines and the compound_stmt allows a single statement.

The idea here is that this is for a console input and that there really should be only one statement being entered at a time.

We could de-couple the relationship between interactive code & printing expression statements though.  Internally they're not really that related so it wouldn't be that hard to do.  If that sounds like it'd help you (or if there's others that would like to see it) you could open a feature request on CodePlex.

> -----Original Message-----
> From: users-bounces at lists.ironpython.com [mailto:users-
> bounces at lists.ironpython.com] On Behalf Of Jeff Slutter
> Sent: Saturday, December 06, 2008 3:33 PM
> To: Discussion of IronPython
> Subject: [IronPython] SourceCodeKind.Statements versus
> SourceCodeKind.InteractiveCode
>
> I'm working on getting an interactive script console into my
> application. Information on how to do this is a bit hard to find but I
> came across an app.py by Jim Hugunin showing how to do it with
> Silverlight.
>
> I based my C# app around that and have something working pretty well.
> I'm always using SourceCodeKind.InteractiveCode when compiling the
> source snippets and it works great. But there is a situation where it
> doesn't do what I expect.
>
> If I pass in the following string:
>
> a = 10 + 3;
> print(a);
>
> (note: two statements, separated with a newline)
>
> It will give me an error saying that the 'print' is an unexpected
> token.
>
> If I pass in the following string:
>
> a = 10 + 3; print(a);
>
> It works as expected (printing out 13).
>
> It works fine if there are multiple lines for a statement like:
>
> if( a > 10 ):
>  print(a);
>
> If I use SourceCodeKind.Statements then all of the above works just
> fine, but I don't get the nice things like the automatic print of the
> returned value of the statement or the "_" variable.
>
> Is there a reason why InteractiveCode does things different like that?
> I
> want to give a consistent interface to my users and I think they would
> expect that they can give two statements at once (especially if they
> can
> do it on the same line).
>
> I can work around this to get what I want by building up my buffer one
> line at a time and testing to see if it is a "complete" statement and
> executing that, then continuing to feed in the next line, etc. But, if
> I
> don't have to, I don't want to do that.
>
> Any ideas?
>
> Thanks
> Jeff
>
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com



From mbw.05 at multiparadigm.com  Sun Dec  7 03:40:12 2008
From: mbw.05 at multiparadigm.com (M. Whitener)
Date: Sat, 06 Dec 2008 19:40:12 -0700
Subject: [IronPython]
 =?utf-8?q?Examples=5Ffor=5Fcalling=5FIronPython=5Ffr?=
 =?utf-8?b?b21fIEMj?=
Message-ID: <20081206194012.0e377ce7397666338d3aeed6efde40f2.0c6673c3ea.wbe@email.secureserver.net>

Sesh, after your changes (below), and replacing using System.Scripting
with using Microsoft.Scripting
it works!

This is great.  Thank you.

You mentioned that planned to post some code showing how to call python
class instance methods.  Did you have a chance to do that?

Thanks again!


-------- Original Message --------
Subject: Re: [IronPython] Examples_for_calling_IronPython_from_  C#
From: Seshadri Pillailokam Vijayaraghavan <seshapv at microsoft.com>
Date: Sat, December 06, 2008 5:02 pm
To: Discussion of IronPython <users at lists.ironpython.com>

See if this helps -
http://blogs.msdn.com/seshadripv/archive/2008/06/30/how-to-invoke-a-python-function-from-c-using-the-dlr-hosting-api.aspx

However, the sample will not compile as some of the runtime
initialization APIs have changed since that sample was posted. To fix
this replace the 'ScriptRuntime.Create()' in this line line -
'ScriptEngine pyEng = ScriptRuntime.Create().GetEngine("python");' with

'IronPython.Hosting.Python.CreateRuntime();'. The new line should look
like ' ScriptEngine pyEng =
IronPython.Hosting.Python.CreateRuntime().GetEngine("python");'

Let me know if you have any questions or need some information on this.

Thanks
Sesh
-----Original Message-----
From: users-bounces at lists.ironpython.com
[mailto:users-bounces at lists.ironpython.com] On Behalf Of M. Whitener
Sent: Saturday, December 06, 2008 2:55 PM
To: Users at lists.ironpython.com
Subject: [IronPython] Examples for calling IronPython from C#

Is there a good source of documentation on how to use functions and
classes in an IronPython program from C#?
I need info on how to do this using C# 3.0 or lower - in other words
_not_ using C# 4.0 dynamic types.
The only example I have found is a very incomplete one from Alex Turner
given at PDC 2008.
Thanks in advance.


_______________________________________________
Users mailing list
Users at lists.ironpython.com
http://lists.ironpython.com/listinfo.cgi/users-ironpython.com

_______________________________________________
Users mailing list
Users at lists.ironpython.com
http://lists.ironpython.com/listinfo.cgi/users-ironpython.com





From seshapv at microsoft.com  Sun Dec  7 04:41:03 2008
From: seshapv at microsoft.com (Seshadri Pillailokam Vijayaraghavan)
Date: Sat, 6 Dec 2008 19:41:03 -0800
Subject: [IronPython] Examples_for_calling_IronPython_from_ C#
In-Reply-To: <20081206194012.0e377ce7397666338d3aeed6efde40f2.0c6673c3ea.wbe@email.secureserver.net>
References: <20081206194012.0e377ce7397666338d3aeed6efde40f2.0c6673c3ea.wbe@email.secureserver.net>
Message-ID: <4F06AAB8EC03C349A5D6D0A797C4881C697E968292@NA-EXMSG-C112.redmond.corp.microsoft.com>

>>>Did you have a chance to do that?
Not in terms of a blog post...

But this example should demonstrate that

        public void InvokeInstance_Example() {
            ScriptEngine engine = _runtime.GetEngine("py");
            ScriptScope scope = engine.CreateScope();
            ObjectOperations operation = engine.CreateOperations(scope);

            string pyCode = @"class Foo(object):
     i = -1
     def Bar(self):
         return i";

            ScriptSource src = engine.CreateScriptSourceFromString(pyCode, SourceCodeKind.Statements);
            src.Execute(scope);

            object FooClass = scope.GetVariable("Foo");

            object newObjectInstance = engine.Operations.CreateInstance(FooClass, new object[] { }); // create new FooClass
            var bar = engine.Operations.GetMember<Func<int>>(newObjectInstance, "Bar");
            int n = bar();
            Assert.AreEqual(-1, n);
        }

Thanks
Sesh

-----Original Message-----
From: users-bounces at lists.ironpython.com [mailto:users-bounces at lists.ironpython.com] On Behalf Of M. Whitener
Sent: Saturday, December 06, 2008 6:40 PM
To: Discussion of IronPython
Subject: Re: [IronPython] Examples_for_calling_IronPython_from_ C#

Sesh, after your changes (below), and replacing using System.Scripting
with using Microsoft.Scripting
it works!

This is great.  Thank you.

You mentioned that planned to post some code showing how to call python
class instance methods.  Did you have a chance to do that?

Thanks again!


-------- Original Message --------
Subject: Re: [IronPython] Examples_for_calling_IronPython_from_  C#
From: Seshadri Pillailokam Vijayaraghavan <seshapv at microsoft.com>
Date: Sat, December 06, 2008 5:02 pm
To: Discussion of IronPython <users at lists.ironpython.com>

See if this helps -
http://blogs.msdn.com/seshadripv/archive/2008/06/30/how-to-invoke-a-python-function-from-c-using-the-dlr-hosting-api.aspx

However, the sample will not compile as some of the runtime
initialization APIs have changed since that sample was posted. To fix
this replace the 'ScriptRuntime.Create()' in this line line -
'ScriptEngine pyEng = ScriptRuntime.Create().GetEngine("python");' with

'IronPython.Hosting.Python.CreateRuntime();'. The new line should look
like ' ScriptEngine pyEng =
IronPython.Hosting.Python.CreateRuntime().GetEngine("python");'

Let me know if you have any questions or need some information on this.

Thanks
Sesh
-----Original Message-----
From: users-bounces at lists.ironpython.com
[mailto:users-bounces at lists.ironpython.com] On Behalf Of M. Whitener
Sent: Saturday, December 06, 2008 2:55 PM
To: Users at lists.ironpython.com
Subject: [IronPython] Examples for calling IronPython from C#

Is there a good source of documentation on how to use functions and
classes in an IronPython program from C#?
I need info on how to do this using C# 3.0 or lower - in other words
_not_ using C# 4.0 dynamic types.
The only example I have found is a very incomplete one from Alex Turner
given at PDC 2008.
Thanks in advance.


_______________________________________________
Users mailing list
Users at lists.ironpython.com
http://lists.ironpython.com/listinfo.cgi/users-ironpython.com

_______________________________________________
Users mailing list
Users at lists.ironpython.com
http://lists.ironpython.com/listinfo.cgi/users-ironpython.com



_______________________________________________
Users mailing list
Users at lists.ironpython.com
http://lists.ironpython.com/listinfo.cgi/users-ironpython.com



From dan.eloff at gmail.com  Sun Dec  7 05:30:01 2008
From: dan.eloff at gmail.com (Dan Eloff)
Date: Sat, 6 Dec 2008 23:30:01 -0500
Subject: [IronPython] SourceCodeKind.Statements versus
	SourceCodeKind.InteractiveCode
In-Reply-To: <493B0BB5.9040408@reactorzero.com>
References: <493B0BB5.9040408@reactorzero.com>
Message-ID: <4817b6fc0812062030w4582d862yca74e7f35ec7a8c8@mail.gmail.com>

On Sat, Dec 6, 2008 at 6:33 PM, Jeff Slutter <jslutter at reactorzero.com> wrote:
> I'm working on getting an interactive script console into my
> application. Information on how to do this is a bit hard to find but I
> came across an app.py by Jim Hugunin showing how to do it with Silverlight.

You may find http://code.google.com/p/silvershell/ to be of some
value, specifically the code in engine.py

-Dan


From dan.eloff at gmail.com  Sun Dec  7 05:51:50 2008
From: dan.eloff at gmail.com (Dan Eloff)
Date: Sat, 6 Dec 2008 23:51:50 -0500
Subject: [IronPython] SilverShell 0.6.1 Released - many "intellisense" fixes
	+ all source
Message-ID: <4817b6fc0812062051j6b13d923teb6a79a8fcdce8a0@mail.gmail.com>

I'm trying to walk a fine line between spamming this list, and
alerting the 200+ people who downloaded SilverShell 0.6.0 that there's
an important new bugfix release. I put a lot of irritating flaws in
the "intellisense" to rest, including a major issue with member
completion that crept into 0.6.0 at the last minute. Had I known there
was going to be such a response I think I would have tested it more
thoroughly.

As promised all source code is in this release. Please notify me of
remaining bugs, either via email, or via the issue tracker on Google
Code. Patches are most welcome.

Thanks for downloading SilverShell!

-Dan


From jdhardy at gmail.com  Sun Dec  7 15:21:48 2008
From: jdhardy at gmail.com (Jeff Hardy)
Date: Sun, 7 Dec 2008 07:21:48 -0700
Subject: [IronPython] CP #20099: Stack Overflow using super() in dict
	subclass
In-Reply-To: <350E7D38B6D819428718949920EC2355564B31B088@NA-EXMSG-C102.redmond.corp.microsoft.com>
References: <ea89d0e50812061257j2d3f495bn3a633d5658b76a2f@mail.gmail.com>
	<350E7D38B6D819428718949920EC2355564B31B088@NA-EXMSG-C102.redmond.corp.microsoft.com>
Message-ID: <ea89d0e50812070621g1e9ba1e8u6d71c728ff26f541@mail.gmail.com>

On Sat, Dec 6, 2008 at 2:16 PM, Dino Viehland <dinov at microsoft.com> wrote:
> The immediate question would be does fixing this unblock all other Django issues?  If there's going to be a long tail of other bugs then it'd be better to just fix them all for 2.0.1.

I doubt it, if past experience is anything to go by. I'll see what
happens when I apply the fix, but there are probably more bugs like
this lurking in there.

>
> If you want to try this out the fix is easy, in PythonDictionary.cs replace the line:
>
>                    this[(object)null] = value;
>
> with:
>
>                    _storage.Add(null, value);
>
>
> You'll probably find that alone isn't sufficient - the getter probably needs a similar change where it does the same thing as this[object key].

If the fix is that simple, I'm fine using a custom build for testing.
I was afraid the fix was going to be much more involved.

Thanks for looking into this so quickly!

- Jeff


From miha.valencic at gmail.com  Sun Dec  7 18:02:22 2008
From: miha.valencic at gmail.com (Miha Valencic)
Date: Sun, 7 Dec 2008 18:02:22 +0100
Subject: [IronPython] NWSGI 0.7 released
In-Reply-To: <ea89d0e50812060807u79d45ceen7ea410acede0dfe@mail.gmail.com>
References: <ea89d0e50812041757o4d5833c7qf647fd38d37b26d6@mail.gmail.com>
	<233dfa1d0812050011v1ac776a8q72f82853f18ea353@mail.gmail.com>
	<ea89d0e50812060807u79d45ceen7ea410acede0dfe@mail.gmail.com>
Message-ID: <233dfa1d0812070902q7e1bbf2bxa936b7cfefe317ad@mail.gmail.com>

Thanks for the info.

Miha

2008/12/6 Jeff Hardy <jdhardy at gmail.com>

> Hi Miha,
> As Dan said, there are issues with str == unicode (although there is a
> [...]
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081207/0b896e03/attachment.html>

From dinov at microsoft.com  Sun Dec  7 20:02:10 2008
From: dinov at microsoft.com (Dino Viehland)
Date: Sun, 7 Dec 2008 11:02:10 -0800
Subject: [IronPython] Last source drop was Nov 30. Is something broken?
In-Reply-To: <4817b6fc0812060538u4c89864apb4a945da5fbd1695@mail.gmail.com>
References: <4817b6fc0812060538u4c89864apb4a945da5fbd1695@mail.gmail.com>
Message-ID: <350E7D38B6D819428718949920EC2355564B31B0CC@NA-EXMSG-C102.redmond.corp.microsoft.com>

Something went wrong on the machine that pushes the source code out and then there were some changes that required the script to be updated.  It's working now and pushed out an update yesterday.

> -----Original Message-----
> From: users-bounces at lists.ironpython.com [mailto:users-
> bounces at lists.ironpython.com] On Behalf Of Dan Eloff
> Sent: Saturday, December 06, 2008 5:39 AM
> To: Discussion of IronPython
> Subject: [IronPython] Last source drop was Nov 30. Is something broken?
>
> Just eager to get my hands on the latest code.
>
> -Dan
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com


From sanxiyn at gmail.com  Sun Dec  7 22:52:15 2008
From: sanxiyn at gmail.com (Seo Sanghyeon)
Date: Mon, 8 Dec 2008 06:52:15 +0900
Subject: [IronPython] Unused files
Message-ID: <5b0248170812071352t4221ce3at75c7b57f79bbc1fa@mail.gmail.com>

Following files seem to be unused but are included in 2.0 RC2 source.
Please remove them before the final release...

Microsoft.Scripting.Core/Actions/RestrictedMetaObject.cs
Microsoft.Scripting.Core/Ast/ExpressionTreeVisitor.Generated.cs
Microsoft.Scripting.Core/Ast/LabeledStatement.cs
Microsoft.Scripting.Core/Com/MetaUnwrappedComObject.cs
IronPython/Runtime/Binding/IPythonCallable.cs
IronPython/Runtime/ThreadLocal.cs

Thanks,

-- 
Seo Sanghyeon


From dan.eloff at gmail.com  Mon Dec  8 00:00:19 2008
From: dan.eloff at gmail.com (Dan Eloff)
Date: Sun, 7 Dec 2008 18:00:19 -0500
Subject: [IronPython] Last source drop was Nov 30. Is something broken?
In-Reply-To: <350E7D38B6D819428718949920EC2355564B31B0CC@NA-EXMSG-C102.redmond.corp.microsoft.com>
References: <4817b6fc0812060538u4c89864apb4a945da5fbd1695@mail.gmail.com>
	<350E7D38B6D819428718949920EC2355564B31B0CC@NA-EXMSG-C102.redmond.corp.microsoft.com>
Message-ID: <4817b6fc0812071500k7d5588aejd74da1ac10eee614@mail.gmail.com>

On Sun, Dec 7, 2008 at 2:02 PM, Dino Viehland <dinov at microsoft.com> wrote:
> Something went wrong on the machine that pushes the source code out and then there were some changes that required the script to be updated.  It's working now and pushed out an update yesterday.
>

Excellent. *rubs hands together gleefully* It's almost like Christmas
when I get a new source drop.

-Dan


From sanxiyn at gmail.com  Mon Dec  8 06:10:03 2008
From: sanxiyn at gmail.com (Seo Sanghyeon)
Date: Mon, 8 Dec 2008 14:10:03 +0900
Subject: [IronPython] IronPython 2.0 RC 2 on Mono
Message-ID: <5b0248170812072110j6a2694c8v8f221a9575cf1f22@mail.gmail.com>

Currently you need Mono SVN (both to run and to compile). Using SVN
r120972 below.

The binary runs fine.

IronPython includes its own copy of ExtensionAttribute, but this
doesn't seem to work with Mono C# compiler: that is, extension methods
don't get recognized as extension methods. This causes trouble where
MetaObject is defined in Microsoft.Scripting.Core.dll, and its
extensions are defined in MetaObjectExtensions in
Microsoft.Scripting.dll. My hack below copies MetaObjectExtensions and
its dependencies to Microsoft.Scripting.Core and patch it to be a
partial class instead of a static class housing extension methods. Any
better idea?

Also there are some unused files in the source distribution.

Below are steps I used to compile the source on Mono.

Download source from
http://www.codeplex.com/IronPython/Release/ProjectReleases.aspx?ReleaseId=19841
$ unzip IronPython-2.0-Src.zip
$ cd IronPython-2.0

Some hacks...
$ svn co -r 597
https://fepy.svn.sourceforge.net/svnroot/fepy/trunk/patches/latest
build

This deletes unused files and does copying mentioned above
$ sh build/pre.sh

Copies NAnt build file: it uses noconfig to avoid pulling System.Core.dll
$ cp build/IronPython.build Src

Patches
$ patch -p1 < build/patch-exthack
$ patch -p1 < build/patch-console

Build
$ cd Src
$ nant

-- 
Seo Sanghyeon


From glenn.k.jones+ipy at gmail.com  Mon Dec  8 12:40:25 2008
From: glenn.k.jones+ipy at gmail.com (Glenn Jones)
Date: Mon, 8 Dec 2008 11:40:25 +0000
Subject: [IronPython] Compiled IronPython mixed with source files
Message-ID: <2679f6510812080340v5ef072f3td5b638fc516ffb8a@mail.gmail.com>

Hi all,

In our work to get Resolver One working on IronPython 2, we've noticed this
interesting behaviour and would like to know whether it's defined behaviour
that we can rely on or just happenstance.

Create a package with the following structure:
+ foo
|----__init__.py
|----foo.py

where foo.py contains
print "foo (compiled)"

and compile it to foo.dll.

Now change foo.py to
print "foo (not compiled)"

and add a bar package:
+ foo
|----__init__.py
|----foo.py
|---+ bar
    |--- __init__.py
    |--- bar.py

where bar contains:
print "bar (not compiled)"

then, the following code:
import clr
clr.AddReference('foo')
from foo import foo
from foo.bar import bar

outputs:
foo(compiled)
bar (not compiled)

and removing the AddReference line outputs:
foo (not compiled)
bar (not compiled)


This makes it clear that we can have a package that is split between
compiled dll and source and that they work together.

Can we rely on this behaviour? Is there a specified order in which imports
from different sources are resolved? How is the multiple identity of the foo
module (compiled and not compiled) handled?

Thanks

Glenn & Michael
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081208/de306493/attachment.html>

From dinov at microsoft.com  Mon Dec  8 18:50:56 2008
From: dinov at microsoft.com (Dino Viehland)
Date: Mon, 8 Dec 2008 09:50:56 -0800
Subject: [IronPython] Compiled IronPython mixed with source files
In-Reply-To: <2679f6510812080340v5ef072f3td5b638fc516ffb8a@mail.gmail.com>
References: <2679f6510812080340v5ef072f3td5b638fc516ffb8a@mail.gmail.com>
Message-ID: <350E7D38B6D819428718949920EC2355564B31B25C@NA-EXMSG-C102.redmond.corp.microsoft.com>

I think you can depend upon this behavior.  Compiled loader support is implemented as a meta path hook and therefore the pure Python version of this looks like:

import sys

class my_loader(object):
                def find_module(self, fullname, path):
                                print 'find', fullname
                                if fullname == 'foo': return module_loader()

class module_loader(object):
                def load_module(self, fullname):
                                print 'load', fullname
                                if fullname == 'foo':
                                                res = __import__('x')
                                                sys.modules['foo'] = res
                                                res.__name__ = 'foo'
                                                res.__path__ = ['foo']
                                                return res


sys.meta_path.append(my_loader())

from foo import foo
from foo.bar import bar

On disk x is a package which has your 1st package structure, foo is a package which has your 2nd package structure.  The loader hook in this case is importing x, renaming it to foo, and then it doesn't resolve foo.bar.  That turns into a full lookup and the importer rules dictate that we'll happily look on disk because foo appears to be a parent package.  This outputs the same on CPython and IronPython so I'm pretty sure we're correct here :)

As to the identity of the modules - the compiled modules always win because the meta path hooks get to execute first.  So you'll only ever successfully import compiled modules if they exist.  So while you can augment an existing package you cannot replace files from it.


From: users-bounces at lists.ironpython.com [mailto:users-bounces at lists.ironpython.com] On Behalf Of Glenn Jones
Sent: Monday, December 08, 2008 3:40 AM
To: Discussion of IronPython
Subject: [IronPython] Compiled IronPython mixed with source files

Hi all,

In our work to get Resolver One working on IronPython 2, we've noticed this interesting behaviour and would like to know whether it's defined behaviour that we can rely on or just happenstance.

Create a package with the following structure:
+ foo
|----__init__.py
|----foo.py

where foo.py contains
print "foo (compiled)"

and compile it to foo.dll.

Now change foo.py to
print "foo (not compiled)"

and add a bar package:
+ foo
|----__init__.py
|----foo.py
|---+ bar
    |--- __init__.py
    |--- bar.py

where bar contains:
print "bar (not compiled)"

then, the following code:
import clr
clr.AddReference('foo')
from foo import foo
from foo.bar import bar

outputs:
foo(compiled)
bar (not compiled)

and removing the AddReference line outputs:
foo (not compiled)
bar (not compiled)


This makes it clear that we can have a package that is split between compiled dll and source and that they work together.

Can we rely on this behaviour? Is there a specified order in which imports from different sources are resolved? How is the multiple identity of the foo module (compiled and not compiled) handled?

Thanks

Glenn & Michael

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081208/5a5d5332/attachment.html>

From dinov at microsoft.com  Mon Dec  8 18:54:03 2008
From: dinov at microsoft.com (Dino Viehland)
Date: Mon, 8 Dec 2008 09:54:03 -0800
Subject: [IronPython] IronPython 2.0 RC 2 on Mono
In-Reply-To: <5b0248170812072110j6a2694c8v8f221a9575cf1f22@mail.gmail.com>
References: <5b0248170812072110j6a2694c8v8f221a9575cf1f22@mail.gmail.com>
Message-ID: <350E7D38B6D819428718949920EC2355564B31B262@NA-EXMSG-C102.redmond.corp.microsoft.com>

Thanks for doing this Seo!  I think we'll remove the extra source files before we release 2.0 final.

If you haven't already I would suggest reporting a bug to Mono about the extension attribute.  Both MS's VB & C# compilers support recognizing it from an arbitrary assembly so it seems like a common pattern.  Unfortunately I don't have any better suggestions.

-----Original Message-----
From: users-bounces at lists.ironpython.com [mailto:users-bounces at lists.ironpython.com] On Behalf Of Seo Sanghyeon
Sent: Sunday, December 07, 2008 9:10 PM
To: Discussion of IronPython
Subject: [IronPython] IronPython 2.0 RC 2 on Mono

Currently you need Mono SVN (both to run and to compile). Using SVN
r120972 below.

The binary runs fine.

IronPython includes its own copy of ExtensionAttribute, but this
doesn't seem to work with Mono C# compiler: that is, extension methods
don't get recognized as extension methods. This causes trouble where
MetaObject is defined in Microsoft.Scripting.Core.dll, and its
extensions are defined in MetaObjectExtensions in
Microsoft.Scripting.dll. My hack below copies MetaObjectExtensions and
its dependencies to Microsoft.Scripting.Core and patch it to be a
partial class instead of a static class housing extension methods. Any
better idea?

Also there are some unused files in the source distribution.

Below are steps I used to compile the source on Mono.

Download source from
http://www.codeplex.com/IronPython/Release/ProjectReleases.aspx?ReleaseId=19841
$ unzip IronPython-2.0-Src.zip
$ cd IronPython-2.0

Some hacks...
$ svn co -r 597
https://fepy.svn.sourceforge.net/svnroot/fepy/trunk/patches/latest
build

This deletes unused files and does copying mentioned above
$ sh build/pre.sh

Copies NAnt build file: it uses noconfig to avoid pulling System.Core.dll
$ cp build/IronPython.build Src

Patches
$ patch -p1 < build/patch-exthack
$ patch -p1 < build/patch-console

Build
$ cd Src
$ nant

--
Seo Sanghyeon
_______________________________________________
Users mailing list
Users at lists.ironpython.com
http://lists.ironpython.com/listinfo.cgi/users-ironpython.com


From yashgt at gmail.com  Tue Dec  9 14:59:22 2008
From: yashgt at gmail.com (Yash Ganthe)
Date: Tue, 9 Dec 2008 19:29:22 +0530
Subject: [IronPython] IronPython Studio and IronPython 2.0
Message-ID: <4f428e7b0812090559l381860f4pe344a564fc721214@mail.gmail.com>

I use IPY 2.0 RC2. I would like to debug through the code in Visual Studio.
Can IronPython Studio be configured to use IPY 2.0?

Thanks,
Yash
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081209/cf4f401c/attachment.html>

From glenn.k.jones+ipy at gmail.com  Tue Dec  9 17:37:04 2008
From: glenn.k.jones+ipy at gmail.com (Glenn Jones)
Date: Tue, 9 Dec 2008 16:37:04 +0000
Subject: [IronPython] subclasses of list with __mul__ defined
Message-ID: <2679f6510812090837o2460a6d8m370441f9dddc12ef@mail.gmail.com>

Hi guys,

Another issue from our upgrade to IPy2:
class thing(list):
    def __init__(self, value):
        self.value = value

    def __mul__(self, other):
        return self.value * other
t1 = thing(3.0)
print t1 * 3.0

results in:

Traceback (most recent call last):
  File "testmat.py", line 12, in testmat.py
  File "Microsoft.Scripting.Core", line unknown, in New
  File "Microsoft.Scripting.Core", line unknown, in New
  File "Microsoft.Scripting.Core", line unknown, in ValidateArgumentTypes
ValueError: Expression of type 'System.Double' cannot be used for
constructor parameter of type 'System.Object'


This has the potential to cause us significant pain since a third-party
(pure python) library that we use relies on this behaving correctly.

We'll raise this as a bug on codeplex.

Thanks
Glenn & Tom
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081209/1a484227/attachment.html>

From dinov at microsoft.com  Wed Dec 10 00:27:45 2008
From: dinov at microsoft.com (Dino Viehland)
Date: Tue, 9 Dec 2008 15:27:45 -0800
Subject: [IronPython] IronPython Studio and IronPython 2.0
In-Reply-To: <4f428e7b0812090559l381860f4pe344a564fc721214@mail.gmail.com>
References: <4f428e7b0812090559l381860f4pe344a564fc721214@mail.gmail.com>
Message-ID: <350E7D38B6D819428718949920EC2355564B31BB5F@NA-EXMSG-C102.redmond.corp.microsoft.com>

I don't believe there is right now w/o a bunch of work to update to the new 2.0 hosting APIs.

From: users-bounces at lists.ironpython.com [mailto:users-bounces at lists.ironpython.com] On Behalf Of Yash Ganthe
Sent: Tuesday, December 09, 2008 5:59 AM
To: users at lists.ironpython.com
Subject: [IronPython] IronPython Studio and IronPython 2.0

I use IPY 2.0 RC2. I would like to debug through the code in Visual Studio. Can IronPython Studio be configured to use IPY 2.0?

Thanks,
Yash
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081209/fcafc20b/attachment.html>

From dinov at microsoft.com  Wed Dec 10 00:43:29 2008
From: dinov at microsoft.com (Dino Viehland)
Date: Tue, 9 Dec 2008 15:43:29 -0800
Subject: [IronPython] subclasses of list with __mul__ defined
In-Reply-To: <2679f6510812090837o2460a6d8m370441f9dddc12ef@mail.gmail.com>
References: <2679f6510812090837o2460a6d8m370441f9dddc12ef@mail.gmail.com>
Message-ID: <350E7D38B6D819428718949920EC2355564B31BB74@NA-EXMSG-C102.redmond.corp.microsoft.com>

The fix for this is pretty simple.  It's going into 2.1 as we speak, and I'll port it back to the 2.0.x branch.  Unfortunately at this point I don't think it'll make it into 2.0 final.

From: users-bounces at lists.ironpython.com [mailto:users-bounces at lists.ironpython.com] On Behalf Of Glenn Jones
Sent: Tuesday, December 09, 2008 8:37 AM
To: Discussion of IronPython
Subject: [IronPython] subclasses of list with __mul__ defined

Hi guys,

Another issue from our upgrade to IPy2:
class thing(list):
    def __init__(self, value):
        self.value = value

    def __mul__(self, other):
        return self.value * other
t1 = thing(3.0)
print t1 * 3.0

results in:

Traceback (most recent call last):
  File "testmat.py", line 12, in testmat.py
  File "Microsoft.Scripting.Core", line unknown, in New
  File "Microsoft.Scripting.Core", line unknown, in New
  File "Microsoft.Scripting.Core", line unknown, in ValidateArgumentTypes
ValueError: Expression of type 'System.Double' cannot be used for constructor parameter of type 'System.Object'


This has the potential to cause us significant pain since a third-party (pure python) library that we use relies on this behaving correctly.

We'll raise this as a bug on codeplex.

Thanks
Glenn & Tom
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081209/4cfc0f6b/attachment.html>

From sanxiyn at gmail.com  Wed Dec 10 11:39:35 2008
From: sanxiyn at gmail.com (Seo Sanghyeon)
Date: Wed, 10 Dec 2008 19:39:35 +0900
Subject: [IronPython] Invalid construct in DLR's MSBuild file?
Message-ID: <5b0248170812100239i2abe78fby4c6c20a7fef6daef@mail.gmail.com>

Hello, now I am trying to use xbuild (open source implementation of
MSBuild language) to build IronPython, and the first error (after
instrumenting the source to make it clear what the error is) is this:

$ xbuild Microsoft.Scripting.ExtensionAttribute.csproj
XBuild Engine Version 0.1
Mono, Version 2.3.0.0
Copyright (C) Marek Sieradzki 2005. All rights reserved.
ConditionParser failed to parse  $(SilverlightBuild) != 'true'

Unhandled Exception:
Microsoft.Build.BuildEngine.InvalidProjectFileException: The requested
feature is not implemented. ---> System.NotImplementedException: The
requested feature is not implemented.
  at Microsoft.Build.BuildEngine.ConditionParser.ParseFactorExpression
() [0x00000]
  at Microsoft.Build.BuildEngine.ConditionParser.ParseRelationalExpression
() [0x00000]
  at Microsoft.Build.BuildEngine.ConditionParser.ParseBooleanOr () [0x00000]
  at Microsoft.Build.BuildEngine.ConditionParser.ParseBooleanAnd () [0x00000]
  at Microsoft.Build.BuildEngine.ConditionParser.ParseBooleanExpression
() [0x00000]
  at Microsoft.Build.BuildEngine.ConditionParser.ParseExpression () [0x00000]
  at Microsoft.Build.BuildEngine.ConditionParser.ParseCondition
(System.String condition) [0x00000]
  --- End of inner exception stack trace ---
  at Microsoft.Build.BuildEngine.Project.DoLoad (System.IO.TextReader
textReader) [0x00000]
  at Microsoft.Build.BuildEngine.Project.Load (System.String
projectFileName) [0x00000]
  at Mono.XBuild.CommandLine.MainClass.Execute () [0x00000]

According to MSDN, MSBuild Conditions
http://msdn.microsoft.com/en-us/library/7szfhaft.aspx conditions
should be of the form 'stringA' != 'stringB', where "single quotes are
not required for simple alphanumeric strings or boolean values". Is
$(SilverlightBuild) this case? If it isn't, shouldn't it be quoted as
'$(SilverlightBuild)'?

If I manually change this condition to be quoted (I also note, all
other uses of $(SilverlightBuild) is actually quoted in this file),
xbuild builds this assembly successfully.

Even if this is actually valid (then I argue documentation is
confusing), I think it would be a good idea to quote this anyway.

-- 
Seo Sanghyeon


From fuzzyman at voidspace.org.uk  Wed Dec 10 14:12:21 2008
From: fuzzyman at voidspace.org.uk (Michael Foord)
Date: Wed, 10 Dec 2008 13:12:21 +0000
Subject: [IronPython] IronPython Studio and IronPython 2.0
In-Reply-To: <350E7D38B6D819428718949920EC2355564B31BB5F@NA-EXMSG-C102.redmond.corp.microsoft.com>
References: <4f428e7b0812090559l381860f4pe344a564fc721214@mail.gmail.com>
	<350E7D38B6D819428718949920EC2355564B31BB5F@NA-EXMSG-C102.redmond.corp.microsoft.com>
Message-ID: <493FC035.30607@voidspace.org.uk>

Dino Viehland wrote:
>
> I don?t believe there is right now w/o a bunch of work to update to 
> the new 2.0 hosting APIs.
>

IronPython Studio has looked suspiciously like abandonware pretty much 
ever since its release.

Michael

> *From:* users-bounces at lists.ironpython.com 
> [mailto:users-bounces at lists.ironpython.com] *On Behalf Of *Yash Ganthe
> *Sent:* Tuesday, December 09, 2008 5:59 AM
> *To:* users at lists.ironpython.com
> *Subject:* [IronPython] IronPython Studio and IronPython 2.0
>
> I use IPY 2.0 RC2. I would like to debug through the code in Visual 
> Studio. Can IronPython Studio be configured to use IPY 2.0?
>
> Thanks,
> Yash
>
> ------------------------------------------------------------------------
>
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>   


-- 
http://www.ironpythoninaction.com/
http://www.voidspace.org.uk/blog




From guy at rzn.co.il  Wed Dec 10 19:04:15 2008
From: guy at rzn.co.il (Guy Rozendorn)
Date: Wed, 10 Dec 2008 20:04:15 +0200
Subject: [IronPython] Questions about compiled .py files - their stub exe
	and dlls
Message-ID: <843870680812101004s537b627am8275da5b1273fbd9@mail.gmail.com>

Hi,

I have a question about compiled python files with the pyc.py example.
I read that the dll must exist from the directory from which the executable
is executed.
For example, let's say I compiled my code to:
myProg.exe
myProgr.dll

and they both reside in C:\Temp\myProg.exe, along with IronPython.dll and
IronPythonModules.dll

If I try to execute myProg.exe from any other directory (i.e - I'm in
C:\Windows, and running C:\Temp\myProg.exe), I get the following error:
Unhandled Exception: System.IO.FileNotFoundException: The system cannot find
the file specified. (Exception from HRESULT
: 0x80070002)
   at System.Reflection.Assembly.nLoadFile(String path, Evidence evidence)
   at System.Reflection.Assembly.LoadFile(String path)
   at PythonMain.Main()

What do I need to do to make myProg.exe run from anywhere?
Specfically, this bothers me because in deploying Windows Services wriiten
in IronPython.
Let's say I just compiled myService.exe, along with myService.dll, and
they're both in C:\Temp (along with IronPython DLLs).
After I register the service, it fails to start, since the current directory
of the services.exe (who executes myService.exe) is %SystemRoot%\system32
If I put myService.dll in the system32 directory, the service works.
But I don't want to put myService.dll in system32 (and many more dlls that I
use).

So, my question is:
* Can I make the magic that will cause myService.exe/myProg.exe to look for
its DLL inside the directory in which he resides in (and not the current
directory)?
* It is possible to include myService.dll and IronPython.dll and
IronPythonModules.dll inside myService.exe (so I could do import sys and
then append to path which ever directries I wish for)?

Thanks,
Guy
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081210/37d6301e/attachment.html>

From dfugate at microsoft.com  Wed Dec 10 22:19:18 2008
From: dfugate at microsoft.com (Dave Fugate)
Date: Wed, 10 Dec 2008 13:19:18 -0800
Subject: [IronPython] Announcing IronPython 2.0
Message-ID: <EECFB552375FDE49A23A13CE85EEC2FD1E9834A4D5@NA-EXMSG-C112.redmond.corp.microsoft.com>

Hello Python Community,

The IronPython and Dynamic Language Runtime teams are proud to announce the release of IronPython 2.0 final.  IronPython 2.0 is the culmination of nearly two years worth of work resulting in a CPython 2.5 compatible release on .NET 2.0 SP1.  By far, the biggest change to 2.0 is that our 1.1 codebase was refactored to run on top of the Dynamic Language Runtime<http://www.codeplex.com/dlr>.  With this we automatically get improvements in many feature areas such as better .NET interop support and hosting Python from managed code.  There have been many other major improvements as well.  The most notable are:

*         An MSI installer for Windows platforms which includes parts of the CPython 2.5 standard library

*         IronPython assemblies targeting Silverlight and tools such as Chiron to improve the Silverlight dynamic development experience

*         The addition of more C-based standard modules such as cmath and _winreg

*         Significant improvements in importing compatibility and features

*         Distribution of IronPython under the Microsoft Public License which has been approved by OSI

*         Performance improvements.  On that note, a new Wiki page has been created for IronPython performance reports - see http://www.codeplex.com/IronPython/Wiki/View.aspx?title=IronPython%20Performance

*         Over 500 bugs have been closed in 2.0.  453 of these were reported on CodePlex

*         Support for precompilation of Python source files into a single dll

This seems like an opportune time to remind everyone that we fix bugs based on the number of votes they have on CodePlex.  As we're planning on releasing IronPython 2.0.1 fairly soon, please vote for your favorite bugs at http://www.codeplex.com/IronPython/WorkItem/AdvancedList.aspx to help ensure they get fixed in time for the next release.

We'd like to extend our gratitude to everyone in the IronPython community who reported bugs thereby making this a better release:  milind, romank, chadaustin, sjmachin, davidfraser, TimothyFitz, drewid, sanxiyn, bashmohandes, pobrien, perhaps, haypo, Undebtedly, ayarrow, tscottw_cp, rope, arman0, eshaish, nivaldo, fuzzyman, CurtHagenlocher, Eloff, brucec, py_sunil, jacobg23, mziller, beaugunderson, gbraad, Oceanborn, tarlano, jbevain, glchapman, anthonybaxter, jdhardy, jjlee, haibo, doubleyewdee, jackeyoo, whit537, sdahlbac, PeteHufnagel, jtenney, nriley, junfeng, grizlupo, rridge, lewisle, JoelBondurant, johnplatt, lthompson, debackerl, googen, tscottw, VoteFw, leppie, Qvin, heyssion2, CriGoT, baxeno, sbergman, Laurion, luntain, oldman, christmas, 05031972, kevgu, wilberforce, Korbinian, lclj, sorokerr, Eriol, tatwright, ais, TraumaPony, pelikhan, asafk, felixpollan, srid, atifaziz, vernondcole, fwereade, zpy, yanne, facorreia, Daneel, zvikag, psykotic, Cavingdeep, BEaton, sborde, orbital56, fbourgeois, antont, krosavcheg, ktc1, awilkins, ben2004uk, paulfelix, axl, JeffreySax, Lawouach, and KKI.

You can download IronPython 2.0 at:  http://www.codeplex.com/IronPython/Release/ProjectReleases.aspx?ReleaseId=8365

The IronPython Team


BTW
Updates to the IronPython 1.0 samples will be released shortly.  Stay tuned to the IronPython mailing list for details.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081210/0ca7dd5d/attachment.html>

From dfugate at microsoft.com  Wed Dec 10 22:50:41 2008
From: dfugate at microsoft.com (Dave Fugate)
Date: Wed, 10 Dec 2008 13:50:41 -0800
Subject: [IronPython] Unused files
In-Reply-To: <5b0248170812071352t4221ce3at75c7b57f79bbc1fa@mail.gmail.com>
References: <5b0248170812071352t4221ce3at75c7b57f79bbc1fa@mail.gmail.com>
Message-ID: <EECFB552375FDE49A23A13CE85EEC2FD1E9834A544@NA-EXMSG-C112.redmond.corp.microsoft.com>

Done and this ended up being the only change we took in for IP 2.0 final.  The change should be live on CodePlex's source repository in a few minutes.

Thanks,

Dave

-----Original Message-----
From: users-bounces at lists.ironpython.com [mailto:users-bounces at lists.ironpython.com] On Behalf Of Seo Sanghyeon
Sent: Sunday, December 07, 2008 1:52 PM
To: Discussion of IronPython
Subject: [IronPython] Unused files

Following files seem to be unused but are included in 2.0 RC2 source.
Please remove them before the final release...

Microsoft.Scripting.Core/Actions/RestrictedMetaObject.cs
Microsoft.Scripting.Core/Ast/ExpressionTreeVisitor.Generated.cs
Microsoft.Scripting.Core/Ast/LabeledStatement.cs
Microsoft.Scripting.Core/Com/MetaUnwrappedComObject.cs
IronPython/Runtime/Binding/IPythonCallable.cs
IronPython/Runtime/ThreadLocal.cs

Thanks,

--
Seo Sanghyeon
_______________________________________________
Users mailing list
Users at lists.ironpython.com
http://lists.ironpython.com/listinfo.cgi/users-ironpython.com



From sh at defuze.org  Wed Dec 10 22:53:35 2008
From: sh at defuze.org (Sylvain Hellegouarch)
Date: Wed, 10 Dec 2008 22:53:35 +0100
Subject: [IronPython] Announcing IronPython 2.0
In-Reply-To: <EECFB552375FDE49A23A13CE85EEC2FD1E9834A4D5@NA-EXMSG-C112.redmond.corp.microsoft.com>
References: <EECFB552375FDE49A23A13CE85EEC2FD1E9834A4D5@NA-EXMSG-C112.redmond.corp.microsoft.com>
Message-ID: <49403A5F.2000903@defuze.org>

Congratulations to everyone who made it happen. That's fantastic news :)

- Sylvain

Dave Fugate a ?crit :
>
> Hello Python Community,
>
> The IronPython and Dynamic Language Runtime teams are proud to 
> announce the release of IronPython 2.0 final. IronPython 2.0 is the 
> culmination of nearly two years worth of work resulting in a CPython 
> 2.5 compatible release on .NET 2.0 SP1. By far, the biggest change to 
> 2.0 is that our 1.1 codebase was refactored to run on top of the 
> Dynamic Language Runtime <http://www.codeplex.com/dlr>. With this we 
> automatically get improvements in many feature areas such as better 
> .NET interop support and hosting Python from managed code. There have 
> been many other major improvements as well. The most notable are:
>
> ? An MSI installer for Windows platforms which includes parts of the 
> CPython 2.5 standard library
>
> ? IronPython assemblies targeting Silverlight and tools such as Chiron 
> to improve the Silverlight dynamic development experience
>
> ? The addition of more C-based standard modules such as /cmath/ and 
> /_winreg/
>
> ? Significant improvements in importing compatibility and features
>
> ? Distribution of IronPython under the Microsoft Public License which 
> has been approved by OSI
>
> ? Performance improvements. On that note, a new Wiki page has been 
> created for IronPython performance reports - see 
> http://www.codeplex.com/IronPython/Wiki/View.aspx?title=IronPython%20Performance
>
> ? Over 500 bugs have been closed in 2.0. 453 of these were reported on 
> CodePlex
>
> ? Support for precompilation of Python source files into a single dll
>
> This seems like an opportune time to remind everyone that we fix bugs 
> based on the number of votes they have on CodePlex. As we?re planning 
> on releasing IronPython 2.0.1 fairly soon, please vote for your 
> favorite bugs at 
> http://www.codeplex.com/IronPython/WorkItem/AdvancedList.aspx to help 
> ensure they get fixed in time for the next release.
>
> We?d like to extend our gratitude to everyone in the IronPython 
> community who reported bugs thereby making this a better release: 
> milind, romank, chadaustin, sjmachin, davidfraser, TimothyFitz, 
> drewid, sanxiyn, bashmohandes, pobrien, perhaps, haypo, Undebtedly, 
> ayarrow, tscottw_cp, rope, arman0, eshaish, nivaldo, fuzzyman, 
> CurtHagenlocher, Eloff, brucec, py_sunil, jacobg23, mziller, 
> beaugunderson, gbraad, Oceanborn, tarlano, jbevain, glchapman, 
> anthonybaxter, jdhardy, jjlee, haibo, doubleyewdee, jackeyoo, whit537, 
> sdahlbac, PeteHufnagel, jtenney, nriley, junfeng, grizlupo, rridge, 
> lewisle, JoelBondurant, johnplatt, lthompson, debackerl, googen, 
> tscottw, VoteFw, leppie, Qvin, heyssion2, CriGoT, baxeno, sbergman, 
> Laurion, luntain, oldman, christmas, 05031972, kevgu, wilberforce, 
> Korbinian, lclj, sorokerr, Eriol, tatwright, ais, TraumaPony, 
> pelikhan, asafk, felixpollan, srid, atifaziz, vernondcole, fwereade, 
> zpy, yanne, facorreia, Daneel, zvikag, psykotic, Cavingdeep, BEaton, 
> sborde, orbital56, fbourgeois, antont, krosavcheg, ktc1, awilkins, 
> ben2004uk, paulfelix, axl, JeffreySax, Lawouach, and KKI.
>
> You can download IronPython 2.0 at: 
> http://www.codeplex.com/IronPython/Release/ProjectReleases.aspx?ReleaseId=8365
>
> The IronPython Team
>
> BTW
>
> Updates to the IronPython 1.0 samples will be released shortly. Stay 
> tuned to the IronPython mailing list for details.
>
> ------------------------------------------------------------------------
>
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>   



From nytrokiss at gmail.com  Wed Dec 10 22:58:08 2008
From: nytrokiss at gmail.com (James Matthews)
Date: Wed, 10 Dec 2008 23:58:08 +0200
Subject: [IronPython] Announcing IronPython 2.0
In-Reply-To: <49403A5F.2000903@defuze.org>
References: <EECFB552375FDE49A23A13CE85EEC2FD1E9834A4D5@NA-EXMSG-C112.redmond.corp.microsoft.com>
	<49403A5F.2000903@defuze.org>
Message-ID: <8a6b8e350812101358g41a89489j80405e3e106136b4@mail.gmail.com>

I agree, I am now downloading and installing it!!

On Wed, Dec 10, 2008 at 11:53 PM, Sylvain Hellegouarch <sh at defuze.org>wrote:

> Congratulations to everyone who made it happen. That's fantastic news :)
>
> - Sylvain
>
> Dave Fugate a ?crit :
>
>>
>> Hello Python Community,
>>
>> The IronPython and Dynamic Language Runtime teams are proud to announce
>> the release of IronPython 2.0 final. IronPython 2.0 is the culmination of
>> nearly two years worth of work resulting in a CPython 2.5 compatible release
>> on .NET 2.0 SP1. By far, the biggest change to 2.0 is that our 1.1 codebase
>> was refactored to run on top of the Dynamic Language Runtime <
>> http://www.codeplex.com/dlr>. With this we automatically get improvements
>> in many feature areas such as better .NET interop support and hosting Python
>> from managed code. There have been many other major improvements as well.
>> The most notable are:
>>
>> ? An MSI installer for Windows platforms which includes parts of the
>> CPython 2.5 standard library
>>
>> ? IronPython assemblies targeting Silverlight and tools such as Chiron to
>> improve the Silverlight dynamic development experience
>>
>> ? The addition of more C-based standard modules such as /cmath/ and
>> /_winreg/
>>
>> ? Significant improvements in importing compatibility and features
>>
>> ? Distribution of IronPython under the Microsoft Public License which has
>> been approved by OSI
>>
>> ? Performance improvements. On that note, a new Wiki page has been created
>> for IronPython performance reports - see
>> http://www.codeplex.com/IronPython/Wiki/View.aspx?title=IronPython%20Performance
>>
>> ? Over 500 bugs have been closed in 2.0. 453 of these were reported on
>> CodePlex
>>
>> ? Support for precompilation of Python source files into a single dll
>>
>> This seems like an opportune time to remind everyone that we fix bugs
>> based on the number of votes they have on CodePlex. As we're planning on
>> releasing IronPython 2.0.1 fairly soon, please vote for your favorite bugs
>> at http://www.codeplex.com/IronPython/WorkItem/AdvancedList.aspx to help
>> ensure they get fixed in time for the next release.
>>
>> We'd like to extend our gratitude to everyone in the IronPython community
>> who reported bugs thereby making this a better release: milind, romank,
>> chadaustin, sjmachin, davidfraser, TimothyFitz, drewid, sanxiyn,
>> bashmohandes, pobrien, perhaps, haypo, Undebtedly, ayarrow, tscottw_cp,
>> rope, arman0, eshaish, nivaldo, fuzzyman, CurtHagenlocher, Eloff, brucec,
>> py_sunil, jacobg23, mziller, beaugunderson, gbraad, Oceanborn, tarlano,
>> jbevain, glchapman, anthonybaxter, jdhardy, jjlee, haibo, doubleyewdee,
>> jackeyoo, whit537, sdahlbac, PeteHufnagel, jtenney, nriley, junfeng,
>> grizlupo, rridge, lewisle, JoelBondurant, johnplatt, lthompson, debackerl,
>> googen, tscottw, VoteFw, leppie, Qvin, heyssion2, CriGoT, baxeno, sbergman,
>> Laurion, luntain, oldman, christmas, 05031972, kevgu, wilberforce,
>> Korbinian, lclj, sorokerr, Eriol, tatwright, ais, TraumaPony, pelikhan,
>> asafk, felixpollan, srid, atifaziz, vernondcole, fwereade, zpy, yanne,
>> facorreia, Daneel, zvikag, psykotic, Cavingdeep, BEaton, sborde, orbital56,
>> fbourgeois, antont, krosavcheg, ktc1, awilkins, ben2004uk, paulfelix, axl,
>> JeffreySax, Lawouach, and KKI.
>>
>> You can download IronPython 2.0 at:
>> http://www.codeplex.com/IronPython/Release/ProjectReleases.aspx?ReleaseId=8365
>>
>> The IronPython Team
>>
>> BTW
>>
>> Updates to the IronPython 1.0 samples will be released shortly. Stay tuned
>> to the IronPython mailing list for details.
>>
>> ------------------------------------------------------------------------
>>
>> _______________________________________________
>> Users mailing list
>> Users at lists.ironpython.com
>> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>>
>>
>
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>



-- 
http://www.goldwatches.com/

http://www.jewelerslounge.com/
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081210/4c819218/attachment.html>

From giles.thomas at resolversystems.com  Thu Dec 11 00:48:28 2008
From: giles.thomas at resolversystems.com (Giles Thomas)
Date: Wed, 10 Dec 2008 23:48:28 +0000
Subject: [IronPython] Announcing IronPython 2.0
In-Reply-To: <8a6b8e350812101358g41a89489j80405e3e106136b4@mail.gmail.com>
References: <EECFB552375FDE49A23A13CE85EEC2FD1E9834A4D5@NA-EXMSG-C112.redmond.corp.microsoft.com>	<49403A5F.2000903@defuze.org>
	<8a6b8e350812101358g41a89489j80405e3e106136b4@mail.gmail.com>
Message-ID: <4940554C.30705@resolversystems.com>

Thirded, well done to all!

James Matthews wrote:

> I agree, I am now downloading and installing it!!
>
> On Wed, Dec 10, 2008 at 11:53 PM, Sylvain Hellegouarch <sh at defuze.org 
> <mailto:sh at defuze.org>> wrote:
>
>     Congratulations to everyone who made it happen. That's fantastic
>     news :)
>
>     - Sylvain
>
>     Dave Fugate a ?crit :
>
>
>         Hello Python Community,
>
>         The IronPython and Dynamic Language Runtime teams are proud to
>         announce the release of IronPython 2.0 final. IronPython 2.0
>         is the culmination of nearly two years worth of work resulting
>         in a CPython 2.5 compatible release on .NET 2.0 SP1. By far,
>         the biggest change to 2.0 is that our 1.1 codebase was
>         refactored to run on top of the Dynamic Language Runtime
>         <http://www.codeplex.com/dlr>. With this we automatically get
>         improvements in many feature areas such as better .NET interop
>         support and hosting Python from managed code. There have been
>         many other major improvements as well. The most notable are:
>
>
>         ? An MSI installer for Windows platforms which includes parts
>         of the CPython 2.5 standard library
>
>         ? IronPython assemblies targeting Silverlight and tools such
>         as Chiron to improve the Silverlight dynamic development
>         experience
>
>         ? The addition of more C-based standard modules such as
>         /cmath/ and /_winreg/
>
>         ? Significant improvements in importing compatibility and features
>
>         ? Distribution of IronPython under the Microsoft Public
>         License which has been approved by OSI
>
>         ? Performance improvements. On that note, a new Wiki page has
>         been created for IronPython performance reports - see
>         http://www.codeplex.com/IronPython/Wiki/View.aspx?title=IronPython%20Performance
>
>         ? Over 500 bugs have been closed in 2.0. 453 of these were
>         reported on CodePlex
>
>         ? Support for precompilation of Python source files into a
>         single dll
>
>         This seems like an opportune time to remind everyone that we
>         fix bugs based on the number of votes they have on CodePlex.
>         As we're planning on releasing IronPython 2.0.1 fairly soon,
>         please vote for your favorite bugs at
>         http://www.codeplex.com/IronPython/WorkItem/AdvancedList.aspx
>         to help ensure they get fixed in time for the next release.
>
>         We'd like to extend our gratitude to everyone in the
>         IronPython community who reported bugs thereby making this a
>         better release: milind, romank, chadaustin, sjmachin,
>         davidfraser, TimothyFitz, drewid, sanxiyn, bashmohandes,
>         pobrien, perhaps, haypo, Undebtedly, ayarrow, tscottw_cp,
>         rope, arman0, eshaish, nivaldo, fuzzyman, CurtHagenlocher,
>         Eloff, brucec, py_sunil, jacobg23, mziller, beaugunderson,
>         gbraad, Oceanborn, tarlano, jbevain, glchapman, anthonybaxter,
>         jdhardy, jjlee, haibo, doubleyewdee, jackeyoo, whit537,
>         sdahlbac, PeteHufnagel, jtenney, nriley, junfeng, grizlupo,
>         rridge, lewisle, JoelBondurant, johnplatt, lthompson,
>         debackerl, googen, tscottw, VoteFw, leppie, Qvin, heyssion2,
>         CriGoT, baxeno, sbergman, Laurion, luntain, oldman, christmas,
>         05031972, kevgu, wilberforce, Korbinian, lclj, sorokerr,
>         Eriol, tatwright, ais, TraumaPony, pelikhan, asafk,
>         felixpollan, srid, atifaziz, vernondcole, fwereade, zpy,
>         yanne, facorreia, Daneel, zvikag, psykotic, Cavingdeep,
>         BEaton, sborde, orbital56, fbourgeois, antont, krosavcheg,
>         ktc1, awilkins, ben2004uk, paulfelix, axl, JeffreySax,
>         Lawouach, and KKI.
>
>         You can download IronPython 2.0 at:
>         http://www.codeplex.com/IronPython/Release/ProjectReleases.aspx?ReleaseId=8365
>
>         The IronPython Team
>
>         BTW
>
>         Updates to the IronPython 1.0 samples will be released
>         shortly. Stay tuned to the IronPython mailing list for details.
>
>         ------------------------------------------------------------------------
>
>         _______________________________________________
>         Users mailing list
>         Users at lists.ironpython.com <mailto:Users at lists.ironpython.com>
>         http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>          
>
>
>     _______________________________________________
>     Users mailing list
>     Users at lists.ironpython.com <mailto:Users at lists.ironpython.com>
>     http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>
>
>
>
> -- 
> http://www.goldwatches.com/
>
> http://www.jewelerslounge.com/
>
>------------------------------------------------------------------------
>
>_______________________________________________
>Users mailing list
>Users at lists.ironpython.com
>http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>  
>


From dfugate at microsoft.com  Thu Dec 11 00:53:26 2008
From: dfugate at microsoft.com (Dave Fugate)
Date: Wed, 10 Dec 2008 15:53:26 -0800
Subject: [IronPython] Samples for IronPython 2.0
Message-ID: <EECFB552375FDE49A23A13CE85EEC2FD1E9834A72B@NA-EXMSG-C112.redmond.corp.microsoft.com>

The 1.0 samples that were ported over to 2.0 have now been added to IronPython-2.0-Samples.zip on the 2.0 release page - see http://www.codeplex.com/IronPython/Release/ProjectReleases.aspx?ReleaseId=8365.  If you encounter any issues, please let us know or file a bug on CodePlex.

Please remember that in addition to the samples produced by the IronPython Team, there is also a community-driven IronPython Cookbook at http://www.ironpython.info.

Thanks,

David Fugate
Microsoft - IronPython
http://knowbody.livejournal.com<http://knowbody.livejournal.com/>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081210/426d9da8/attachment.html>

From artie.ziff at gmail.com  Thu Dec 11 01:33:53 2008
From: artie.ziff at gmail.com (Artie Ziff)
Date: Wed, 10 Dec 2008 16:33:53 -0800
Subject: [IronPython] def within a def
Message-ID: <49405FF1.9050701@gmail.com>

Hi,

Apologies if this is actually a general python inquiry. Due to the
dotnet statement related to adding a handler for an event, I will ask here.

Looking at the examples on the "IronPython & Windows Forms" page, I see
a code that I am not familiar. In the 2nd example on this page,
<http://www.voidspace.org.uk/ironpython/winforms/part5.shtml>, I am
unfamiliar with placing a method (in this case "update") inside a
constructor.

I see that the method "update" is being added to the button's event
handler in the constructor, __init__. Why must this method be defined
inside the constructor? Is there a name for this? Is this an example of
a PEP 309, Partial Function Application?

Cheers,
Art


From me at alcidesfonseca.com  Thu Dec 11 01:38:49 2008
From: me at alcidesfonseca.com (Alcides Fonseca)
Date: Thu, 11 Dec 2008 00:38:49 +0000
Subject: [IronPython] def within a def
In-Reply-To: <49405FF1.9050701@gmail.com>
References: <49405FF1.9050701@gmail.com>
Message-ID: <95648DCC-46CE-4BA5-A404-B3553D1DF567@alcidesfonseca.com>


Em 2008/12/11, ?s 00:33, Artie Ziff escreveu:

> Looking at the examples on the "IronPython & Windows Forms" page, I  
> see
> a code that I am not familiar. In the 2nd example on this page,
> <http://www.voidspace.org.uk/ironpython/winforms/part5.shtml>, I am
> unfamiliar with placing a method (in this case "update") inside a
> constructor.
>
> I see that the method "update" is being added to the button's event
> handler in the constructor, __init__. Why must this method be defined
> inside the constructor? Is there a name for this? Is this an example  
> of
> a PEP 309, Partial Function Application?

Actually, it's not a method. Methods take self as the first argument.  
that update is just a function that takes two parameters and prints  
some value. It's the callback of that event.

After you call

self.button1.Click += update

Everytime button1 is clicked, it is going to call the function update.

You can define functions anywhere in your code, inside classes, inside  
other functions, at the module level, etc... Methods are the ones that  
are restricted to be inside the class and taking self as the first  
parameter*

Alcides

* Actually you can monkey patch, but that's an advanced feature.



From billchi at microsoft.com  Thu Dec 11 03:09:41 2008
From: billchi at microsoft.com (Bill Chiles)
Date: Wed, 10 Dec 2008 18:09:41 -0800
Subject: [IronPython] DLR Codeplex Project v0.9 Released with IPy v2.0
Message-ID: <8DD7BECAEB597D4782F7C9CEFDCA6B92867AB38FCE@NA-EXMSG-C125.redmond.corp.microsoft.com>

Today we're excited to announce the final v0.9 release of the Dynamic Language Runtime (DLR).  The DLR project provides one-stop shopping for folks who want to:
*       implement a language on .NET using the DLR
*       add dynamic features to their existing language like C#'s 'dynamic' (http://code.msdn.microsoft.com/Project/Download/FileDownload.aspx?ProjectName=csharpfuture&DownloadId=3550 )
*       add scripting to their applications.
*       create .NET libraries with dynamic objects
This Codeplex project includes the DLR sources as well as sources for IronPython, IronRuby, and samples (such as ToyScript).  This is a classic open source version 0.9 release with high quality code through much of the system.  We're shipping some of this in the .NET 4.0 release coming up.  Some parts are still open to design changes as we solidify those parts for post .NET 4.0, and we'll continue to add more docs and samples.   We will have coordinated releases or sync points across IronPython and IronRuby sites for major releases so that if you're not pulling from the DLR project, you can still see which sources or releases are consistent across our projects.

Jim Hugunin did a great talk at PDC 08 on DLR architecture, concepts, and coding with the .NET 4.0 functionality (http://channel9.msdn.com/pdc2008/TL10/ ).  In his talk, he also outlines what's in .NET 4.0:
*       fast dynamic dispatch with polymorphic inline caching
*       dynamic object interoperability across languages and libraries (including C# 'dynamic' consuming IronPython and IronRuby objects naturally, including any language whose objects participate in the DLR's dynamic object protocol)
*       support for library authors to easily make their model objects consumable with nice looking and lightweight code in languages that support the DLR dynamic object protocol (for example, with C#'s 'dynamic', you can write xml.Customer.Name instead of xml.GetChild("Customer").GetChild("Name")
*       ability to mix binding logic from various languages and library objects in a single dynamic call site cache
*       Expression Trees v2 with support for control flow, assignments, etc.
*       COM IDispatch binding as DLR dynamic objects
The following highlights the support shipping only on Codeplex until a future .NET release:
*       common hosting model for languages built on or supporting the DLR hosting model
*       helpers such as a default .NET binder, complex numbers, tuples
*       more Expression Tree support, such as globals access in hosted scenarios and iterator/generator functions.

All the code is available for what the DLR Team is shipping in .NET 4.0.  Furthermore, we're releasing all the code for LINQ Expression Trees v1 since the DLR merged its ASTs with LINQ trees.  Some of you have noticed this code has been on the IronPython site for a couple of months now.  We will continue indefinitely shipping all of our code open source through .NET 4.0 shipping and future releases of .NET as we move more and more of the DLR into .NET (for example, the common hosting APIs and language implementer APIs).  We have no plan in place at this time for source take back from the community.

It is worth noting that the sources and binaries in the .zip files are circa mid-October, and the daily source tree pushes have several small but pervasive cleanups to the API (name changes), perf improvements, and so on.  We're busy here solidifying final design changes for the DLR parts going into .NET 4.0.  We wanted to have our release match IronPython 2.0, and they had to lock down sources before we did.  The documents in the release speak to the latest sources.  Some of the documents are our internal working specs, but in general we prefer you know where we're planning to end up with the code.  There are no significant cognitive changes in the APIs, so you'll easily map from the docs to the actual code.

For a consistent IronPython release, go to www.codeplex.com/ironpython<http://www.codeplex.com/ironpython> and get their v2.0 RC2.  We won't normally have consistent release across all projects for RCs, betas, etc., and we don't have one for IronRuby today.  When we RTW IPy v2 and DLR v0.9 shortly, we'll have consistent releases for all three projects.

If you want to read some documents without downloading sources or bits, go to http://www.codeplex.com/dlr/Wiki/View.aspx?title=Docs%20and%20specs :
*       dlr-overview.doc
*       dlr-spec-hosting.doc
*       sites-binders-dyn-objs-spec.doc
*       libraries-author-introduction
*       expr-tree-spec.doc

There is a discussion list at http://www.codeplex.com/dlr/Thread/List.aspx (click on the "get email notifications" link).  All mail there goes to dlr at microsoft.com<mailto:dlr at microsoft.com> which has all the folks working on the DLR and our languages on it.

Thanks,
Bill


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081210/79fe4937/attachment.html>

From szport at gmail.com  Thu Dec 11 09:16:23 2008
From: szport at gmail.com (Zaur Shibzoukhov)
Date: Thu, 11 Dec 2008 11:16:23 +0300
Subject: [IronPython] COM Object Issue
In-Reply-To: <350E7D38B6D819428718949920EC2355564A4DF775@NA-EXMSG-C102.redmond.corp.microsoft.com>
References: <e2a7590810270050w30120cb0r5acb66925ea722a7@mail.gmail.com>
	<350E7D38B6D819428718949920EC23554FF6DC08FB@NA-EXMSG-C102.redmond.corp.microsoft.com>
	<350E7D38B6D819428718949920EC23554FF6DC08FE@NA-EXMSG-C102.redmond.corp.microsoft.com>
	<e2a7590811260150u2244de65o4ebe7651e4b166b7@mail.gmail.com>
	<b7eb4a700811260551sa53daf1t19e7680c427417b5@mail.gmail.com>
	<350E7D38B6D819428718949920EC2355564A4DF775@NA-EXMSG-C102.redmond.corp.microsoft.com>
Message-ID: <e2a7590812110016q6e0ee438k2b265cdb11777835@mail.gmail.com>

I just discovered strange thing with COM (IP 2.0):

D:\Downloads\IronPython-2.0-Bin\IronPython-2.0>ipy.exe
-X:ExceptionDetail -X:ShowClrExceptions
IronPython 2.0 (2.0.0.0) on .NET 2.0.50727.1433
Type "help", "copyright", "credits" or "license" for more information.
>>> import System
>>> wt=System.Type.GetTypeFromProgID("Word.Application")
>>> wa=System.Activator.CreateInstance(wt)
>>> wa.Documents.Add()
<Microsoft.Office.Interop.Word.DocumentClass (_Document) object at
0x000000000000002B>
>>> wa.Documents[1]
Error while invoking Item.
   ? Microsoft.Scripting.Com.ComRuntimeHelpers.CheckThrowException(Int32
hresult, ExcepInfo& excepInfo, UInt32 argErr, String message)
   ? _stub_$26##25(Closure , CallSite , DispCallable )
   ? Microsoft.Scripting.Actions.MatchCaller.Call1[T0,TRet](Func`3
target, CallSite site, Object[] args)
   ? Microsoft.Scripting.Actions.CallSite`1.UpdateAndExecute(Object[] args)
   ? Microsoft.Scripting.Actions.UpdateDelegates.Update1[T,T0,TRet](CallSite
site, T0 arg0)
   ? _stub_$25##24(Closure , CallSite , ComObject , Int32 )
   ? Microsoft.Scripting.Actions.MatchCaller.Call2[T0,T1,TRet](Func`4
target, CallSite site, Object[] args)
   ? Microsoft.Scripting.Actions.CallSite`1.UpdateAndExecute(Object[] args)
   ? Microsoft.Scripting.Actions.UpdateDelegates.Update2[T,T0,T1,TRet](CallSite
site, T0 arg0, T1 arg1)
   ? _stub_$24##23(Closure , CallSite , Object , Int32 )
   ? Microsoft.Scripting.Actions.MatchCaller.Call2[T0,T1,TRet](Func`4
target, CallSite site, Object[] args)
   ? Microsoft.Scripting.Actions.CallSite`1.UpdateAndExecute(Object[] args)
   ? Microsoft.Scripting.Actions.UpdateDelegates.Update2[T,T0,T1,TRet](CallSite
site, T0 arg0, T1 arg1)
   ? <module>$23##22(Closure , Scope , LanguageContext )
   ? Microsoft.Scripting.Runtime.OptimizedScriptCode.InvokeTarget(LambdaExpression
code, Scope scope)
   ? Microsoft.Scripting.ScriptCode.Run(Scope scope)
   ? IronPython.Hosting.PythonCommandLine.<>c__DisplayClass1.<RunOneInteraction>b__0()
StandardError: Error while invoking Item.
CLR Exception:
    TargetParameterCountException
:
Error while invoking Item.
>>> wa.Documents[None,1]
<Microsoft.Office.Interop.Word.DocumentClass (_Document) object at
0x000000000000002B>
>>> wa.Documents[100000000000000000000,1]
<Microsoft.Office.Interop.Word.DocumentClass (_Document) object at
0x000000000000002B>

It looks like IP's wa.Documents indexer make wrong call of .Item.

-- Zaur

From giulio.petrucci at gmail.com  Thu Dec 11 09:42:10 2008
From: giulio.petrucci at gmail.com (Giulio Petrucci)
Date: Thu, 11 Dec 2008 09:42:10 +0100
Subject: [IronPython] Newbie Questions
In-Reply-To: <350E7D38B6D819428718949920EC2355564A7FB258@NA-EXMSG-C102.redmond.corp.microsoft.com>
References: <c138ac70812050611i7d96d027vcd0c03c79896e12@mail.gmail.com>
	<350E7D38B6D819428718949920EC2355564A7FB258@NA-EXMSG-C102.redmond.corp.microsoft.com>
Message-ID: <c138ac70812110042w2a1912ddufcce5d59004ce3cb@mail.gmail.com>

Hi Dino,

first of all, thanks for your reply

2008/12/5 Dino Viehland <dinov at microsoft.com>:
> You can add a reference to the assembly and then import both the class and the enum.  For example:
>
> import clr
> clr.AddReference('MyLibrary')
> from MyLibrary import MyClass, MyEnum
> MyClass().SomeMethod(MyEnum.SomeValue)

Ok, I've just tested it and everything seems to work fine.

> The 2nd question... is this an app.exe.config file that you're reading?  If so this might be helpful:
> http://www.devnewsgroups.net/group/microsoft.public.dotnet.framework.windowsforms/topic64732.aspx

I couldn't realize actually the utility of the article. I'll try to
explain my problem wit a litlle bit more of words. I've created a .NET
library which reads some settings in the application configuration
file. The library isn't actually used by any Console/WinForm/whatever
application, but I just want to test it using the IronPython
interactive console. So I created a MyLibrary.dll.config file, I run
the IronPython console and I add the reference to my library. How can
I "force" my library to read the proper .config file? Can I explicitly
load a .config file?

Thanks in advance,
Giulio
-- 
OnAir:
http://www.giuliopetrucci.it
http://www.fujikomonamour.com


From dinov at microsoft.com  Thu Dec 11 09:57:27 2008
From: dinov at microsoft.com (Dino Viehland)
Date: Thu, 11 Dec 2008 00:57:27 -0800
Subject: [IronPython] Newbie Questions
In-Reply-To: <c138ac70812110042w2a1912ddufcce5d59004ce3cb@mail.gmail.com>
References: <c138ac70812050611i7d96d027vcd0c03c79896e12@mail.gmail.com>
	<350E7D38B6D819428718949920EC2355564A7FB258@NA-EXMSG-C102.redmond.corp.microsoft.com>
	<c138ac70812110042w2a1912ddufcce5d59004ce3cb@mail.gmail.com>
Message-ID: <350E7D38B6D819428718949920EC2355564B62DFDF@NA-EXMSG-C102.redmond.corp.microsoft.com>

.NET config doesn't really have any concept of configuration for an individual DLL.  Instead it only has a config for the exe which kicked things off (or actually for the current app domain).  So the concept of a "MyLibrary.dll.config" file doesn't really exist AFAIK.

So that basically leaves you with a few options.  You could add your configuration to an ipy.exe.config file.  Or you could try and force the .NET config APIs to explicitly load a config file (which is what I thought the blog post I linked to showed via ExeConfigurationFileMap/OpenMappedExeConfiguration).  Or your DLL could provide its own configuration mechanism if you really feel like its settings should be independent from the app using it.

But ultimately I think these are all the same problems you'll encounter when using your DLL from any other EXE.

> -----Original Message-----
> From: users-bounces at lists.ironpython.com [mailto:users-
> bounces at lists.ironpython.com] On Behalf Of Giulio Petrucci
> Sent: Thursday, December 11, 2008 12:42 AM
> To: IronPython mailing list
> Subject: Re: [IronPython] Newbie Questions
>
> Hi Dino,
>
> first of all, thanks for your reply
>
> 2008/12/5 Dino Viehland <dinov at microsoft.com>:
> > You can add a reference to the assembly and then import both the
> class and the enum.  For example:
> >
> > import clr
> > clr.AddReference('MyLibrary')
> > from MyLibrary import MyClass, MyEnum
> > MyClass().SomeMethod(MyEnum.SomeValue)
>
> Ok, I've just tested it and everything seems to work fine.
>
> > The 2nd question... is this an app.exe.config file that you're
> reading?  If so this might be helpful:
> >
> http://www.devnewsgroups.net/group/microsoft.public.dotnet.framework.wi
> ndowsforms/topic64732.aspx
>
> I couldn't realize actually the utility of the article. I'll try to
> explain my problem wit a litlle bit more of words. I've created a .NET
> library which reads some settings in the application configuration
> file. The library isn't actually used by any Console/WinForm/whatever
> application, but I just want to test it using the IronPython
> interactive console. So I created a MyLibrary.dll.config file, I run
> the IronPython console and I add the reference to my library. How can
> I "force" my library to read the proper .config file? Can I explicitly
> load a .config file?
>
> Thanks in advance,
> Giulio
> --
> OnAir:
> http://www.giuliopetrucci.it
> http://www.fujikomonamour.com
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com



From giulio.petrucci at gmail.com  Thu Dec 11 10:13:12 2008
From: giulio.petrucci at gmail.com (Giulio Petrucci)
Date: Thu, 11 Dec 2008 10:13:12 +0100
Subject: [IronPython] Newbie Questions
In-Reply-To: <350E7D38B6D819428718949920EC2355564B62DFDF@NA-EXMSG-C102.redmond.corp.microsoft.com>
References: <c138ac70812050611i7d96d027vcd0c03c79896e12@mail.gmail.com>
	<350E7D38B6D819428718949920EC2355564A7FB258@NA-EXMSG-C102.redmond.corp.microsoft.com>
	<c138ac70812110042w2a1912ddufcce5d59004ce3cb@mail.gmail.com>
	<350E7D38B6D819428718949920EC2355564B62DFDF@NA-EXMSG-C102.redmond.corp.microsoft.com>
Message-ID: <c138ac70812110113ta89e6a6wb1838daa32a7e842@mail.gmail.com>

Hi Dino,

2008/12/11 Dino Viehland <dinov at microsoft.com>:
> .NET config doesn't really have any concept of configuration for an individual DLL.  Instead it only has a config for the exe
> which kicked things off (or actually for the current app domain).  So the concept of a "MyLibrary.dll.config" file doesn't
> really exist AFAIK.

Yes, I know. In fact the "MyLibrary.dll.config" is a conventional name
- AFAIK - given by VisualStudio when creating a .config file for a
.dll

> So that basically leaves you with a few options.  You could add your configuration to an ipy.exe.config file.

It's an interesting option... ;-) I'll play a bit around an let you know. But...

> Or you could
> try and force the .NET config APIs to explicitly load a config file (which is what I thought the blog post I linked to showed
> via ExeConfigurationFileMap/OpenMappedExeConfiguration).

...and this is the option I prefer. I tryed to follow the example you
showd but it didn't work. Basically what I need is that each
ConfigurationManager.AppSettings[*] reads the "*" setting from my
config file.
I'll investigate and, if I find anything interesting, I'll let you know.

Thanks,
Giulio

-- 
OnAir:
http://www.giuliopetrucci.it
http://www.fujikomonamour.com


From taozuhong at qq.com  Thu Dec 11 12:33:43 2008
From: taozuhong at qq.com (=?gbk?B?QW5keS5UYW8=?=)
Date: Thu, 11 Dec 2008 19:33:43 +0800
Subject: [IronPython] Announcing IronPython 2.0
Message-ID: <tencent_3226308F1550EEA96ADA15E6@qq.com>

good news, thanks for yours' hard working...
  
 I noticed that the binary package haven't contain the help document, could you tell us what time the latest API document will be released?
  
 thanks
 ------------------
  Andy Tao[???]
 ??????? http://www.zuhong.cn
 ???????????????????

  
  
 ------------------ Original ------------------
  From:  "Dave Fugate"<dfugate at microsoft.com>;
 Date:  Thu, Dec 11, 2008 05:19 AM
 To:  "Discussion of IronPython"<users at lists.ironpython.com>; 
 Cc:  "python-announce-list at python.org"<python-announce-list at python.org>; 
 Subject:  [IronPython] Announcing IronPython 2.0

  
   
Hello Python Community,
 
 
 
The IronPython and Dynamic Language Runtime teams are proud to announce the release of IronPython 2.0 final.  IronPython 2.0 is the culmination of nearly two years worth of work resulting in a CPython 2.5 compatible release on .NET 2.0 SP1.  By far, the biggest change to 2.0 is that our 1.1 codebase was refactored to run on top of the Dynamic Language Runtime.  With this we automatically get improvements in many feature areas such as better ..NET interop support and hosting Python from managed code.  There have been many other major improvements as well.  The most notable are:
 
?        An MSI installer for Windows platforms which includes parts of the CPython 2.5 standard library
 
?        IronPython assemblies targeting Silverlight and tools such as Chiron to improve the Silverlight dynamic development experience
 
?        The addition of more C-based standard modules such as cmath and _winreg
 
?        Significant improvements in importing compatibility and features
 
?        Distribution of IronPython under the Microsoft Public License which has been approved by OSI
 
?        Performance improvements.  On that note, a new Wiki page has been created for IronPython performance reports - see http://www.codeplex.com/IronPython/Wiki/View.aspx?title=IronPython%20Performance
 
?        Over 500 bugs have been closed in 2.0.  453 of these were reported on CodePlex
 
?        Support for precompilation of Python source files into a single dll
 
 
 
This seems like an opportune time to remind everyone that we fix bugs based on the number of votes they have on CodePlex.  As we?re planning on releasing IronPython 2.0.1 fairly soon, please vote for your favorite bugs at http://www.codeplex.com/IronPython/WorkItem/AdvancedList.aspx to help ensure they get fixed in time for the next release.
 
 
 
We?d like to extend our gratitude to everyone in the IronPython community who reported bugs thereby making this a better release: milind, romank, chadaustin, sjmachin, davidfraser, TimothyFitz, drewid, sanxiyn, bashmohandes, pobrien, perhaps, haypo, Undebtedly, ayarrow, tscottw_cp, rope, arman0, eshaish, nivaldo, fuzzyman, CurtHagenlocher, Eloff, brucec, py_sunil, jacobg23, mziller, beaugunderson, gbraad, Oceanborn, tarlano, jbevain, glchapman, anthonybaxter, jdhardy, jjlee, haibo, doubleyewdee, jackeyoo, whit537, sdahlbac, PeteHufnagel, jtenney, nriley, junfeng, grizlupo, rridge, lewisle, JoelBondurant, johnplatt, lthompson, debackerl, googen, tscottw, VoteFw, leppie, Qvin, heyssion2, CriGoT, baxeno, sbergman, Laurion, luntain, oldman, christmas, 05031972, kevgu, wilberforce, Korbinian, lclj, sorokerr, Eriol, tatwright, ais, TraumaPony, pelikhan, asafk, felixpollan, srid, atifaziz, vernondcole, fwereade, zpy, yanne, facorreia, Daneel, zvikag, psykotic, Cavingdeep, BEaton, sborde, orbital56, fbourgeois, antont, krosavcheg, ktc1, awilkins, ben2004uk, paulfelix, axl, JeffreySax, Lawouach, and KKI.
 
 
 
You can download IronPython 2.0 at:  http://www.codeplex.com/IronPython/Release/ProjectReleases.aspx?ReleaseId=8365
 
 
 
The IronPython Team
 
 
 
 
 
BTW
 
Updates to the IronPython 1.0 samples will be released shortly.  Stay tuned to the IronPython mailing list for details.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081211/de00ae6e/attachment.html>

From jslutter at reactorzero.com  Thu Dec 11 16:58:33 2008
From: jslutter at reactorzero.com (Jeff Slutter)
Date: Thu, 11 Dec 2008 10:58:33 -0500
Subject: [IronPython] Announcing IronPython 2.0
In-Reply-To: <EECFB552375FDE49A23A13CE85EEC2FD1E9834A4D5@NA-EXMSG-C112.redmond.corp.microsoft.com>
References: <EECFB552375FDE49A23A13CE85EEC2FD1E9834A4D5@NA-EXMSG-C112.redmond.corp.microsoft.com>
Message-ID: <494138A9.6040504@reactorzero.com>

Excuse me if this is known or not, but searching around has only turned
up articles/mailing lists posts that are two years old, but does
IronPython 2.0 support .NET attributes yet?

I've seen some examples of doing it using stub classes in C# and then
deriving from the in Python, but I'm not sure if this will work for me.
I need to put an Attribute on a class, or method, in Python that takes a
string argument. I can't think of a workaround to get this support.

For example, in C# the class would be:

class Foo
{
     [MyNameAttribute("Test name")]
     public void Method()
     {
           Console.Write("test");
     }
}

And the string "Test name" could be different for each use of
MyNameAttribute. I use it for internal registration purposes in my C# code.

How would I write this class in IP?





From curt at hagenlocher.org  Thu Dec 11 17:46:49 2008
From: curt at hagenlocher.org (Curt Hagenlocher)
Date: Thu, 11 Dec 2008 08:46:49 -0800
Subject: [IronPython] Announcing IronPython 2.0
In-Reply-To: <494138A9.6040504@reactorzero.com>
References: <EECFB552375FDE49A23A13CE85EEC2FD1E9834A4D5@NA-EXMSG-C112.redmond.corp.microsoft.com>
	<494138A9.6040504@reactorzero.com>
Message-ID: <d2155e360812110846r3caa0f2ct7a05c5f4fae59087@mail.gmail.com>

We don't support this in IronPython 2.0, no.  And oddly, no one seems to
have added this as a "desired feature" in CodePlex, even though it's come up
a few times before on this list.
The most likely way that we'd initially provide support for this is to let
you override the "CLS backing class" that's being used to represent your
Python class -- this would mean that you'd need to do the code generation
for the attribute yourself, though.

On Thu, Dec 11, 2008 at 7:58 AM, Jeff Slutter <jslutter at reactorzero.com>wrote:

> Excuse me if this is known or not, but searching around has only turned
> up articles/mailing lists posts that are two years old, but does
> IronPython 2.0 support .NET attributes yet?
>
> I've seen some examples of doing it using stub classes in C# and then
> deriving from the in Python, but I'm not sure if this will work for me.
> I need to put an Attribute on a class, or method, in Python that takes a
> string argument. I can't think of a workaround to get this support.
>
> For example, in C# the class would be:
>
> class Foo
> {
>     [MyNameAttribute("Test name")]
>     public void Method()
>     {
>           Console.Write("test");
>     }
> }
>
> And the string "Test name" could be different for each use of
> MyNameAttribute. I use it for internal registration purposes in my C# code.
>
> How would I write this class in IP?
>
>
>
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081211/199eb56c/attachment.html>

From billchi at microsoft.com  Thu Dec 11 21:03:47 2008
From: billchi at microsoft.com (Bill Chiles)
Date: Thu, 11 Dec 2008 12:03:47 -0800
Subject: [IronPython] Announcing IronPython 2.0
In-Reply-To: <tencent_3226308F1550EEA96ADA15E6@qq.com>
References: <tencent_3226308F1550EEA96ADA15E6@qq.com>
Message-ID: <8DD7BECAEB597D4782F7C9CEFDCA6B92867AB3941A@NA-EXMSG-C125.redmond.corp.microsoft.com>

You can always get the latest docs from http://www.codeplex.com/dlr/Wiki/View.aspx?title=Docs%20and%20specs&referringTitle=Home where I push specs and docs.  IPy specific docs like the tutorial and whatnot are only in the IPy zips.

bill

From: users-bounces at lists.ironpython.com [mailto:users-bounces at lists.ironpython.com] On Behalf Of Andy.Tao
Sent: Thursday, December 11, 2008 3:34 AM
To: Discussion of IronPython
Subject: Re: [IronPython] Announcing IronPython 2.0

good news, thanks for yours' hard working...

I noticed that the binary package haven't contain the help document, could you tell us what time the latest API document will be released?

thanks
------------------
Andy Tao[???]
??????? http://www.zuhong.cn<http://www.zuhong.cn/>
???????????????????


------------------ Original ------------------
From:  "Dave Fugate"<dfugate at microsoft.com>;
Date:  Thu, Dec 11, 2008 05:19 AM
To:  "Discussion of IronPython"<users at lists.ironpython.com>;
Cc:  "python-announce-list at python.org"<python-announce-list at python.org>;
Subject:  [IronPython] Announcing IronPython 2.0

Hello Python Community,

The IronPython and Dynamic Language Runtime teams are proud to announce the release of IronPython 2.0 final.  IronPython 2.0 is the culmination of nearly two years worth of work resulting in a CPython 2.5 compatible release on .NET 2.0 SP1.  By far, the biggest change to 2.0 is that our 1.1 codebase was refactored to run on top of the Dynamic Language Runtime<http://www.codeplex.com/dlr>.  With this we automatically get improvements in many feature areas such as better ..NET interop support and hosting Python from managed code.  There have been many other major improvements as well.  The most notable are:

*        An MSI installer for Windows platforms which includes parts of the CPython 2.5 standard library

*        IronPython assemblies targeting Silverlight and tools such as Chiron to improve the Silverlight dynamic development experience

*        The addition of more C-based standard modules such as cmath and _winreg

*        Significant improvements in importing compatibility and features

*        Distribution of IronPython under the Microsoft Public License which has been approved by OSI

*        Performance improvements.  On that note, a new Wiki page has been created for IronPython performance reports - see http://www.codeplex.com/IronPython/Wiki/View.aspx?title=IronPython%20Performance

*        Over 500 bugs have been closed in 2.0.  453 of these were reported on CodePlex

*        Support for precompilation of Python source files into a single dll

This seems like an opportune time to remind everyone that we fix bugs based on the number of votes they have on CodePlex.  As we?re planning on releasing IronPython 2.0.1 fairly soon, please vote for your favorite bugs at http://www.codeplex.com/IronPython/WorkItem/AdvancedList.aspx to help ensure they get fixed in time for the next release.

We?d like to extend our gratitude to everyone in the IronPython community who reported bugs thereby making this a better release: milind, romank, chadaustin, sjmachin, davidfraser, TimothyFitz, drewid, sanxiyn, bashmohandes, pobrien, perhaps, haypo, Undebtedly, ayarrow, tscottw_cp, rope, arman0, eshaish, nivaldo, fuzzyman, CurtHagenlocher, Eloff, brucec, py_sunil, jacobg23, mziller, beaugunderson, gbraad, Oceanborn, tarlano, jbevain, glchapman, anthonybaxter, jdhardy, jjlee, haibo, doubleyewdee, jackeyoo, whit537, sdahlbac, PeteHufnagel, jtenney, nriley, junfeng, grizlupo, rridge, lewisle, JoelBondurant, johnplatt, lthompson, debackerl, googen, tscottw, VoteFw, leppie, Qvin, heyssion2, CriGoT, baxeno, sbergman, Laurion, luntain, oldman, christmas, 05031972, kevgu, wilberforce, Korbinian, lclj, sorokerr, Eriol, tatwright, ais, TraumaPony, pelikhan, asafk, felixpollan, srid, atifaziz, vernondcole, fwereade, zpy, yanne, facorreia, Daneel, zvikag, psykotic, Cavingdeep, BEaton, sborde, orbital56, fbourgeois, antont, krosavcheg, ktc1, awilkins, ben2004uk, paulfelix, axl, JeffreySax, Lawouach, and KKI.

You can download IronPython 2.0 at:  http://www.codeplex.com/IronPython/Release/ProjectReleases.aspx?ReleaseId=8365

The IronPython Team


BTW
Updates to the IronPython 1.0 samples will be released shortly.  Stay tuned to the IronPython mailing list for details.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081211/63d64d38/attachment.html>

From kfarmer at thuban.org  Thu Dec 11 23:05:13 2008
From: kfarmer at thuban.org (Keith J. Farmer)
Date: Thu, 11 Dec 2008 14:05:13 -0800
Subject: [IronPython] Announcing IronPython 2.0
References: <tencent_3226308F1550EEA96ADA15E6@qq.com>
	<8DD7BECAEB597D4782F7C9CEFDCA6B92867AB3941A@NA-EXMSG-C125.redmond.corp.microsoft.com>
Message-ID: <EEDCD954DD4A0D488BBDBDDCC26075442CC947@meta-exchange.circle-cross.home>

Actually, IP has error messages about help that refer to things that don't exist -- such as the CHM help included in other distributions.  At this point, I suspect IP is emulating CPython behavior too closely ;)
 
So congrats on v2!  I recall there was talk long ago about a PyPan-like service (including dependency tracking and such).  I wonder if that's gone anywhere?

________________________________

From: users-bounces at lists.ironpython.com on behalf of Bill Chiles
Sent: Thu 12/11/2008 12:03 PM
To: Discussion of IronPython
Subject: Re: [IronPython] Announcing IronPython 2.0



You can always get the latest docs from http://www.codeplex.com/dlr/Wiki/View.aspx?title=Docs%20and%20specs&referringTitle=Home where I push specs and docs.  IPy specific docs like the tutorial and whatnot are only in the IPy zips.

 

bill

 

From: users-bounces at lists.ironpython.com [mailto:users-bounces at lists.ironpython.com] On Behalf Of Andy.Tao
Sent: Thursday, December 11, 2008 3:34 AM
To: Discussion of IronPython
Subject: Re: [IronPython] Announcing IronPython 2.0

 

good news, thanks for yours' hard working...

 

I noticed that the binary package haven't contain the help document, could you tell us what time the latest API document will be released?

 

thanks

------------------

Andy Tao[???]

??????? http://www.zuhong.cn <http://www.zuhong.cn/> 

????,????,????????!

 

 

------------------ Original ------------------

From:  "Dave Fugate"<dfugate at microsoft.com>;

Date:  Thu, Dec 11, 2008 05:19 AM

To:  "Discussion of IronPython"<users at lists.ironpython.com>; 

Cc:  "python-announce-list at python.org"<python-announce-list at python.org>; 

Subject:  [IronPython] Announcing IronPython 2.0

 

Hello Python Community,

 

The IronPython and Dynamic Language Runtime teams are proud to announce the release of IronPython 2.0 final.  IronPython 2.0 is the culmination of nearly two years worth of work resulting in a CPython 2.5 compatible release on .NET 2.0 SP1.  By far, the biggest change to 2.0 is that our 1.1 codebase was refactored to run on top of the Dynamic Language Runtime <http://www.codeplex.com/dlr> .  With this we automatically get improvements in many feature areas such as better ..NET interop support and hosting Python from managed code.  There have been many other major improvements as well.  The most notable are:

?        An MSI installer for Windows platforms which includes parts of the CPython 2.5 standard library

?        IronPython assemblies targeting Silverlight and tools such as Chiron to improve the Silverlight dynamic development experience

?        The addition of more C-based standard modules such as cmath and _winreg

?        Significant improvements in importing compatibility and features

?        Distribution of IronPython under the Microsoft Public License which has been approved by OSI

?        Performance improvements.  On that note, a new Wiki page has been created for IronPython performance reports - see http://www.codeplex.com/IronPython/Wiki/View.aspx?title=IronPython%20Performance

?        Over 500 bugs have been closed in 2.0.  453 of these were reported on CodePlex

?        Support for precompilation of Python source files into a single dll

 

This seems like an opportune time to remind everyone that we fix bugs based on the number of votes they have on CodePlex.  As we're planning on releasing IronPython 2.0.1 fairly soon, please vote for your favorite bugs at http://www.codeplex.com/IronPython/WorkItem/AdvancedList.aspx to help ensure they get fixed in time for the next release.

 

We'd like to extend our gratitude to everyone in the IronPython community who reported bugs thereby making this a better release: milind, romank, chadaustin, sjmachin, davidfraser, TimothyFitz, drewid, sanxiyn, bashmohandes, pobrien, perhaps, haypo, Undebtedly, ayarrow, tscottw_cp, rope, arman0, eshaish, nivaldo, fuzzyman, CurtHagenlocher, Eloff, brucec, py_sunil, jacobg23, mziller, beaugunderson, gbraad, Oceanborn, tarlano, jbevain, glchapman, anthonybaxter, jdhardy, jjlee, haibo, doubleyewdee, jackeyoo, whit537, sdahlbac, PeteHufnagel, jtenney, nriley, junfeng, grizlupo, rridge, lewisle, JoelBondurant, johnplatt, lthompson, debackerl, googen, tscottw, VoteFw, leppie, Qvin, heyssion2, CriGoT, baxeno, sbergman, Laurion, luntain, oldman, christmas, 05031972, kevgu, wilberforce, Korbinian, lclj, sorokerr, Eriol, tatwright, ais, TraumaPony, pelikhan, asafk, felixpollan, srid, atifaziz, vernondcole, fwereade, zpy, yanne, facorreia, Daneel, zvikag, psykotic, Cavingdeep, BEaton, sborde, orbital56, fbourgeois, antont, krosavcheg, ktc1, awilkins, ben2004uk, paulfelix, axl, JeffreySax, Lawouach, and KKI.

 

You can download IronPython 2.0 at:  http://www.codeplex.com/IronPython/Release/ProjectReleases.aspx?ReleaseId=8365

 

The IronPython Team

 

 

BTW

Updates to the IronPython 1.0 samples will be released shortly.  Stay tuned to the IronPython mailing list for details.

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081211/180bec4d/attachment.html>

From mtraudt at alum.mit.edu  Fri Dec 12 14:48:51 2008
From: mtraudt at alum.mit.edu (Mark Traudt)
Date: Fri, 12 Dec 2008 05:48:51 -0800 (PST)
Subject: [IronPython] Problem inheriting from C# abstract base class
 with overloaded constructors
In-Reply-To: <350E7D38B6D819428718949920EC2355564A70AA3E@NA-EXMSG-C102.redmond.corp.microsoft.com>
References: <20804327.post@talk.nabble.com>
	<350E7D38B6D819428718949920EC2355564A70AA3E@NA-EXMSG-C102.redmond.corp.microsoft.com>
Message-ID: <20975961.post@talk.nabble.com>



Dino Viehland-2 wrote:
> 
> This is bug #20021 -
> http://www.codeplex.com/IronPython/WorkItem/View.aspx?WorkItemId=20021
> 
> I checked a fix for this in today into the Main branch (which is 2.1).  I
> think that should show up on CodePlex tonight or tomorrow if you're
> willing to build IronPython from source.  If not we'll definitely fix this
> in 2.0.1 if we don't spin a new build for 2.0 which would also likely
> include the fix.
> 
> 

I built Main from latest source and the problem is fixed.  Thanks.


-- 
View this message in context: http://www.nabble.com/Problem-inheriting-from-C--abstract-base-class-with-overloaded-constructors-tp20804327p20975961.html
Sent from the IronPython mailing list archive at Nabble.com.



From glenn.k.jones+ipy at gmail.com  Fri Dec 12 16:27:27 2008
From: glenn.k.jones+ipy at gmail.com (Glenn Jones)
Date: Fri, 12 Dec 2008 15:27:27 +0000
Subject: [IronPython] Issue about string.upper and string.lower
Message-ID: <2679f6510812120727o58a85ec4ub686aeac4dc4820b@mail.gmail.com>

Hello everybody,

We ran across
http://www.codeplex.com/IronPython/WorkItem/View.aspx?WorkItemId=13629(turkish
collation issues) today while trying to port Resolver One to
IronPython 2.0.

This is in a test that tries to import decimal while in the turkish locale
(it was actually reported by a user!). It does an .upper on a string with an
'i', and that doesn't give the expected results.

This is a very big issue because all the python code out there expects
string transformations to be locale-independent, and there may be strange
bugs in strange places.

Is mapping .upper and .lower to ToUpperInvariant and ToLowerInvariant an
acceptable solution? People that want to do locale-dependent transformations
can always use the .NET specific ToUpper/ToLower.

We can work around the decimal being unimportable by hacking it, but clearly
this is not a general solution. We will report other modules that might fail
from this as we find them.

Thanks,
Glenn and Orestis
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081212/6c7fd4df/attachment.html>

From glenn.k.jones+ipy at gmail.com  Fri Dec 12 17:27:10 2008
From: glenn.k.jones+ipy at gmail.com (Glenn Jones)
Date: Fri, 12 Dec 2008 16:27:10 +0000
Subject: [IronPython] Compiled dlls and builtin modules
Message-ID: <2679f6510812120827x173452fbg97ba9ac332d63e92@mail.gmail.com>

Hi all,

We are compiling the python standard library (so it can be ngend when it is
shipped with Resolver One) and we've discovered that modules in the dll take
precedence over builtins. We noticed this because re and socket are included
in the standard Lib directory (installed with IPy2) and when we ran against
the dll built from the Lib, importing them failed.

We have removed the following files from the standard library (and we may
have to remove some others):
socket, copy_reg, re, sre_compile, sre_constants, sre_parse (the sre_* were
referenced in re)

A better solution would be for builtin modules to take precedence over
modules in referenced assemblies.

As an experiment, we created a datetime module that just printed that it was
being imported. When we included it in a dll and referenced it, it replaced
the builtin datetime.

Glenn & Orestis
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081212/227563e7/attachment.html>

From fuzzyman at voidspace.org.uk  Sat Dec 13 15:26:28 2008
From: fuzzyman at voidspace.org.uk (Michael Foord)
Date: Sat, 13 Dec 2008 14:26:28 +0000
Subject: [IronPython] Questions about compiled .py files - their stub
 exe	and dlls
In-Reply-To: <843870680812101004s537b627am8275da5b1273fbd9@mail.gmail.com>
References: <843870680812101004s537b627am8275da5b1273fbd9@mail.gmail.com>
Message-ID: <4943C614.8020705@voidspace.org.uk>

Guy Rozendorn wrote:
> Hi,
>
> I have a question about compiled python files with the pyc.py example.
> I read that the dll must exist from the directory from which the 
> executable is executed.
> For example, let's say I compiled my code to:
> myProg.exe
> myProgr.dll
>
> and they both reside in C:\Temp\myProg.exe, along with IronPython.dll 
> and IronPythonModules.dll

It looks like you are using the Pyc.py sample with IronPython 1. Can you 
update to IronPython 2 and try again.

Thanks

Michael Foord

>
> If I try to execute myProg.exe from any other directory (i.e - I'm in 
> C:\Windows, and running C:\Temp\myProg.exe), I get the following error:
> Unhandled Exception: System.IO.FileNotFoundException: The system 
> cannot find the file specified. (Exception from HRESULT
> : 0x80070002)
>    at System.Reflection.Assembly.nLoadFile(String path, Evidence evidence)
>    at System.Reflection.Assembly.LoadFile(String path)
>    at PythonMain.Main()
>
> What do I need to do to make myProg.exe run from anywhere?
> Specfically, this bothers me because in deploying Windows Services 
> wriiten in IronPython.
> Let's say I just compiled myService.exe, along with myService.dll, and 
> they're both in C:\Temp (along with IronPython DLLs).
> After I register the service, it fails to start, since the current 
> directory of the services.exe (who executes myService.exe) is 
> %SystemRoot%\system32
> If I put myService.dll in the system32 directory, the service works.
> But I don't want to put myService.dll in system32 (and many more dlls 
> that I use).
>
> So, my question is:
> * Can I make the magic that will cause myService.exe/myProg.exe to 
> look for its DLL inside the directory in which he resides in (and not 
> the current directory)?
> * It is possible to include myService.dll and IronPython.dll and 
> IronPythonModules.dll inside myService.exe (so I could do import sys 
> and then append to path which ever directries I wish for)?
>
> Thanks,
> Guy
> ------------------------------------------------------------------------
>
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>   


-- 
http://www.ironpythoninaction.com/
http://www.voidspace.org.uk/blog




From guy at rzn.co.il  Sat Dec 13 15:40:31 2008
From: guy at rzn.co.il (Guy Rozendorn)
Date: Sat, 13 Dec 2008 16:40:31 +0200
Subject: [IronPython] Questions about compiled .py files - their stub
	exe and dlls
In-Reply-To: <4943C614.8020705@voidspace.org.uk>
References: <843870680812101004s537b627am8275da5b1273fbd9@mail.gmail.com>
	<4943C614.8020705@voidspace.org.uk>
Message-ID: <843870680812130640i639e7692h8506137188d8528f@mail.gmail.com>

I'm using the IronPython 2.0, final version, and the corresponding pyc.py
sample


>> and they both reside in C:\Temp\myProg.exe, along with IronPython.dll and
>> IronPythonModules.dll
>>
>
> It looks like you are using the Pyc.py sample with IronPython 1. Can you
> update to IronPython 2 and try again.
>
> Thanks
>
> Michael Foord
>
>
>> If I try to execute myProg.exe from any other directory (i.e - I'm in
>> C:\Windows, and running C:\Temp\myProg.exe), I get the following error:
>> Unhandled Exception: System.IO.FileNotFoundException: The system cannot
>> find the file specified. (Exception from HRESULT
>> : 0x80070002)
>>   at System.Reflection.Assembly.nLoadFile(String path, Evidence evidence)
>>   at System.Reflection.Assembly.LoadFile(String path)
>>   at PythonMain.Main()
>>
>> What do I need to do to make myProg.exe run from anywhere?
>> Specfically, this bothers me because in deploying Windows Services wriiten
>> in IronPython.
>> Let's say I just compiled myService.exe, along with myService.dll, and
>> they're both in C:\Temp (along with IronPython DLLs).
>> After I register the service, it fails to start, since the current
>> directory of the services.exe (who executes myService.exe) is
>> %SystemRoot%\system32
>> If I put myService.dll in the system32 directory, the service works.
>> But I don't want to put myService.dll in system32 (and many more dlls that
>> I use).
>>
>> So, my question is:
>> * Can I make the magic that will cause myService.exe/myProg.exe to look
>> for its DLL inside the directory in which he resides in (and not the current
>> directory)?
>> * It is possible to include myService.dll and IronPython.dll and
>> IronPythonModules.dll inside myService.exe (so I could do import sys and
>> then append to path which ever directries I wish for)?
>>
>> Thanks,
>> Guy
>> ------------------------------------------------------------------------
>>
>> _______________________________________________
>> Users mailing list
>> Users at lists.ironpython.com
>> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>>
>>
>
>
> --
> http://www.ironpythoninaction.com/
> http://www.voidspace.org.uk/blog
>
>
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081213/da1d539b/attachment.html>

From fuzzyman at voidspace.org.uk  Sat Dec 13 15:43:25 2008
From: fuzzyman at voidspace.org.uk (Michael Foord)
Date: Sat, 13 Dec 2008 14:43:25 +0000
Subject: [IronPython] Questions about compiled .py files - their stub
 exe and dlls
In-Reply-To: <843870680812130640i639e7692h8506137188d8528f@mail.gmail.com>
References: <843870680812101004s537b627am8275da5b1273fbd9@mail.gmail.com>	<4943C614.8020705@voidspace.org.uk>
	<843870680812130640i639e7692h8506137188d8528f@mail.gmail.com>
Message-ID: <4943CA0D.80907@voidspace.org.uk>

Guy Rozendorn wrote:
> I'm using the IronPython 2.0, final version, and the corresponding pyc.py
Well, in which case you need to include *all* the IronPython assemblies 
and not just IronPython.dll and IronPythonModules.dll.

All the best,

Michael Foord

>
>
>         and they both reside in C:\Temp\myProg.exe, along with
>         IronPython.dll and IronPythonModules.dll
>
>
>     It looks like you are using the Pyc.py sample with IronPython 1.
>     Can you update to IronPython 2 and try again.
>
>     Thanks
>
>     Michael Foord
>
>
>         If I try to execute myProg.exe from any other directory (i.e -
>         I'm in C:\Windows, and running C:\Temp\myProg.exe), I get the
>         following error:
>         Unhandled Exception: System.IO.FileNotFoundException: The
>         system cannot find the file specified. (Exception from HRESULT
>         : 0x80070002)
>           at System.Reflection.Assembly.nLoadFile(String path,
>         Evidence evidence)
>           at System.Reflection.Assembly.LoadFile(String path)
>           at PythonMain.Main()
>
>         What do I need to do to make myProg.exe run from anywhere?
>         Specfically, this bothers me because in deploying Windows
>         Services wriiten in IronPython.
>         Let's say I just compiled myService.exe, along with
>         myService.dll, and they're both in C:\Temp (along with
>         IronPython DLLs).
>         After I register the service, it fails to start, since the
>         current directory of the services.exe (who executes
>         myService.exe) is %SystemRoot%\system32
>         If I put myService.dll in the system32 directory, the service
>         works.
>         But I don't want to put myService.dll in system32 (and many
>         more dlls that I use).
>
>         So, my question is:
>         * Can I make the magic that will cause
>         myService.exe/myProg.exe to look for its DLL inside the
>         directory in which he resides in (and not the current directory)?
>         * It is possible to include myService.dll and IronPython.dll
>         and IronPythonModules.dll inside myService.exe (so I could do
>         import sys and then append to path which ever directries I
>         wish for)?
>
>         Thanks,
>         Guy
>         ------------------------------------------------------------------------
>
>         _______________________________________________
>         Users mailing list
>         Users at lists.ironpython.com <mailto:Users at lists.ironpython.com>
>         http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>          
>
>
>
>     -- 
>     http://www.ironpythoninaction.com/
>     http://www.voidspace.org.uk/blog
>
>
>     _______________________________________________
>     Users mailing list
>     Users at lists.ironpython.com <mailto:Users at lists.ironpython.com>
>     http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>
>
> ------------------------------------------------------------------------
>
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>   


-- 
http://www.ironpythoninaction.com/
http://www.voidspace.org.uk/blog




From guy at rzn.co.il  Sat Dec 13 15:56:20 2008
From: guy at rzn.co.il (Guy Rozendorn)
Date: Sat, 13 Dec 2008 16:56:20 +0200
Subject: [IronPython] Questions about compiled .py files - their stub
	exe and dlls
In-Reply-To: <4943CA0D.80907@voidspace.org.uk>
References: <843870680812101004s537b627am8275da5b1273fbd9@mail.gmail.com>
	<4943C614.8020705@voidspace.org.uk>
	<843870680812130640i639e7692h8506137188d8528f@mail.gmail.com>
	<4943CA0D.80907@voidspace.org.uk>
Message-ID: <843870680812130656s44e39cf7w8a78f8f7a9ab11bb@mail.gmail.com>

Include them where?

Let's look at the following example:
bar.py only has one line: print dir()
I compile using pyc.py to bar.dll and bar.exe
I place bar* and IronPyhton*.dll in C:\Temp\yyy\
If I execute bar.exe from insdie C:\Temp\yyy, it runs without problems.
here's the output:
C:\Temp\yyy\
bar.exe
['__builtins__', '__file__', '__name__']

But, if I try to execute bar.exe from any other directory (for example,
C:\Temp\), it fails:
C:\Temp\
yyy\bar.exe

Unhandled Exception: System.IO.FileNotFoundException: The system cannot find
the file specified. (Exception from HRESULT: 0x80070002)
   at System.Reflection.Assembly.nLoadFile(String path, Evidence evidence)
   at System.Reflection.Assembly.LoadFile(String path)
   at PythonMain.Main()

====

All I want is to be able to execute the compiled python scripts from
whatever directory I want, and not just the directory in which the binary
and all assemblies are in. Is it possilble?


On Sat, Dec 13, 2008 at 4:43 PM, Michael Foord <fuzzyman at voidspace.org.uk>wrote:

> Guy Rozendorn wrote:
>
>> I'm using the IronPython 2.0, final version, and the corresponding pyc.py
>>
> Well, in which case you need to include *all* the IronPython assemblies and
> not just IronPython.dll and IronPythonModules.dll.
>
> All the best,
>
> Michael Foord
>
>
>>
>>        and they both reside in C:\Temp\myProg.exe, along with
>>        IronPython.dll and IronPythonModules.dll
>>
>>
>>    It looks like you are using the Pyc.py sample with IronPython 1.
>>    Can you update to IronPython 2 and try again.
>>
>>    Thanks
>>
>>    Michael Foord
>>
>>
>>        If I try to execute myProg.exe from any other directory (i.e -
>>        I'm in C:\Windows, and running C:\Temp\myProg.exe), I get the
>>        following error:
>>        Unhandled Exception: System.IO.FileNotFoundException: The
>>        system cannot find the file specified. (Exception from HRESULT
>>        : 0x80070002)
>>          at System.Reflection.Assembly.nLoadFile(String path,
>>        Evidence evidence)
>>          at System.Reflection.Assembly.LoadFile(String path)
>>          at PythonMain.Main()
>>
>>        What do I need to do to make myProg.exe run from anywhere?
>>        Specfically, this bothers me because in deploying Windows
>>        Services wriiten in IronPython.
>>        Let's say I just compiled myService.exe, along with
>>        myService.dll, and they're both in C:\Temp (along with
>>        IronPython DLLs).
>>        After I register the service, it fails to start, since the
>>        current directory of the services.exe (who executes
>>        myService.exe) is %SystemRoot%\system32
>>        If I put myService.dll in the system32 directory, the service
>>        works.
>>        But I don't want to put myService.dll in system32 (and many
>>        more dlls that I use).
>>
>>        So, my question is:
>>        * Can I make the magic that will cause
>>        myService.exe/myProg.exe to look for its DLL inside the
>>        directory in which he resides in (and not the current directory)?
>>        * It is possible to include myService.dll and IronPython.dll
>>        and IronPythonModules.dll inside myService.exe (so I could do
>>        import sys and then append to path which ever directries I
>>        wish for)?
>>
>>        Thanks,
>>        Guy
>>
>>  ------------------------------------------------------------------------
>>
>>        _______________________________________________
>>        Users mailing list
>>        Users at lists.ironpython.com <mailto:Users at lists.ironpython.com>
>>        http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>>
>>
>>
>>    --    http://www.ironpythoninaction.com/
>>    http://www.voidspace.org.uk/blog
>>
>>
>>    _______________________________________________
>>    Users mailing list
>>    Users at lists.ironpython.com <mailto:Users at lists.ironpython.com>
>>    http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>>
>>
>> ------------------------------------------------------------------------
>>
>> _______________________________________________
>> Users mailing list
>> Users at lists.ironpython.com
>> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>>
>>
>
>
> --
> http://www.ironpythoninaction.com/
> http://www.voidspace.org.uk/blog
>
>
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081213/5a18cbdf/attachment.html>

From fuzzyman at voidspace.org.uk  Sat Dec 13 15:57:57 2008
From: fuzzyman at voidspace.org.uk (Michael Foord)
Date: Sat, 13 Dec 2008 14:57:57 +0000
Subject: [IronPython] Questions about compiled .py files - their stub
 exe and dlls
In-Reply-To: <843870680812130656s44e39cf7w8a78f8f7a9ab11bb@mail.gmail.com>
References: <843870680812101004s537b627am8275da5b1273fbd9@mail.gmail.com>	<4943C614.8020705@voidspace.org.uk>	<843870680812130640i639e7692h8506137188d8528f@mail.gmail.com>	<4943CA0D.80907@voidspace.org.uk>
	<843870680812130656s44e39cf7w8a78f8f7a9ab11bb@mail.gmail.com>
Message-ID: <4943CD75.2030101@voidspace.org.uk>

Guy Rozendorn wrote:
> Include them where?
>
> Let's look at the following example:
> bar.py only has one line: print dir()
> I compile using pyc.py to bar.dll and bar.exe
> I place bar* and IronPyhton*.dll in C:\Temp\yyy\

What about the Microsoft.Scripting etc dlls - are you including those?

Michael

> If I execute bar.exe from insdie C:\Temp\yyy, it runs without 
> problems. here's the output:
> C:\Temp\yyy\
> bar.exe
> ['__builtins__', '__file__', '__name__']
>
> But, if I try to execute bar.exe from any other directory (for 
> example, C:\Temp\), it fails:
> C:\Temp\
> yyy\bar.exe
>
> Unhandled Exception: System.IO.FileNotFoundException: The system 
> cannot find the file specified. (Exception from HRESULT: 0x80070002)
>    at System.Reflection.Assembly.nLoadFile(String path, Evidence evidence)
>    at System.Reflection.Assembly.LoadFile(String path)
>    at PythonMain.Main()
>
> ====
>
> All I want is to be able to execute the compiled python scripts from 
> whatever directory I want, and not just the directory in which the 
> binary and all assemblies are in. Is it possilble?
>
>
> On Sat, Dec 13, 2008 at 4:43 PM, Michael Foord 
> <fuzzyman at voidspace.org.uk <mailto:fuzzyman at voidspace.org.uk>> wrote:
>
>     Guy Rozendorn wrote:
>
>         I'm using the IronPython 2.0, final version, and the
>         corresponding pyc.py
>
>     Well, in which case you need to include *all* the IronPython
>     assemblies and not just IronPython.dll and IronPythonModules.dll.
>
>     All the best,
>
>     Michael Foord
>
>
>
>                and they both reside in C:\Temp\myProg.exe, along with
>                IronPython.dll and IronPythonModules.dll
>
>
>            It looks like you are using the Pyc.py sample with
>         IronPython 1.
>            Can you update to IronPython 2 and try again.
>
>            Thanks
>
>            Michael Foord
>
>
>                If I try to execute myProg.exe from any other directory
>         (i.e -
>                I'm in C:\Windows, and running C:\Temp\myProg.exe), I
>         get the
>                following error:
>                Unhandled Exception: System.IO.FileNotFoundException: The
>                system cannot find the file specified. (Exception from
>         HRESULT
>                : 0x80070002)
>                  at System.Reflection.Assembly.nLoadFile(String path,
>                Evidence evidence)
>                  at System.Reflection.Assembly.LoadFile(String path)
>                  at PythonMain.Main()
>
>                What do I need to do to make myProg.exe run from anywhere?
>                Specfically, this bothers me because in deploying Windows
>                Services wriiten in IronPython.
>                Let's say I just compiled myService.exe, along with
>                myService.dll, and they're both in C:\Temp (along with
>                IronPython DLLs).
>                After I register the service, it fails to start, since the
>                current directory of the services.exe (who executes
>                myService.exe) is %SystemRoot%\system32
>                If I put myService.dll in the system32 directory, the
>         service
>                works.
>                But I don't want to put myService.dll in system32 (and many
>                more dlls that I use).
>
>                So, my question is:
>                * Can I make the magic that will cause
>                myService.exe/myProg.exe to look for its DLL inside the
>                directory in which he resides in (and not the current
>         directory)?
>                * It is possible to include myService.dll and
>         IronPython.dll
>                and IronPythonModules.dll inside myService.exe (so I
>         could do
>                import sys and then append to path which ever directries I
>                wish for)?
>
>                Thanks,
>                Guy
>              
>          ------------------------------------------------------------------------
>
>                _______________________________________________
>                Users mailing list
>                Users at lists.ironpython.com
>         <mailto:Users at lists.ironpython.com>
>         <mailto:Users at lists.ironpython.com
>         <mailto:Users at lists.ironpython.com>>
>
>              
>          http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>                
>
>
>            --    http://www.ironpythoninaction.com/
>            http://www.voidspace.org.uk/blog
>
>
>            _______________________________________________
>            Users mailing list
>            Users at lists.ironpython.com
>         <mailto:Users at lists.ironpython.com>
>         <mailto:Users at lists.ironpython.com
>         <mailto:Users at lists.ironpython.com>>
>
>            http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>
>
>         ------------------------------------------------------------------------
>
>         _______________________________________________
>         Users mailing list
>         Users at lists.ironpython.com <mailto:Users at lists.ironpython.com>
>         http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>          
>
>
>
>     -- 
>     http://www.ironpythoninaction.com/
>     http://www.voidspace.org.uk/blog
>
>
>     _______________________________________________
>     Users mailing list
>     Users at lists.ironpython.com <mailto:Users at lists.ironpython.com>
>     http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>
>
> ------------------------------------------------------------------------
>
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>   


-- 
http://www.ironpythoninaction.com/
http://www.voidspace.org.uk/blog




From guy at rzn.co.il  Sat Dec 13 16:49:13 2008
From: guy at rzn.co.il (Guy Rozendorn)
Date: Sat, 13 Dec 2008 17:49:13 +0200
Subject: [IronPython] Questions about compiled .py files - their stub
	exe and dlls
In-Reply-To: <4943CD75.2030101@voidspace.org.uk>
References: <843870680812101004s537b627am8275da5b1273fbd9@mail.gmail.com>
	<4943C614.8020705@voidspace.org.uk>
	<843870680812130640i639e7692h8506137188d8528f@mail.gmail.com>
	<4943CA0D.80907@voidspace.org.uk>
	<843870680812130656s44e39cf7w8a78f8f7a9ab11bb@mail.gmail.com>
	<4943CD75.2030101@voidspace.org.uk>
Message-ID: <843870680812130749n5ff2d172x92227ab7ffee15cc@mail.gmail.com>

Yes, forgot to mention that - also, the Microsoft*.dll form the IronPython
2.0 directory are in C:\Temp\yyy\

Chevy Chase  - "Parrots make great pets. They have more personality than
goldfish."

On Sat, Dec 13, 2008 at 4:57 PM, Michael Foord <fuzzyman at voidspace.org.uk>wrote:

> Guy Rozendorn wrote:
>
>> Include them where?
>>
>> Let's look at the following example:
>> bar.py only has one line: print dir()
>> I compile using pyc.py to bar.dll and bar.exe
>> I place bar* and IronPyhton*.dll in C:\Temp\yyy\
>>
>
> What about the Microsoft.Scripting etc dlls - are you including those?
>
> Michael
>
>  If I execute bar.exe from insdie C:\Temp\yyy, it runs without problems.
>> here's the output:
>> C:\Temp\yyy\
>> bar.exe
>> ['__builtins__', '__file__', '__name__']
>>
>> But, if I try to execute bar.exe from any other directory (for example,
>> C:\Temp\), it fails:
>> C:\Temp\
>> yyy\bar.exe
>>
>> Unhandled Exception: System.IO.FileNotFoundException: The system cannot
>> find the file specified. (Exception from HRESULT: 0x80070002)
>>   at System.Reflection.Assembly.nLoadFile(String path, Evidence evidence)
>>   at System.Reflection.Assembly.LoadFile(String path)
>>   at PythonMain.Main()
>>
>> ====
>>
>> All I want is to be able to execute the compiled python scripts from
>> whatever directory I want, and not just the directory in which the binary
>> and all assemblies are in. Is it possilble?
>>
>>
>> On Sat, Dec 13, 2008 at 4:43 PM, Michael Foord <fuzzyman at voidspace.org.uk<mailto:
>> fuzzyman at voidspace.org.uk>> wrote:
>>
>>    Guy Rozendorn wrote:
>>
>>        I'm using the IronPython 2.0, final version, and the
>>        corresponding pyc.py
>>
>>    Well, in which case you need to include *all* the IronPython
>>    assemblies and not just IronPython.dll and IronPythonModules.dll.
>>
>>    All the best,
>>
>>    Michael Foord
>>
>>
>>
>>               and they both reside in C:\Temp\myProg.exe, along with
>>               IronPython.dll and IronPythonModules.dll
>>
>>
>>           It looks like you are using the Pyc.py sample with
>>        IronPython 1.
>>           Can you update to IronPython 2 and try again.
>>
>>           Thanks
>>
>>           Michael Foord
>>
>>
>>               If I try to execute myProg.exe from any other directory
>>        (i.e -
>>               I'm in C:\Windows, and running C:\Temp\myProg.exe), I
>>        get the
>>               following error:
>>               Unhandled Exception: System.IO.FileNotFoundException: The
>>               system cannot find the file specified. (Exception from
>>        HRESULT
>>               : 0x80070002)
>>                 at System.Reflection.Assembly.nLoadFile(String path,
>>               Evidence evidence)
>>                 at System.Reflection.Assembly.LoadFile(String path)
>>                 at PythonMain.Main()
>>
>>               What do I need to do to make myProg.exe run from anywhere?
>>               Specfically, this bothers me because in deploying Windows
>>               Services wriiten in IronPython.
>>               Let's say I just compiled myService.exe, along with
>>               myService.dll, and they're both in C:\Temp (along with
>>               IronPython DLLs).
>>               After I register the service, it fails to start, since the
>>               current directory of the services.exe (who executes
>>               myService.exe) is %SystemRoot%\system32
>>               If I put myService.dll in the system32 directory, the
>>        service
>>               works.
>>               But I don't want to put myService.dll in system32 (and many
>>               more dlls that I use).
>>
>>               So, my question is:
>>               * Can I make the magic that will cause
>>               myService.exe/myProg.exe to look for its DLL inside the
>>               directory in which he resides in (and not the current
>>        directory)?
>>               * It is possible to include myService.dll and
>>        IronPython.dll
>>               and IronPythonModules.dll inside myService.exe (so I
>>        could do
>>               import sys and then append to path which ever directries I
>>               wish for)?
>>
>>               Thanks,
>>               Guy
>>
>> ------------------------------------------------------------------------
>>
>>               _______________________________________________
>>               Users mailing list
>>               Users at lists.ironpython.com
>>        <mailto:Users at lists.ironpython.com>
>>        <mailto:Users at lists.ironpython.com
>>        <mailto:Users at lists.ironpython.com>>
>>
>>
>> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>>
>>
>>           --    http://www.ironpythoninaction.com/
>>           http://www.voidspace.org.uk/blog
>>
>>
>>           _______________________________________________
>>           Users mailing list
>>           Users at lists.ironpython.com
>>        <mailto:Users at lists.ironpython.com>
>>        <mailto:Users at lists.ironpython.com
>>        <mailto:Users at lists.ironpython.com>>
>>
>>           http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>>
>>
>>
>>  ------------------------------------------------------------------------
>>
>>        _______________________________________________
>>        Users mailing list
>>        Users at lists.ironpython.com <mailto:Users at lists.ironpython.com>
>>        http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>>
>>
>>
>>    --    http://www.ironpythoninaction.com/
>>    http://www.voidspace.org.uk/blog
>>
>>
>>    _______________________________________________
>>    Users mailing list
>>    Users at lists.ironpython.com <mailto:Users at lists.ironpython.com>
>>    http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>>
>>
>> ------------------------------------------------------------------------
>>
>> _______________________________________________
>> Users mailing list
>> Users at lists.ironpython.com
>> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>>
>>
>
>
> --
> http://www.ironpythoninaction.com/
> http://www.voidspace.org.uk/blog
>
>
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081213/72d1e6cd/attachment.html>

From mbrown at dynamicfunds.com  Sun Dec 14 22:31:21 2008
From: mbrown at dynamicfunds.com (Brown, Matt)
Date: Sun, 14 Dec 2008 16:31:21 -0500 (EST)
Subject: [IronPython] subprocess module
Message-ID: <059cf3f17480523c780fb749efcbc8c6.squirrel@www.dynamicfunds.com>

Hello,

I was glad to see that a subprocess module had been written for use with
IronPython; as I'd like to get stdout from plink and os.popen will not
handle this, but subprocess.popen will.
(http://www.ironpython.info/index.php/The_subprocess_module)

However, when I attempt build an IronPython Studio project, it errors upon
build citing an error in Program.py:
- from Form1 import *
- PythonSyntaxErrorException was unhandled by user code (unexpected token if)


This seems to happen whenever there is an issue with an import (note that
I've sys.path.appended the path containing subprocess.py).

Note that I'm completely new to Python, so I assume it's something simple.

Any input is appreciated.


Thanks,

Matt Brown
DYNAMIC Capital Management LLC



From hernan.martinez at ecc.es  Mon Dec 15 13:52:27 2008
From: hernan.martinez at ecc.es (=?iso-8859-1?Q?Hern=E1n_Mart=EDnez_Foffani?=)
Date: Mon, 15 Dec 2008 13:52:27 +0100
Subject: [IronPython] Help contents
Message-ID: <EDEELHKACGEJPBNKEMDBGEKBDLAA.hernan.martinez@ecc.es>

Where does IP find the help content for .NET libraries?
I mean, for instance, the output of
  >>> help(System.DateTime.ToBoolean)

Does it parse the Help.2 files? If so where does it look
for them? Only under "%ALLUSERSPROFILE%\{Program Data}\Microsoft Help"?

Regards,
-H.




From yashgt at gmail.com  Mon Dec 15 14:46:12 2008
From: yashgt at gmail.com (Yash Ganthe)
Date: Mon, 15 Dec 2008 19:16:12 +0530
Subject: [IronPython] Preprocessor directives in IronPython 2.0
Message-ID: <4f428e7b0812150546l2ed24c73td1f492f320f5a95a@mail.gmail.com>

In C# we can code for conditional compilation such as:
public static void Main()
{
#if DEBUG
Console.WriteLine("DEBUG is defined");
#else
Console.WriteLine("DEBUG is not defined");
#endif
}

Is it possible to do the same in IronPython? Does it support preprocessor
directives?

Thanks,
Yash
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081215/76a8c929/attachment.html>

From curt at hagenlocher.org  Mon Dec 15 15:27:05 2008
From: curt at hagenlocher.org (Curt Hagenlocher)
Date: Mon, 15 Dec 2008 06:27:05 -0800
Subject: [IronPython] Preprocessor directives in IronPython 2.0
In-Reply-To: <4f428e7b0812150546l2ed24c73td1f492f320f5a95a@mail.gmail.com>
References: <4f428e7b0812150546l2ed24c73td1f492f320f5a95a@mail.gmail.com>
Message-ID: <d2155e360812150627kabee1b1o7edd85f94327a99f@mail.gmail.com>

Nope.  Why do you think you want them?  Or rather, what's wrong with this:

debug = False

def Main():
    if debug:
        print 'DEBUG is set'
    else:
        print 'DEBUG is not set'

On Mon, Dec 15, 2008 at 5:46 AM, Yash Ganthe <yashgt at gmail.com> wrote:

> In C# we can code for conditional compilation such as:
> public static void Main()
> {
> #if DEBUG
> Console.WriteLine("DEBUG is defined");
> #else
> Console.WriteLine("DEBUG is not defined");
> #endif
> }
>
> Is it possible to do the same in IronPython? Does it support preprocessor
> directives?
>
> Thanks,
> Yash
>
>
>
>
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081215/7dad4c5f/attachment.html>

From tom.wright at resolversystems.com  Mon Dec 15 17:03:38 2008
From: tom.wright at resolversystems.com (Tom Wright)
Date: Mon, 15 Dec 2008 16:03:38 +0000
Subject: [IronPython] IronPython and file descriptors
Message-ID: <49467FDA.5000101@resolversystems.com>

Hi,

At the moment file.fileno() returns an arbitrary identifier of a python 
file rather than a true file descriptor. This is semi-blocking the 
Ironclad port of PIL.

Code in PIL gets an integer using fileno() and passes it directly to C 
code where a call to write() is made. To fix this  one would have to 
patch PIL, IronPython's file objects or the write function provided to C 
code - none of these feel like a good course of action.

When ctypes is ported to IronPython this may create similar problems. 
Ideally fileno() would return a real file descriptor. Would this be 
possible?

Thanks,

Tom
Resolver Systems


From curt at hagenlocher.org  Mon Dec 15 18:28:04 2008
From: curt at hagenlocher.org (Curt Hagenlocher)
Date: Mon, 15 Dec 2008 09:28:04 -0800
Subject: [IronPython] IronPython and file descriptors
In-Reply-To: <49467FDA.5000101@resolversystems.com>
References: <49467FDA.5000101@resolversystems.com>
Message-ID: <d2155e360812150928i6e18806k309ce32ab98fde4e@mail.gmail.com>

There's no such thing as a file descriptor number in .NET -- or, for that
matter, in Windows itself! :)  (The latter is something of a semantic point,
of course, as a HANDLE serves something of the same role in Win32 as a file
descriptor number does in Unix.)
If you have a FileStream, I think you can turn it into a Windows HANDLE by
saying
IntPtr handle = stream.SafeFileHandle.DangerousGetHandle();
The C library should have a way to convert a HANDLE into a "file
descriptor", though I wasn't able to identify the function name with a quick
google.
I don't see a way to get handles for stdin, stdout or stderr, though, except
that these ought to be easy to translate by hand.

 On Mon, Dec 15, 2008 at 8:03 AM, Tom Wright <tom.wright at resolversystems.com
> wrote:

> Hi,
>
> At the moment file.fileno() returns an arbitrary identifier of a python
> file rather than a true file descriptor. This is semi-blocking the Ironclad
> port of PIL.
>
> Code in PIL gets an integer using fileno() and passes it directly to C code
> where a call to write() is made. To fix this  one would have to patch PIL,
> IronPython's file objects or the write function provided to C code - none of
> these feel like a good course of action.
>
> When ctypes is ported to IronPython this may create similar problems.
> Ideally fileno() would return a real file descriptor. Would this be
> possible?
>
> Thanks,
>
> Tom
> Resolver Systems
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081215/1bb887d4/attachment.html>

From tom.wright at resolversystems.com  Mon Dec 15 18:37:04 2008
From: tom.wright at resolversystems.com (Tom Wright)
Date: Mon, 15 Dec 2008 17:37:04 +0000
Subject: [IronPython] IronPython and file descriptors
In-Reply-To: <d2155e360812150928i6e18806k309ce32ab98fde4e@mail.gmail.com>
References: <49467FDA.5000101@resolversystems.com>
	<d2155e360812150928i6e18806k309ce32ab98fde4e@mail.gmail.com>
Message-ID: <494695C0.9040401@resolversystems.com>

Not so: Though admittedly you have to do quite a bit of work to get the 
file descriptor into .NET.
    http://msdn.microsoft.com/en-us/library/bdts1c9x(VS.71).aspx

To clarify - the C library was written to work with Python (not 
IronPython) and is not my code.

Thanks,
Tom

Curt Hagenlocher wrote:
>
> There's no such thing as a file descriptor number in .NET -- or, for 
> that matter, in Windows itself! :)  (The latter is something of a 
> semantic point, of course, as a HANDLE serves something of the same 
> role in Win32 as a file descriptor number does in Unix.)
>
> If you have a FileStream, I think you can turn it into a Windows 
> HANDLE by saying
> IntPtr handle = stream.SafeFileHandle.DangerousGetHandle();
> The C library should have a way to convert a HANDLE into a "file 
> descriptor", though I wasn't able to identify the function name with a 
> quick google.
> I don't see a way to get handles for stdin, stdout or stderr, though, 
> except that these ought to be easy to translate by hand.
>  
> On Mon, Dec 15, 2008 at 8:03 AM, Tom Wright 
> <tom.wright at resolversystems.com 
> <mailto:tom.wright at resolversystems.com>> wrote:
>
>     Hi,
>
>     At the moment file.fileno() returns an arbitrary identifier of a
>     python file rather than a true file descriptor. This is
>     semi-blocking the Ironclad port of PIL.
>
>     Code in PIL gets an integer using fileno() and passes it directly
>     to C code where a call to write() is made. To fix this  one would
>     have to patch PIL, IronPython's file objects or the write function
>     provided to C code - none of these feel like a good course of action.
>
>     When ctypes is ported to IronPython this may create similar
>     problems. Ideally fileno() would return a real file descriptor.
>     Would this be possible?
>
>     Thanks,
>
>     Tom
>     Resolver Systems
>     _______________________________________________
>     Users mailing list
>     Users at lists.ironpython.com <mailto:Users at lists.ironpython.com>
>     http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>
>
> ------------------------------------------------------------------------
>
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>   



From curt at hagenlocher.org  Mon Dec 15 18:47:19 2008
From: curt at hagenlocher.org (Curt Hagenlocher)
Date: Mon, 15 Dec 2008 09:47:19 -0800
Subject: [IronPython] IronPython and file descriptors
In-Reply-To: <494695C0.9040401@resolversystems.com>
References: <49467FDA.5000101@resolversystems.com>
	<d2155e360812150928i6e18806k309ce32ab98fde4e@mail.gmail.com>
	<494695C0.9040401@resolversystems.com>
Message-ID: <d2155e360812150947vccb612ey9d32711bef718395@mail.gmail.com>

Ah, that was the function I was looking for.

Sure, the file descriptor exists in the C library under Windows.  But the C
library is basically doing exactly the same thing as IronPython is; it's
maintaining the file descriptor number as an abstraction on top of the
HANDLE that the operating system knows about.  So we're dealing two similar
abstractions with entirely unrelated interfaces:

IronPython fd built on top of .NET stream built on top of Windows HANDLE, vs
clib fd built directly on top of Windows HANDLE

To get from the IronPython fd to a clib fd, you need to translate down the
first chain and up the second.
On Mon, Dec 15, 2008 at 9:37 AM, Tom Wright
<tom.wright at resolversystems.com>wrote:

> Not so: Though admittedly you have to do quite a bit of work to get the
> file descriptor into .NET.
>   http://msdn.microsoft.com/en-us/library/bdts1c9x(VS.71).aspx
>
> To clarify - the C library was written to work with Python (not IronPython)
> and is not my code.
>
> Thanks,
> Tom
>
> Curt Hagenlocher wrote:
>
>>
>> There's no such thing as a file descriptor number in .NET -- or, for that
>> matter, in Windows itself! :)  (The latter is something of a semantic point,
>> of course, as a HANDLE serves something of the same role in Win32 as a file
>> descriptor number does in Unix.)
>>
>> If you have a FileStream, I think you can turn it into a Windows HANDLE by
>> saying
>> IntPtr handle = stream.SafeFileHandle.DangerousGetHandle();
>> The C library should have a way to convert a HANDLE into a "file
>> descriptor", though I wasn't able to identify the function name with a quick
>> google.
>> I don't see a way to get handles for stdin, stdout or stderr, though,
>> except that these ought to be easy to translate by hand.
>>  On Mon, Dec 15, 2008 at 8:03 AM, Tom Wright <
>> tom.wright at resolversystems.com <mailto:tom.wright at resolversystems.com>>
>> wrote:
>>
>>    Hi,
>>
>>    At the moment file.fileno() returns an arbitrary identifier of a
>>    python file rather than a true file descriptor. This is
>>    semi-blocking the Ironclad port of PIL.
>>
>>    Code in PIL gets an integer using fileno() and passes it directly
>>    to C code where a call to write() is made. To fix this  one would
>>    have to patch PIL, IronPython's file objects or the write function
>>    provided to C code - none of these feel like a good course of action.
>>
>>    When ctypes is ported to IronPython this may create similar
>>    problems. Ideally fileno() would return a real file descriptor.
>>    Would this be possible?
>>
>>    Thanks,
>>
>>    Tom
>>    Resolver Systems
>>    _______________________________________________
>>    Users mailing list
>>    Users at lists.ironpython.com <mailto:Users at lists.ironpython.com>
>>    http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>>
>>
>> ------------------------------------------------------------------------
>>
>> _______________________________________________
>> Users mailing list
>> Users at lists.ironpython.com
>> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>>
>>
>
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081215/0553683e/attachment.html>

From curt at hagenlocher.org  Mon Dec 15 18:56:04 2008
From: curt at hagenlocher.org (Curt Hagenlocher)
Date: Mon, 15 Dec 2008 09:56:04 -0800
Subject: [IronPython] Help contents
In-Reply-To: <EDEELHKACGEJPBNKEMDBGEKBDLAA.hernan.martinez@ecc.es>
References: <EDEELHKACGEJPBNKEMDBGEKBDLAA.hernan.martinez@ecc.es>
Message-ID: <d2155e360812150956hf67ebd1o838a6977a55f6c24@mail.gmail.com>

This is generated from the XML document that lives adjacent to the assembly
containing the code.  For instance, mscorlib is typically located someplace
like C:\Windows\Microsoft.NET\Framework\v2.0.50727\mscorlib.dll, so the XML
file is taken from
C:\Windows\Microsoft.NET\Framework\v2.0.50727\en\mscorlib.xml (assuming
CurrentCulture is English).

The source code for this can be found in DocBuilder.cs

On Mon, Dec 15, 2008 at 4:52 AM, Hern?n Mart?nez Foffani <
hernan.martinez at ecc.es> wrote:

> Where does IP find the help content for .NET libraries?
> I mean, for instance, the output of
>  >>> help(System.DateTime.ToBoolean)
>
> Does it parse the Help.2 files? If so where does it look
> for them? Only under "%ALLUSERSPROFILE%\{Program Data}\Microsoft Help"?
>
> Regards,
> -H.
>
>
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081215/001711d8/attachment.html>

From tom.wright at resolversystems.com  Mon Dec 15 19:18:46 2008
From: tom.wright at resolversystems.com (Tom Wright)
Date: Mon, 15 Dec 2008 18:18:46 +0000
Subject: [IronPython] IronPython and file descriptors
In-Reply-To: <d2155e360812150947vccb612ey9d32711bef718395@mail.gmail.com>
References: <49467FDA.5000101@resolversystems.com>	<d2155e360812150928i6e18806k309ce32ab98fde4e@mail.gmail.com>	<494695C0.9040401@resolversystems.com>
	<d2155e360812150947vccb612ey9d32711bef718395@mail.gmail.com>
Message-ID: <49469F86.8040901@resolversystems.com>

Agreed. It is certainly possible with some work to get a file descriptor 
and pass it to C code.

This is not the problem, however. Ironclad's aim (eventually) is to 
allow *arbitrary* C extensions  to work with Ironpython without changing 
the extensions. Ctypes aim is to allow arbitrary C code to be run by 
python (possibly in other python libraries which you really don't want 
to modify).

In CPython .fileno() is a file descriptor and some modules use this fact 
(specifically PIL) by passing file descriptors as integers into C code. 
I do not know how one can make this code work in IronPython unless 
.fileno() returns a file descriptor.

The game here is not making a particular C library work with IronPython 
by modifying this library but making as many libraries as possible work 
without modification. Of course whether this is worthwhile depends on 
how many libraries rely on this fact.

Sorry - I hope this is a little clearer.

Tom


Curt Hagenlocher wrote:
> Ah, that was the function I was looking for.
>  
> Sure, the file descriptor exists in the C library under Windows.  But 
> the C library is basically doing exactly the same thing as IronPython 
> is; it's maintaining the file descriptor number as an abstraction on 
> top of the HANDLE that the operating system knows about.  So we're 
> dealing two similar abstractions with entirely unrelated interfaces:
>
> IronPython fd built on top of .NET stream built on top of Windows 
> HANDLE, vs
> clib fd built directly on top of Windows HANDLE
>  
> To get from the IronPython fd to a clib fd, you need to translate down 
> the first chain and up the second.
> On Mon, Dec 15, 2008 at 9:37 AM, Tom Wright 
> <tom.wright at resolversystems.com 
> <mailto:tom.wright at resolversystems.com>> wrote:
>
>     Not so: Though admittedly you have to do quite a bit of work to
>     get the file descriptor into .NET.
>       http://msdn.microsoft.com/en-us/library/bdts1c9x(VS.71).aspx
>     <http://msdn.microsoft.com/en-us/library/bdts1c9x%28VS.71%29.aspx>
>
>     To clarify - the C library was written to work with Python (not
>     IronPython) and is not my code.
>
>     Thanks,
>     Tom
>
>     Curt Hagenlocher wrote:
>
>
>         There's no such thing as a file descriptor number in .NET --
>         or, for that matter, in Windows itself! :)  (The latter is
>         something of a semantic point, of course, as a HANDLE serves
>         something of the same role in Win32 as a file descriptor
>         number does in Unix.)
>
>         If you have a FileStream, I think you can turn it into a
>         Windows HANDLE by saying
>         IntPtr handle = stream.SafeFileHandle.DangerousGetHandle();
>         The C library should have a way to convert a HANDLE into a
>         "file descriptor", though I wasn't able to identify the
>         function name with a quick google.
>         I don't see a way to get handles for stdin, stdout or stderr,
>         though, except that these ought to be easy to translate by hand.
>          On Mon, Dec 15, 2008 at 8:03 AM, Tom Wright
>         <tom.wright at resolversystems.com
>         <mailto:tom.wright at resolversystems.com>
>         <mailto:tom.wright at resolversystems.com
>         <mailto:tom.wright at resolversystems.com>>> wrote:
>
>            Hi,
>
>            At the moment file.fileno() returns an arbitrary identifier
>         of a
>            python file rather than a true file descriptor. This is
>            semi-blocking the Ironclad port of PIL.
>
>            Code in PIL gets an integer using fileno() and passes it
>         directly
>            to C code where a call to write() is made. To fix this  one
>         would
>            have to patch PIL, IronPython's file objects or the write
>         function
>            provided to C code - none of these feel like a good course
>         of action.
>
>            When ctypes is ported to IronPython this may create similar
>            problems. Ideally fileno() would return a real file descriptor.
>            Would this be possible?
>
>            Thanks,
>
>            Tom
>            Resolver Systems
>            _______________________________________________
>            Users mailing list
>            Users at lists.ironpython.com
>         <mailto:Users at lists.ironpython.com>
>         <mailto:Users at lists.ironpython.com
>         <mailto:Users at lists.ironpython.com>>
>
>            http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>
>
>         ------------------------------------------------------------------------
>
>
>
>         _______________________________________________
>         Users mailing list
>         Users at lists.ironpython.com <mailto:Users at lists.ironpython.com>
>         http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>          
>
>
>     _______________________________________________
>     Users mailing list
>     Users at lists.ironpython.com <mailto:Users at lists.ironpython.com>
>     http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>
>
> ------------------------------------------------------------------------
>
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>   



From slide.o.mix at gmail.com  Mon Dec 15 19:21:27 2008
From: slide.o.mix at gmail.com (Slide)
Date: Mon, 15 Dec 2008 11:21:27 -0700
Subject: [IronPython] IronPython and file descriptors
In-Reply-To: <49469F86.8040901@resolversystems.com>
References: <49467FDA.5000101@resolversystems.com>
	<d2155e360812150928i6e18806k309ce32ab98fde4e@mail.gmail.com>
	<494695C0.9040401@resolversystems.com>
	<d2155e360812150947vccb612ey9d32711bef718395@mail.gmail.com>
	<49469F86.8040901@resolversystems.com>
Message-ID: <a623f7d10812151021u7e6dbba3k3b0971f4d42295f3@mail.gmail.com>

On Mon, Dec 15, 2008 at 11:18 AM, Tom Wright
<tom.wright at resolversystems.com> wrote:
> Agreed. It is certainly possible with some work to get a file descriptor and
> pass it to C code.
>
> This is not the problem, however. Ironclad's aim (eventually) is to allow
> *arbitrary* C extensions  to work with Ironpython without changing the
> extensions. Ctypes aim is to allow arbitrary C code to be run by python
> (possibly in other python libraries which you really don't want to modify).
>
> In CPython .fileno() is a file descriptor and some modules use this fact
> (specifically PIL) by passing file descriptors as integers into C code. I do
> not know how one can make this code work in IronPython unless .fileno()
> returns a file descriptor.
>
> The game here is not making a particular C library work with IronPython by
> modifying this library but making as many libraries as possible work without
> modification. Of course whether this is worthwhile depends on how many
> libraries rely on this fact.
>
> Sorry - I hope this is a little clearer.
>
> Tom


How does CPython get the file descriptor for fileno()? Does it just
return the HANDLE on Windows?

slide


From dinov at microsoft.com  Mon Dec 15 19:20:47 2008
From: dinov at microsoft.com (Dino Viehland)
Date: Mon, 15 Dec 2008 10:20:47 -0800
Subject: [IronPython] Preprocessor directives in IronPython 2.0
In-Reply-To: <d2155e360812150627kabee1b1o7edd85f94327a99f@mail.gmail.com>
References: <4f428e7b0812150546l2ed24c73td1f492f320f5a95a@mail.gmail.com>
	<d2155e360812150627kabee1b1o7edd85f94327a99f@mail.gmail.com>
Message-ID: <350E7D38B6D819428718949920EC2355564B62EA68@NA-EXMSG-C102.redmond.corp.microsoft.com>

The one form this is available is with the assert keyword.  Assert statements will be removed in optimized code so you don't have the if debug check.  But they won't let you have the else branch.

From: users-bounces at lists.ironpython.com [mailto:users-bounces at lists.ironpython.com] On Behalf Of Curt Hagenlocher
Sent: Monday, December 15, 2008 6:27 AM
To: Discussion of IronPython
Subject: Re: [IronPython] Preprocessor directives in IronPython 2.0

Nope.  Why do you think you want them?  Or rather, what's wrong with this:

debug = False

def Main():
    if debug:
        print 'DEBUG is set'
    else:
        print 'DEBUG is not set'
On Mon, Dec 15, 2008 at 5:46 AM, Yash Ganthe <yashgt at gmail.com<mailto:yashgt at gmail.com>> wrote:
In C# we can code for conditional compilation such as:
public static void Main()
{
#if DEBUG
Console.WriteLine("DEBUG is defined");
#else
Console.WriteLine("DEBUG is not defined");
#endif
}

Is it possible to do the same in IronPython? Does it support preprocessor directives?

Thanks,
Yash




_______________________________________________
Users mailing list
Users at lists.ironpython.com<mailto:Users at lists.ironpython.com>
http://lists.ironpython.com/listinfo.cgi/users-ironpython.com

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081215/1ce29f78/attachment.html>

From curt at hagenlocher.org  Mon Dec 15 19:27:35 2008
From: curt at hagenlocher.org (Curt Hagenlocher)
Date: Mon, 15 Dec 2008 10:27:35 -0800
Subject: [IronPython] IronPython and file descriptors
In-Reply-To: <49469F86.8040901@resolversystems.com>
References: <49467FDA.5000101@resolversystems.com>
	<d2155e360812150928i6e18806k309ce32ab98fde4e@mail.gmail.com>
	<494695C0.9040401@resolversystems.com>
	<d2155e360812150947vccb612ey9d32711bef718395@mail.gmail.com>
	<49469F86.8040901@resolversystems.com>
Message-ID: <d2155e360812151027p3dc17bdan4072e64ba7d14c68@mail.gmail.com>

I think I understand the motivation pretty well; what I'm saying is that I
don't see that this is possible.  IronPython has no access to the table
being used by the C library to map between Windows HANDLEs and clib file
descriptors, so we can't return a clib-compatible fd.

Even if we were to return a Windows HANDLE as the fd from IronPython
(something which is bad in several ways), this wouldn't be a value that
could just be used directly by the C extension through the C library.
On Mon, Dec 15, 2008 at 10:18 AM, Tom Wright <tom.wright at resolversystems.com
> wrote:

> Agreed. It is certainly possible with some work to get a file descriptor
> and pass it to C code.
>
> This is not the problem, however. Ironclad's aim (eventually) is to allow
> *arbitrary* C extensions  to work with Ironpython without changing the
> extensions. Ctypes aim is to allow arbitrary C code to be run by python
> (possibly in other python libraries which you really don't want to modify).
>
> In CPython .fileno() is a file descriptor and some modules use this fact
> (specifically PIL) by passing file descriptors as integers into C code. I do
> not know how one can make this code work in IronPython unless .fileno()
> returns a file descriptor.
>
> The game here is not making a particular C library work with IronPython by
> modifying this library but making as many libraries as possible work without
> modification. Of course whether this is worthwhile depends on how many
> libraries rely on this fact.
>
> Sorry - I hope this is a little clearer.
>
> Tom
>
>
> Curt Hagenlocher wrote:
>
>> Ah, that was the function I was looking for.
>>  Sure, the file descriptor exists in the C library under Windows.  But the
>> C library is basically doing exactly the same thing as IronPython is; it's
>> maintaining the file descriptor number as an abstraction on top of the
>> HANDLE that the operating system knows about.  So we're dealing two similar
>> abstractions with entirely unrelated interfaces:
>>
>> IronPython fd built on top of .NET stream built on top of Windows HANDLE,
>> vs
>> clib fd built directly on top of Windows HANDLE
>>  To get from the IronPython fd to a clib fd, you need to translate down
>> the first chain and up the second.
>> On Mon, Dec 15, 2008 at 9:37 AM, Tom Wright <
>> tom.wright at resolversystems.com <mailto:tom.wright at resolversystems.com>>
>> wrote:
>>
>>    Not so: Though admittedly you have to do quite a bit of work to
>>    get the file descriptor into .NET.
>>      http://msdn.microsoft.com/en-us/library/bdts1c9x(VS.71).aspx
>>    <http://msdn.microsoft.com/en-us/library/bdts1c9x%28VS.71%29.aspx>
>>
>>
>>    To clarify - the C library was written to work with Python (not
>>    IronPython) and is not my code.
>>
>>    Thanks,
>>    Tom
>>
>>    Curt Hagenlocher wrote:
>>
>>
>>        There's no such thing as a file descriptor number in .NET --
>>        or, for that matter, in Windows itself! :)  (The latter is
>>        something of a semantic point, of course, as a HANDLE serves
>>        something of the same role in Win32 as a file descriptor
>>        number does in Unix.)
>>
>>        If you have a FileStream, I think you can turn it into a
>>        Windows HANDLE by saying
>>        IntPtr handle = stream.SafeFileHandle.DangerousGetHandle();
>>        The C library should have a way to convert a HANDLE into a
>>        "file descriptor", though I wasn't able to identify the
>>        function name with a quick google.
>>        I don't see a way to get handles for stdin, stdout or stderr,
>>        though, except that these ought to be easy to translate by hand.
>>         On Mon, Dec 15, 2008 at 8:03 AM, Tom Wright
>>        <tom.wright at resolversystems.com
>>        <mailto:tom.wright at resolversystems.com>
>>        <mailto:tom.wright at resolversystems.com
>>        <mailto:tom.wright at resolversystems.com>>> wrote:
>>
>>           Hi,
>>
>>           At the moment file.fileno() returns an arbitrary identifier
>>        of a
>>           python file rather than a true file descriptor. This is
>>           semi-blocking the Ironclad port of PIL.
>>
>>           Code in PIL gets an integer using fileno() and passes it
>>        directly
>>           to C code where a call to write() is made. To fix this  one
>>        would
>>           have to patch PIL, IronPython's file objects or the write
>>        function
>>           provided to C code - none of these feel like a good course
>>        of action.
>>
>>           When ctypes is ported to IronPython this may create similar
>>           problems. Ideally fileno() would return a real file descriptor.
>>           Would this be possible?
>>
>>           Thanks,
>>
>>           Tom
>>           Resolver Systems
>>           _______________________________________________
>>           Users mailing list
>>           Users at lists.ironpython.com
>>        <mailto:Users at lists.ironpython.com>
>>        <mailto:Users at lists.ironpython.com
>>        <mailto:Users at lists.ironpython.com>>
>>
>>           http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>>
>>
>>
>>  ------------------------------------------------------------------------
>>
>>
>>
>>        _______________________________________________
>>        Users mailing list
>>        Users at lists.ironpython.com <mailto:Users at lists.ironpython.com>
>>        http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>>
>>
>>    _______________________________________________
>>    Users mailing list
>>    Users at lists.ironpython.com <mailto:Users at lists.ironpython.com>
>>    http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>>
>>
>> ------------------------------------------------------------------------
>>
>> _______________________________________________
>> Users mailing list
>> Users at lists.ironpython.com
>> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>>
>>
>
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081215/059b3517/attachment.html>

From hernan.martinez at ecc.es  Mon Dec 15 19:45:38 2008
From: hernan.martinez at ecc.es (=?iso-8859-1?Q?Hern=E1n_Mart=EDnez_Foffani?=)
Date: Mon, 15 Dec 2008 19:45:38 +0100
Subject: [IronPython] Help contents
In-Reply-To: <d2155e360812150956hf67ebd1o838a6977a55f6c24@mail.gmail.com>
Message-ID: <EDEELHKACGEJPBNKEMDBEEKDDLAA.hernan.martinez@ecc.es>

It seems clear now, thank you.

Reading DocBuilder.cs:GetXPathDocument(..) I assume that if I want
to let our users to be able to get IP console help on our own C#
libraries, we can place the XML docs in a culture subdirectory under
the DLLs install directory, right?


> This is generated from the XML document that lives adjacent to the
> assembly containing the code.  For instance, mscorlib is typically
> located someplace like
> C:\Windows\Microsoft.NET\Framework\v2.0.50727\mscorlib.dll, so the
> XML file is taken from
> C:\Windows\Microsoft.NET\Framework\v2.0.50727\en\mscorlib.xml
> (assuming CurrentCulture is English).
>
> The source code for this can be found in DocBuilder.cs
>
>
> On Mon, Dec 15, 2008 at 4:52 AM, Hern?n Mart?nez Foffani
> <hernan.martinez at ecc.es> wrote:
>
> Where does IP find the help content for .NET libraries?
> I mean, for instance, the output of
>  >>> help(System.DateTime.ToBoolean)
>
> Does it parse the Help.2 files? If so where does it look
> for them? Only under "%ALLUSERSPROFILE%\{Program Data}\Microsoft
> Help"?
>



From dinov at microsoft.com  Mon Dec 15 19:49:46 2008
From: dinov at microsoft.com (Dino Viehland)
Date: Mon, 15 Dec 2008 10:49:46 -0800
Subject: [IronPython] IronPython and file descriptors
In-Reply-To: <a623f7d10812151021u7e6dbba3k3b0971f4d42295f3@mail.gmail.com>
References: <49467FDA.5000101@resolversystems.com>
	<d2155e360812150928i6e18806k309ce32ab98fde4e@mail.gmail.com>
	<494695C0.9040401@resolversystems.com>
	<d2155e360812150947vccb612ey9d32711bef718395@mail.gmail.com>
	<49469F86.8040901@resolversystems.com>
	<a623f7d10812151021u7e6dbba3k3b0971f4d42295f3@mail.gmail.com>
Message-ID: <350E7D38B6D819428718949920EC2355564B62EAC4@NA-EXMSG-C102.redmond.corp.microsoft.com>

Presumably CPython is linking against msvcrt and is using the C abstraction rather than calling the Win32 APIs which return handles directly.

As Curt said the biggest problem for us is that .NET does not expose C file descriptors.  Therefore we could fix this by P/Invoking out to msvcrt.  For users on other platforms we'd need to either add support for their platform or hope that their handles == C runtime file descriptors.  For something like fileno maybe this is ok because it's an interop point only - or are there any non-interop uses of fileno in the world?

What do people think of this?  This would be the 1st place where we would add a P/Invoke so I'd want to tread lightly and make sure this is really the right thing to do.

-----Original Message-----
From: users-bounces at lists.ironpython.com [mailto:users-bounces at lists.ironpython.com] On Behalf Of Slide
Sent: Monday, December 15, 2008 10:21 AM
To: Discussion of IronPython
Subject: Re: [IronPython] IronPython and file descriptors

On Mon, Dec 15, 2008 at 11:18 AM, Tom Wright
<tom.wright at resolversystems.com> wrote:
> Agreed. It is certainly possible with some work to get a file descriptor and
> pass it to C code.
>
> This is not the problem, however. Ironclad's aim (eventually) is to allow
> *arbitrary* C extensions  to work with Ironpython without changing the
> extensions. Ctypes aim is to allow arbitrary C code to be run by python
> (possibly in other python libraries which you really don't want to modify).
>
> In CPython .fileno() is a file descriptor and some modules use this fact
> (specifically PIL) by passing file descriptors as integers into C code. I do
> not know how one can make this code work in IronPython unless .fileno()
> returns a file descriptor.
>
> The game here is not making a particular C library work with IronPython by
> modifying this library but making as many libraries as possible work without
> modification. Of course whether this is worthwhile depends on how many
> libraries rely on this fact.
>
> Sorry - I hope this is a little clearer.
>
> Tom


How does CPython get the file descriptor for fileno()? Does it just
return the HANDLE on Windows?

slide
_______________________________________________
Users mailing list
Users at lists.ironpython.com
http://lists.ironpython.com/listinfo.cgi/users-ironpython.com


From curt at hagenlocher.org  Mon Dec 15 19:50:19 2008
From: curt at hagenlocher.org (Curt Hagenlocher)
Date: Mon, 15 Dec 2008 10:50:19 -0800
Subject: [IronPython] Help contents
In-Reply-To: <EDEELHKACGEJPBNKEMDBEEKDDLAA.hernan.martinez@ecc.es>
References: <d2155e360812150956hf67ebd1o838a6977a55f6c24@mail.gmail.com>
	<EDEELHKACGEJPBNKEMDBEEKDDLAA.hernan.martinez@ecc.es>
Message-ID: <d2155e360812151050i6185f2fdk807189e5ed535d90@mail.gmail.com>

If you have only one version/culture, I believe you can just place the XML
doc in the same directory as the DLL.

On Mon, Dec 15, 2008 at 10:45 AM, Hern?n Mart?nez Foffani <
hernan.martinez at ecc.es> wrote:

> It seems clear now, thank you.
>
> Reading DocBuilder.cs:GetXPathDocument(..) I assume that if I want
> to let our users to be able to get IP console help on our own C#
> libraries, we can place the XML docs in a culture subdirectory under
> the DLLs install directory, right?
>
>
> > This is generated from the XML document that lives adjacent to the
> > assembly containing the code.  For instance, mscorlib is typically
> > located someplace like
> > C:\Windows\Microsoft.NET\Framework\v2.0.50727\mscorlib.dll, so the
> > XML file is taken from
> > C:\Windows\Microsoft.NET\Framework\v2.0.50727\en\mscorlib.xml
> > (assuming CurrentCulture is English).
> >
> > The source code for this can be found in DocBuilder.cs
> >
> >
> > On Mon, Dec 15, 2008 at 4:52 AM, Hern?n Mart?nez Foffani
> > <hernan.martinez at ecc.es> wrote:
> >
> > Where does IP find the help content for .NET libraries?
> > I mean, for instance, the output of
> >  >>> help(System.DateTime.ToBoolean)
> >
> > Does it parse the Help.2 files? If so where does it look
> > for them? Only under "%ALLUSERSPROFILE%\{Program Data}\Microsoft
> > Help"?
> >
>
>  _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081215/3d9f4597/attachment.html>

From curt at hagenlocher.org  Mon Dec 15 20:32:03 2008
From: curt at hagenlocher.org (Curt Hagenlocher)
Date: Mon, 15 Dec 2008 11:32:03 -0800
Subject: [IronPython] IronPython and file descriptors
In-Reply-To: <350E7D38B6D819428718949920EC2355564B62EAC4@NA-EXMSG-C102.redmond.corp.microsoft.com>
References: <49467FDA.5000101@resolversystems.com>
	<d2155e360812150928i6e18806k309ce32ab98fde4e@mail.gmail.com>
	<494695C0.9040401@resolversystems.com>
	<d2155e360812150947vccb612ey9d32711bef718395@mail.gmail.com>
	<49469F86.8040901@resolversystems.com>
	<a623f7d10812151021u7e6dbba3k3b0971f4d42295f3@mail.gmail.com>
	<350E7D38B6D819428718949920EC2355564B62EAC4@NA-EXMSG-C102.redmond.corp.microsoft.com>
Message-ID: <d2155e360812151132g24f4cdbdr40a9aa0bf4cc7165@mail.gmail.com>

I guess I'm too much of a purist when it comes to managed code.  But it
would make me sad to P/Invoke by default for this purpose, even if "Mono
compatibility" is the only real counterargument I can muster.

On Mon, Dec 15, 2008 at 10:49 AM, Dino Viehland <dinov at microsoft.com> wrote:

> Presumably CPython is linking against msvcrt and is using the C abstraction
> rather than calling the Win32 APIs which return handles directly.
>
> As Curt said the biggest problem for us is that .NET does not expose C file
> descriptors.  Therefore we could fix this by P/Invoking out to msvcrt.  For
> users on other platforms we'd need to either add support for their platform
> or hope that their handles == C runtime file descriptors.  For something
> like fileno maybe this is ok because it's an interop point only - or are
> there any non-interop uses of fileno in the world?
>
> What do people think of this?  This would be the 1st place where we would
> add a P/Invoke so I'd want to tread lightly and make sure this is really the
> right thing to do.
>
> -----Original Message-----
> From: users-bounces at lists.ironpython.com [mailto:
> users-bounces at lists.ironpython.com] On Behalf Of Slide
> Sent: Monday, December 15, 2008 10:21 AM
> To: Discussion of IronPython
> Subject: Re: [IronPython] IronPython and file descriptors
>
> On Mon, Dec 15, 2008 at 11:18 AM, Tom Wright
> <tom.wright at resolversystems.com> wrote:
> > Agreed. It is certainly possible with some work to get a file descriptor
> and
> > pass it to C code.
> >
> > This is not the problem, however. Ironclad's aim (eventually) is to allow
> > *arbitrary* C extensions  to work with Ironpython without changing the
> > extensions. Ctypes aim is to allow arbitrary C code to be run by python
> > (possibly in other python libraries which you really don't want to
> modify).
> >
> > In CPython .fileno() is a file descriptor and some modules use this fact
> > (specifically PIL) by passing file descriptors as integers into C code. I
> do
> > not know how one can make this code work in IronPython unless .fileno()
> > returns a file descriptor.
> >
> > The game here is not making a particular C library work with IronPython
> by
> > modifying this library but making as many libraries as possible work
> without
> > modification. Of course whether this is worthwhile depends on how many
> > libraries rely on this fact.
> >
> > Sorry - I hope this is a little clearer.
> >
> > Tom
>
>
> How does CPython get the file descriptor for fileno()? Does it just
> return the HANDLE on Windows?
>
> slide
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081215/8c1364cb/attachment.html>

From jslutter at reactorzero.com  Mon Dec 15 21:14:51 2008
From: jslutter at reactorzero.com (Jeff Slutter)
Date: Mon, 15 Dec 2008 15:14:51 -0500
Subject: [IronPython] Determine the classes/interfaces a Python implements
Message-ID: <4946BABB.2080905@reactorzero.com>

I have a Python script that creates a class within it. This Python class
is derived off of a class, or interface, I made in C# - something like:

class MyClass(Test.MainForm.IScript):
    ...

Now, back in C#, I have gotten access to "MyClass" by:

object myclass = someScope.GetVariable("MyClass");

Is there a way to determine either:
a) what classes/interfaces MyClass implements OR
b) if it implements a specific class/interface

I want to know if the "object myclass" is supposed to implement
Test.MainForm.IScript or not.

I don't want to create an instance of MyClass as this would cause
problems, executing things I'm not ready to execute.

Also, related but not as important, implementing an interface (as above)
doesn't cause any compiler errors if I'm missing functions - is there a
way to enforce this?


Thanks,
Jeff


From fuzzyman at gmail.com  Mon Dec 15 21:52:43 2008
From: fuzzyman at gmail.com (Michael Foord)
Date: Mon, 15 Dec 2008 20:52:43 +0000
Subject: [IronPython] IronPython and file descriptors
In-Reply-To: <350E7D38B6D819428718949920EC2355564B62EAC4@NA-EXMSG-C102.redmond.corp.microsoft.com>
References: <49467FDA.5000101@resolversystems.com>
	<d2155e360812150928i6e18806k309ce32ab98fde4e@mail.gmail.com>
	<494695C0.9040401@resolversystems.com>
	<d2155e360812150947vccb612ey9d32711bef718395@mail.gmail.com>
	<49469F86.8040901@resolversystems.com>
	<a623f7d10812151021u7e6dbba3k3b0971f4d42295f3@mail.gmail.com>
	<350E7D38B6D819428718949920EC2355564B62EAC4@NA-EXMSG-C102.redmond.corp.microsoft.com>
Message-ID: <6f4025010812151252p2bb355d7p1b9d915c0b033016@mail.gmail.com>

2008/12/15 Dino Viehland <dinov at microsoft.com>

> Presumably CPython is linking against msvcrt and is using the C abstraction
> rather than calling the Win32 APIs which return handles directly.
>
> As Curt said the biggest problem for us is that .NET does not expose C file
> descriptors.  Therefore we could fix this by P/Invoking out to msvcrt.  For
> users on other platforms we'd need to either add support for their platform
> or hope that their handles == C runtime file descriptors.  For something
> like fileno maybe this is ok because it's an interop point only - or are
> there any non-interop uses of fileno in the world?
>
> What do people think of this?  This would be the 1st place where we would
> add a P/Invoke so I'd want to tread lightly and make sure this is really the
> right thing to do.



Well, some way of doing this is needed if certain Python C extensions are to
work with Ironclad. Perhaps a global or per engine setting that allows file
handles to be associated with a C descriptor. Users of Ironclad could switch
it on if they wished and take the consequences.

Michael Foord


>
>
> -----Original Message-----
> From: users-bounces at lists.ironpython.com [mailto:
> users-bounces at lists.ironpython.com] On Behalf Of Slide
> Sent: Monday, December 15, 2008 10:21 AM
> To: Discussion of IronPython
> Subject: Re: [IronPython] IronPython and file descriptors
>
> On Mon, Dec 15, 2008 at 11:18 AM, Tom Wright
> <tom.wright at resolversystems.com> wrote:
> > Agreed. It is certainly possible with some work to get a file descriptor
> and
> > pass it to C code.
> >
> > This is not the problem, however. Ironclad's aim (eventually) is to allow
> > *arbitrary* C extensions  to work with Ironpython without changing the
> > extensions. Ctypes aim is to allow arbitrary C code to be run by python
> > (possibly in other python libraries which you really don't want to
> modify).
> >
> > In CPython .fileno() is a file descriptor and some modules use this fact
> > (specifically PIL) by passing file descriptors as integers into C code. I
> do
> > not know how one can make this code work in IronPython unless .fileno()
> > returns a file descriptor.
> >
> > The game here is not making a particular C library work with IronPython
> by
> > modifying this library but making as many libraries as possible work
> without
> > modification. Of course whether this is worthwhile depends on how many
> > libraries rely on this fact.
> >
> > Sorry - I hope this is a little clearer.
> >
> > Tom
>
>
> How does CPython get the file descriptor for fileno()? Does it just
> return the HANDLE on Windows?
>
> slide
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>



-- 
http://www.ironpythoninaction.com/
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081215/81cda530/attachment.html>

From fuzzyman at voidspace.org.uk  Mon Dec 15 22:20:18 2008
From: fuzzyman at voidspace.org.uk (Michael Foord)
Date: Mon, 15 Dec 2008 21:20:18 +0000
Subject: [IronPython] Determine the classes/interfaces a Python
	implements
In-Reply-To: <4946BABB.2080905@reactorzero.com>
References: <4946BABB.2080905@reactorzero.com>
Message-ID: <4946CA12.4050600@voidspace.org.uk>


Hi Jeff,

Probably the easiest way of doing this is to define a Python function 
that uses issubtype. You can use this as a delegate from the C# side 
(warning untested):

ScriptScope scope = engine.CreateScope();
ScriptSource imports = engine.CreateScriptSourceFromString("from System 
import ISomething", SourceCodeKind.Staatements);
imports.Execute(scope);
ScriptSource source = engine.CreateScriptSourceFromString("lambda x: 
issubtype(x, ISomething)", SourceCodeKind.Expression);
Func<object, bool> Implements = (scope)source.Execute(scope);

bool result = Implements(some_object);


Note that issubtype will barf if you give it anything other than a type, 
so it may make more sense to define a function with exception handling 
and pull it out of the scope instead.

HTH

Michael Foord


Jeff Slutter wrote:
> I have a Python script that creates a class within it. This Python class
> is derived off of a class, or interface, I made in C# - something like:
>
> class MyClass(Test.MainForm.IScript):
>     ...
>
> Now, back in C#, I have gotten access to "MyClass" by:
>
> object myclass = someScope.GetVariable("MyClass");
>
> Is there a way to determine either:
> a) what classes/interfaces MyClass implements OR
> b) if it implements a specific class/interface
>
> I want to know if the "object myclass" is supposed to implement
> Test.MainForm.IScript or not.
>
> I don't want to create an instance of MyClass as this would cause
> problems, executing things I'm not ready to execute.
>
> Also, related but not as important, implementing an interface (as above)
> doesn't cause any compiler errors if I'm missing functions - is there a
> way to enforce this?
>
>
> Thanks,
> Jeff
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>   


-- 
http://www.ironpythoninaction.com/
http://www.voidspace.org.uk/blog




From sanxiyn at gmail.com  Mon Dec 15 22:30:39 2008
From: sanxiyn at gmail.com (Seo Sanghyeon)
Date: Tue, 16 Dec 2008 06:30:39 +0900
Subject: [IronPython] Determine the classes/interfaces a Python
	implements
In-Reply-To: <4946CA12.4050600@voidspace.org.uk>
References: <4946BABB.2080905@reactorzero.com>
	<4946CA12.4050600@voidspace.org.uk>
Message-ID: <5b0248170812151330u7a6754d4paa688086c0735944@mail.gmail.com>

2008/12/16 Michael Foord <fuzzyman at voidspace.org.uk>:
> Probably the easiest way of doing this is to define a Python function that
> uses issubtype. You can use this as a delegate from the C# side (warning
> untested):

You mean issubclass...

-- 
Seo Sanghyeon


From dinov at microsoft.com  Mon Dec 15 22:29:15 2008
From: dinov at microsoft.com (Dino Viehland)
Date: Mon, 15 Dec 2008 13:29:15 -0800
Subject: [IronPython] Determine the classes/interfaces a Python
 implements
In-Reply-To: <4946BABB.2080905@reactorzero.com>
References: <4946BABB.2080905@reactorzero.com>
Message-ID: <350E7D38B6D819428718949920EC2355564B62EC27@NA-EXMSG-C102.redmond.corp.microsoft.com>

You should call PythonOps.IsSubClass and pass in the PythonType and the interface you want to compare it with.  The interface can be either a .NET type object or another Python type object (or a tuple, etc...)

The not raising on missing functions is a feature :)  You could build a meta-class which would look at the incoming arguments and validate that all of the methods in an interface are implemented.  That shouldn't be too hard - probably just seomthing like:

assert (set(dir(someInterface)) & set(newTypeDict)) == set(dir(someInterface))

-----Original Message-----
From: users-bounces at lists.ironpython.com [mailto:users-bounces at lists.ironpython.com] On Behalf Of Jeff Slutter
Sent: Monday, December 15, 2008 12:15 PM
To: Discussion of IronPython
Subject: [IronPython] Determine the classes/interfaces a Python implements

I have a Python script that creates a class within it. This Python class
is derived off of a class, or interface, I made in C# - something like:

class MyClass(Test.MainForm.IScript):
    ...

Now, back in C#, I have gotten access to "MyClass" by:

object myclass = someScope.GetVariable("MyClass");

Is there a way to determine either:
a) what classes/interfaces MyClass implements OR
b) if it implements a specific class/interface

I want to know if the "object myclass" is supposed to implement
Test.MainForm.IScript or not.

I don't want to create an instance of MyClass as this would cause
problems, executing things I'm not ready to execute.

Also, related but not as important, implementing an interface (as above)
doesn't cause any compiler errors if I'm missing functions - is there a
way to enforce this?


Thanks,
Jeff
_______________________________________________
Users mailing list
Users at lists.ironpython.com
http://lists.ironpython.com/listinfo.cgi/users-ironpython.com



From dinov at microsoft.com  Mon Dec 15 22:24:03 2008
From: dinov at microsoft.com (Dino Viehland)
Date: Mon, 15 Dec 2008 13:24:03 -0800
Subject: [IronPython] Determine the classes/interfaces a
	Python	implements
In-Reply-To: <4946CA12.4050600@voidspace.org.uk>
References: <4946BABB.2080905@reactorzero.com>
	<4946CA12.4050600@voidspace.org.uk>
Message-ID: <350E7D38B6D819428718949920EC2355564B62EC17@NA-EXMSG-C102.redmond.corp.microsoft.com>

Can you open a feature request on CodePlex?  It's certainly an interesting idea to ponder and I'm leaning towards it but there's lots of details to be gotten right.

Do you know if this needs to work w/ sockets as well?  (There's also the question of can we make it work with sockets?  :))

There'll be a bunch of places we need to update (nt, socket, file, select, etc...) so I think it'll have to wait until 2.1 instead of coming in a minor update like 2.0.1.

-----Original Message-----
From: users-bounces at lists.ironpython.com [mailto:users-bounces at lists.ironpython.com] On Behalf Of Michael Foord
Sent: Monday, December 15, 2008 1:20 PM
To: Discussion of IronPython
Subject: Re: [IronPython] Determine the classes/interfaces a Python implements


Hi Jeff,

Probably the easiest way of doing this is to define a Python function
that uses issubtype. You can use this as a delegate from the C# side
(warning untested):

ScriptScope scope = engine.CreateScope();
ScriptSource imports = engine.CreateScriptSourceFromString("from System
import ISomething", SourceCodeKind.Staatements);
imports.Execute(scope);
ScriptSource source = engine.CreateScriptSourceFromString("lambda x:
issubtype(x, ISomething)", SourceCodeKind.Expression);
Func<object, bool> Implements = (scope)source.Execute(scope);

bool result = Implements(some_object);


Note that issubtype will barf if you give it anything other than a type,
so it may make more sense to define a function with exception handling
and pull it out of the scope instead.

HTH

Michael Foord


Jeff Slutter wrote:
> I have a Python script that creates a class within it. This Python class
> is derived off of a class, or interface, I made in C# - something like:
>
> class MyClass(Test.MainForm.IScript):
>     ...
>
> Now, back in C#, I have gotten access to "MyClass" by:
>
> object myclass = someScope.GetVariable("MyClass");
>
> Is there a way to determine either:
> a) what classes/interfaces MyClass implements OR
> b) if it implements a specific class/interface
>
> I want to know if the "object myclass" is supposed to implement
> Test.MainForm.IScript or not.
>
> I don't want to create an instance of MyClass as this would cause
> problems, executing things I'm not ready to execute.
>
> Also, related but not as important, implementing an interface (as above)
> doesn't cause any compiler errors if I'm missing functions - is there a
> way to enforce this?
>
>
> Thanks,
> Jeff
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>


--
http://www.ironpythoninaction.com/
http://www.voidspace.org.uk/blog


_______________________________________________
Users mailing list
Users at lists.ironpython.com
http://lists.ironpython.com/listinfo.cgi/users-ironpython.com



From curt at hagenlocher.org  Mon Dec 15 22:40:17 2008
From: curt at hagenlocher.org (Curt Hagenlocher)
Date: Mon, 15 Dec 2008 13:40:17 -0800
Subject: [IronPython] Determine the classes/interfaces a Python
	implements
In-Reply-To: <350E7D38B6D819428718949920EC2355564B62EC17@NA-EXMSG-C102.redmond.corp.microsoft.com>
References: <4946BABB.2080905@reactorzero.com>
	<4946CA12.4050600@voidspace.org.uk>
	<350E7D38B6D819428718949920EC2355564B62EC17@NA-EXMSG-C102.redmond.corp.microsoft.com>
Message-ID: <d2155e360812151340g5ebb4208g9cfa563183c1f6ae@mail.gmail.com>

(Dino clearly meant this to be a reply to the other thread with respect to C
file descriptors. :)

On Mon, Dec 15, 2008 at 1:24 PM, Dino Viehland <dinov at microsoft.com> wrote:

> Can you open a feature request on CodePlex?  It's certainly an interesting
> idea to ponder and I'm leaning towards it but there's lots of details to be
> gotten right.
>
> Do you know if this needs to work w/ sockets as well?  (There's also the
> question of can we make it work with sockets?  :))
>
> There'll be a bunch of places we need to update (nt, socket, file, select,
> etc...) so I think it'll have to wait until 2.1 instead of coming in a minor
> update like 2.0.1.
>
> -----Original Message-----
> From: users-bounces at lists.ironpython.com [mailto:
> users-bounces at lists.ironpython.com] On Behalf Of Michael Foord
> Sent: Monday, December 15, 2008 1:20 PM
> To: Discussion of IronPython
>  Subject: Re: [IronPython] Determine the classes/interfaces a Python
> implements
>
>
> Hi Jeff,
>
> Probably the easiest way of doing this is to define a Python function
> that uses issubtype. You can use this as a delegate from the C# side
> (warning untested):
>
> ScriptScope scope = engine.CreateScope();
> ScriptSource imports = engine.CreateScriptSourceFromString("from System
> import ISomething", SourceCodeKind.Staatements);
> imports.Execute(scope);
> ScriptSource source = engine.CreateScriptSourceFromString("lambda x:
> issubtype(x, ISomething)", SourceCodeKind.Expression);
> Func<object, bool> Implements = (scope)source.Execute(scope);
>
> bool result = Implements(some_object);
>
>
> Note that issubtype will barf if you give it anything other than a type,
> so it may make more sense to define a function with exception handling
> and pull it out of the scope instead.
>
> HTH
>
> Michael Foord
>
>
> Jeff Slutter wrote:
> > I have a Python script that creates a class within it. This Python class
> > is derived off of a class, or interface, I made in C# - something like:
> >
> > class MyClass(Test.MainForm.IScript):
> >     ...
> >
> > Now, back in C#, I have gotten access to "MyClass" by:
> >
> > object myclass = someScope.GetVariable("MyClass");
> >
> > Is there a way to determine either:
> > a) what classes/interfaces MyClass implements OR
> > b) if it implements a specific class/interface
> >
> > I want to know if the "object myclass" is supposed to implement
> > Test.MainForm.IScript or not.
> >
> > I don't want to create an instance of MyClass as this would cause
> > problems, executing things I'm not ready to execute.
> >
> > Also, related but not as important, implementing an interface (as above)
> > doesn't cause any compiler errors if I'm missing functions - is there a
> > way to enforce this?
> >
> >
> > Thanks,
> > Jeff
> > _______________________________________________
> > Users mailing list
> > Users at lists.ironpython.com
> > http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
> >
>
>
> --
> http://www.ironpythoninaction.com/
> http://www.voidspace.org.uk/blog
>
>
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081215/32c8dc64/attachment.html>

From dinov at microsoft.com  Mon Dec 15 22:39:35 2008
From: dinov at microsoft.com (Dino Viehland)
Date: Mon, 15 Dec 2008 13:39:35 -0800
Subject: [IronPython] IronPython and file descriptors
In-Reply-To: <6f4025010812151252p2bb355d7p1b9d915c0b033016@mail.gmail.com>
References: <49467FDA.5000101@resolversystems.com>
	<d2155e360812150928i6e18806k309ce32ab98fde4e@mail.gmail.com>
	<494695C0.9040401@resolversystems.com>
	<d2155e360812150947vccb612ey9d32711bef718395@mail.gmail.com>
	<49469F86.8040901@resolversystems.com>
	<a623f7d10812151021u7e6dbba3k3b0971f4d42295f3@mail.gmail.com>
	<350E7D38B6D819428718949920EC2355564B62EAC4@NA-EXMSG-C102.redmond.corp.microsoft.com>
	<6f4025010812151252p2bb355d7p1b9d915c0b033016@mail.gmail.com>
Message-ID: <350E7D38B6D819428718949920EC2355564B62EC42@NA-EXMSG-C102.redmond.corp.microsoft.com>

(now I've replied to the correct thread...)



Can you open a feature request on CodePlex?  It's certainly an interesting idea to ponder and I'm leaning towards it but there's lots of details to be gotten right.



Do you know if this needs to work w/ sockets as well?  (There's also the question of can we make it work with sockets?  :))



There'll be a bunch of places we need to update (nt, socket, file, select, etc...) so I think it'll have to wait until 2.1 instead of coming in a minor update like 2.0.1.


From: users-bounces at lists.ironpython.com [mailto:users-bounces at lists.ironpython.com] On Behalf Of Michael Foord
Sent: Monday, December 15, 2008 12:53 PM
To: Discussion of IronPython
Subject: Re: [IronPython] IronPython and file descriptors


2008/12/15 Dino Viehland <dinov at microsoft.com<mailto:dinov at microsoft.com>>
Presumably CPython is linking against msvcrt and is using the C abstraction rather than calling the Win32 APIs which return handles directly.

As Curt said the biggest problem for us is that .NET does not expose C file descriptors.  Therefore we could fix this by P/Invoking out to msvcrt.  For users on other platforms we'd need to either add support for their platform or hope that their handles == C runtime file descriptors.  For something like fileno maybe this is ok because it's an interop point only - or are there any non-interop uses of fileno in the world?

What do people think of this?  This would be the 1st place where we would add a P/Invoke so I'd want to tread lightly and make sure this is really the right thing to do.


Well, some way of doing this is needed if certain Python C extensions are to work with Ironclad. Perhaps a global or per engine setting that allows file handles to be associated with a C descriptor. Users of Ironclad could switch it on if they wished and take the consequences.

Michael Foord



-----Original Message-----
From: users-bounces at lists.ironpython.com<mailto:users-bounces at lists.ironpython.com> [mailto:users-bounces at lists.ironpython.com<mailto:users-bounces at lists.ironpython.com>] On Behalf Of Slide
Sent: Monday, December 15, 2008 10:21 AM
To: Discussion of IronPython
Subject: Re: [IronPython] IronPython and file descriptors

On Mon, Dec 15, 2008 at 11:18 AM, Tom Wright
<tom.wright at resolversystems.com<mailto:tom.wright at resolversystems.com>> wrote:
> Agreed. It is certainly possible with some work to get a file descriptor and
> pass it to C code.
>
> This is not the problem, however. Ironclad's aim (eventually) is to allow
> *arbitrary* C extensions  to work with Ironpython without changing the
> extensions. Ctypes aim is to allow arbitrary C code to be run by python
> (possibly in other python libraries which you really don't want to modify).
>
> In CPython .fileno() is a file descriptor and some modules use this fact
> (specifically PIL) by passing file descriptors as integers into C code. I do
> not know how one can make this code work in IronPython unless .fileno()
> returns a file descriptor.
>
> The game here is not making a particular C library work with IronPython by
> modifying this library but making as many libraries as possible work without
> modification. Of course whether this is worthwhile depends on how many
> libraries rely on this fact.
>
> Sorry - I hope this is a little clearer.
>
> Tom


How does CPython get the file descriptor for fileno()? Does it just
return the HANDLE on Windows?

slide
_______________________________________________
Users mailing list
Users at lists.ironpython.com<mailto:Users at lists.ironpython.com>
http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
_______________________________________________
Users mailing list
Users at lists.ironpython.com<mailto:Users at lists.ironpython.com>
http://lists.ironpython.com/listinfo.cgi/users-ironpython.com



--
http://www.ironpythoninaction.com/

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081215/6c91d099/attachment.html>

From fuzzyman at voidspace.org.uk  Mon Dec 15 22:56:24 2008
From: fuzzyman at voidspace.org.uk (Michael Foord)
Date: Mon, 15 Dec 2008 21:56:24 +0000
Subject: [IronPython] IronPython and file descriptors
In-Reply-To: <350E7D38B6D819428718949920EC2355564B62EC42@NA-EXMSG-C102.redmond.corp.microsoft.com>
References: <49467FDA.5000101@resolversystems.com>	<d2155e360812150928i6e18806k309ce32ab98fde4e@mail.gmail.com>	<494695C0.9040401@resolversystems.com>	<d2155e360812150947vccb612ey9d32711bef718395@mail.gmail.com>	<49469F86.8040901@resolversystems.com>	<a623f7d10812151021u7e6dbba3k3b0971f4d42295f3@mail.gmail.com>	<350E7D38B6D819428718949920EC2355564B62EAC4@NA-EXMSG-C102.redmond.corp.microsoft.com>	<6f4025010812151252p2bb355d7p1b9d915c0b033016@mail.gmail.com>
	<350E7D38B6D819428718949920EC2355564B62EC42@NA-EXMSG-C102.redmond.corp.microsoft.com>
Message-ID: <4946D288.5020909@voidspace.org.uk>

Dino Viehland wrote:
>
> (now I?ve replied to the correct thread?)
>
> Can you open a feature request on CodePlex? It's certainly an 
> interesting idea to ponder and I'm leaning towards it but there's lots 
> of details to be gotten right.
>
> Do you know if this needs to work w/ sockets as well? (There's also 
> the question of can we make it work with sockets? :))
>
> There'll be a bunch of places we need to update (nt, socket, file, 
> select, etc...) so I think it'll have to wait until 2.1 instead of 
> coming in a minor update like 2.0.1.
>

Tom will have to comment on whether it is needed for sockets as well. 
Not yet I guess. :-)

Tom created an issue on Codeplex - I've added a comment to that:

http://www.codeplex.com/IronPython/WorkItem/View.aspx?WorkItemId=20242

Michael
>
> *From:* users-bounces at lists.ironpython.com 
> [mailto:users-bounces at lists.ironpython.com] *On Behalf Of *Michael Foord
> *Sent:* Monday, December 15, 2008 12:53 PM
> *To:* Discussion of IronPython
> *Subject:* Re: [IronPython] IronPython and file descriptors
>
> 2008/12/15 Dino Viehland <dinov at microsoft.com 
> <mailto:dinov at microsoft.com>>
>
> Presumably CPython is linking against msvcrt and is using the C 
> abstraction rather than calling the Win32 APIs which return handles 
> directly.
>
> As Curt said the biggest problem for us is that .NET does not expose C 
> file descriptors. Therefore we could fix this by P/Invoking out to 
> msvcrt. For users on other platforms we'd need to either add support 
> for their platform or hope that their handles == C runtime file 
> descriptors. For something like fileno maybe this is ok because it's 
> an interop point only - or are there any non-interop uses of fileno in 
> the world?
>
> What do people think of this? This would be the 1st place where we 
> would add a P/Invoke so I'd want to tread lightly and make sure this 
> is really the right thing to do.
>
>
>
> Well, some way of doing this is needed if certain Python C extensions 
> are to work with Ironclad. Perhaps a global or per engine setting that 
> allows file handles to be associated with a C descriptor. Users of 
> Ironclad could switch it on if they wished and take the consequences.
>
> Michael Foord
>
>
>     -----Original Message-----
>     From: users-bounces at lists.ironpython.com
>     <mailto:users-bounces at lists.ironpython.com>
>     [mailto:users-bounces at lists.ironpython.com
>     <mailto:users-bounces at lists.ironpython.com>] On Behalf Of Slide
>     Sent: Monday, December 15, 2008 10:21 AM
>     To: Discussion of IronPython
>     Subject: Re: [IronPython] IronPython and file descriptors
>
>     On Mon, Dec 15, 2008 at 11:18 AM, Tom Wright
>     <tom.wright at resolversystems.com
>     <mailto:tom.wright at resolversystems.com>> wrote:
>     > Agreed. It is certainly possible with some work to get a file
>     descriptor and
>     > pass it to C code.
>     >
>     > This is not the problem, however. Ironclad's aim (eventually) is
>     to allow
>     > *arbitrary* C extensions to work with Ironpython without
>     changing the
>     > extensions. Ctypes aim is to allow arbitrary C code to be run by
>     python
>     > (possibly in other python libraries which you really don't want
>     to modify).
>     >
>     > In CPython .fileno() is a file descriptor and some modules use
>     this fact
>     > (specifically PIL) by passing file descriptors as integers into
>     C code. I do
>     > not know how one can make this code work in IronPython unless
>     .fileno()
>     > returns a file descriptor.
>     >
>     > The game here is not making a particular C library work with
>     IronPython by
>     > modifying this library but making as many libraries as possible
>     work without
>     > modification. Of course whether this is worthwhile depends on
>     how many
>     > libraries rely on this fact.
>     >
>     > Sorry - I hope this is a little clearer.
>     >
>     > Tom
>
>
>     How does CPython get the file descriptor for fileno()? Does it just
>     return the HANDLE on Windows?
>
>     slide
>     _______________________________________________
>     Users mailing list
>     Users at lists.ironpython.com <mailto:Users at lists.ironpython.com>
>     http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>     _______________________________________________
>     Users mailing list
>     Users at lists.ironpython.com <mailto:Users at lists.ironpython.com>
>     http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>
>
>
>
> -- 
> http://www.ironpythoninaction.com/
>
> ------------------------------------------------------------------------
>
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>   


-- 
http://www.ironpythoninaction.com/
http://www.voidspace.org.uk/blog




From dinov at microsoft.com  Tue Dec 16 00:00:26 2008
From: dinov at microsoft.com (Dino Viehland)
Date: Mon, 15 Dec 2008 15:00:26 -0800
Subject: [IronPython] Compiled dlls and builtin modules
In-Reply-To: <2679f6510812120827x173452fbg97ba9ac332d63e92@mail.gmail.com>
References: <2679f6510812120827x173452fbg97ba9ac332d63e92@mail.gmail.com>
Message-ID: <350E7D38B6D819428718949920EC2355564B62ECF8@NA-EXMSG-C102.redmond.corp.microsoft.com>

I think this behavior is currently by design because we're using sys.meta_path which does take precedence over the built-in modules.  We could switch to using sys.path_hooks in 2.1 or 3.0 if the consensus is this behavior in undesirable (which would also give control over when pre-compiled modules take precedence over other places on sys.path).  Personally I'm +0 to make the change but this would probably break the other recently reported behavior w/ combinations on disk as .py files & precompiled files :)

From: users-bounces at lists.ironpython.com [mailto:users-bounces at lists.ironpython.com] On Behalf Of Glenn Jones
Sent: Friday, December 12, 2008 8:27 AM
To: Discussion of IronPython
Subject: [IronPython] Compiled dlls and builtin modules

Hi all,

We are compiling the python standard library (so it can be ngend when it is shipped with Resolver One) and we've discovered that modules in the dll take precedence over builtins. We noticed this because re and socket are included in the standard Lib directory (installed with IPy2) and when we ran against the dll built from the Lib, importing them failed.

We have removed the following files from the standard library (and we may have to remove some others):
socket, copy_reg, re, sre_compile, sre_constants, sre_parse (the sre_* were referenced in re)

A better solution would be for builtin modules to take precedence over modules in referenced assemblies.

As an experiment, we created a datetime module that just printed that it was being imported. When we included it in a dll and referenced it, it replaced the builtin datetime.

Glenn & Orestis
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081215/cbafe09d/attachment.html>

From fuzzyman at voidspace.org.uk  Tue Dec 16 00:04:05 2008
From: fuzzyman at voidspace.org.uk (Michael Foord)
Date: Mon, 15 Dec 2008 23:04:05 +0000
Subject: [IronPython] Compiled dlls and builtin modules
In-Reply-To: <350E7D38B6D819428718949920EC2355564B62ECF8@NA-EXMSG-C102.redmond.corp.microsoft.com>
References: <2679f6510812120827x173452fbg97ba9ac332d63e92@mail.gmail.com>
	<350E7D38B6D819428718949920EC2355564B62ECF8@NA-EXMSG-C102.redmond.corp.microsoft.com>
Message-ID: <4946E265.9000802@voidspace.org.uk>

Dino Viehland wrote:
>
> I think this behavior is currently by design because we?re using 
> sys.meta_path which does take precedence over the built-in modules. We 
> could switch to using sys.path_hooks in 2.1 or 3.0 if the consensus is 
> this behavior in undesirable (which would also give control over when 
> pre-compiled modules take precedence over other places on sys.path). 
> Personally I?m +0 to make the change but this would probably break the 
> other recently reported behavior w/ combinations on disk as .py files 
> & precompiled files J
>
Another suggestion is that you stop including these unneeded modules in 
the version of the standard library shipped with the IronPython 2 
installer.

Michael

> *From:* users-bounces at lists.ironpython.com 
> [mailto:users-bounces at lists.ironpython.com] *On Behalf Of *Glenn Jones
> *Sent:* Friday, December 12, 2008 8:27 AM
> *To:* Discussion of IronPython
> *Subject:* [IronPython] Compiled dlls and builtin modules
>
> Hi all,
>
> We are compiling the python standard library (so it can be ngend when 
> it is shipped with Resolver One) and we've discovered that modules in 
> the dll take precedence over builtins. We noticed this because re and 
> socket are included in the standard Lib directory (installed with 
> IPy2) and when we ran against the dll built from the Lib, importing 
> them failed.
>
> We have removed the following files from the standard library (and we 
> may have to remove some others):
> socket, copy_reg, re, sre_compile, sre_constants, sre_parse (the sre_* 
> were referenced in re)
>
> A better solution would be for builtin modules to take precedence over 
> modules in referenced assemblies.
>
> As an experiment, we created a datetime module that just printed that 
> it was being imported. When we included it in a dll and referenced it, 
> it replaced the builtin datetime.
>
> Glenn & Orestis
>
> ------------------------------------------------------------------------
>
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>   


-- 
http://www.ironpythoninaction.com/
http://www.voidspace.org.uk/blog




From curt at hagenlocher.org  Tue Dec 16 00:49:45 2008
From: curt at hagenlocher.org (Curt Hagenlocher)
Date: Mon, 15 Dec 2008 15:49:45 -0800
Subject: [IronPython] Compiled dlls and builtin modules
In-Reply-To: <4946E265.9000802@voidspace.org.uk>
References: <2679f6510812120827x173452fbg97ba9ac332d63e92@mail.gmail.com>
	<350E7D38B6D819428718949920EC2355564B62ECF8@NA-EXMSG-C102.redmond.corp.microsoft.com>
	<4946E265.9000802@voidspace.org.uk>
Message-ID: <d2155e360812151549o750621f4o4a8335189bceb6ad@mail.gmail.com>

Sounds waaay too easy :P.

In the long run, I'd prefer that we implement only the "C" part of these
modules and share the Python part with CPython.  But this is not only a
potentially breaking change, it also would require modification to the
standard library for at least the "socket" module.
On Mon, Dec 15, 2008 at 3:04 PM, Michael Foord <fuzzyman at voidspace.org.uk>wrote:

> Dino Viehland wrote:
>
>>
>> I think this behavior is currently by design because we're using
>> sys.meta_path which does take precedence over the built-in modules. We could
>> switch to using sys.path_hooks in 2.1 or 3.0 if the consensus is this
>> behavior in undesirable (which would also give control over when
>> pre-compiled modules take precedence over other places on sys.path).
>> Personally I'm +0 to make the change but this would probably break the other
>> recently reported behavior w/ combinations on disk as .py files &
>> precompiled files J
>>
>> Another suggestion is that you stop including these unneeded modules in
> the version of the standard library shipped with the IronPython 2 installer.
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081215/02b8e574/attachment.html>

From curt at hagenlocher.org  Tue Dec 16 01:29:02 2008
From: curt at hagenlocher.org (Curt Hagenlocher)
Date: Mon, 15 Dec 2008 16:29:02 -0800
Subject: [IronPython] Invalid construct in DLR's MSBuild file?
In-Reply-To: <5b0248170812100239i2abe78fby4c6c20a7fef6daef@mail.gmail.com>
References: <5b0248170812100239i2abe78fby4c6c20a7fef6daef@mail.gmail.com>
Message-ID: <d2155e360812151629v55c27776r3b3548977a07dbd1@mail.gmail.com>

This looks like a problem in quite a few of our csproj files.  Could you
open up a bug report on CodePlex for this?

Thanks,
-Curt

On Wed, Dec 10, 2008 at 2:39 AM, Seo Sanghyeon <sanxiyn at gmail.com> wrote:

> Hello, now I am trying to use xbuild (open source implementation of
> MSBuild language) to build IronPython, and the first error (after
> instrumenting the source to make it clear what the error is) is this:
>
> $ xbuild Microsoft.Scripting.ExtensionAttribute.csproj
> XBuild Engine Version 0.1
> Mono, Version 2.3.0.0
> Copyright (C) Marek Sieradzki 2005. All rights reserved.
> ConditionParser failed to parse  $(SilverlightBuild) != 'true'
>
> Unhandled Exception:
> Microsoft.Build.BuildEngine.InvalidProjectFileException: The requested
> feature is not implemented. ---> System.NotImplementedException: The
> requested feature is not implemented.
>  at Microsoft.Build.BuildEngine.ConditionParser.ParseFactorExpression
> () [0x00000]
>  at Microsoft.Build.BuildEngine.ConditionParser.ParseRelationalExpression
> () [0x00000]
>  at Microsoft.Build.BuildEngine.ConditionParser.ParseBooleanOr () [0x00000]
>  at Microsoft.Build.BuildEngine.ConditionParser.ParseBooleanAnd ()
> [0x00000]
>  at Microsoft.Build.BuildEngine.ConditionParser.ParseBooleanExpression
> () [0x00000]
>  at Microsoft.Build.BuildEngine.ConditionParser.ParseExpression ()
> [0x00000]
>  at Microsoft.Build.BuildEngine.ConditionParser.ParseCondition
> (System.String condition) [0x00000]
>  --- End of inner exception stack trace ---
>  at Microsoft.Build.BuildEngine.Project.DoLoad (System.IO.TextReader
> textReader) [0x00000]
>  at Microsoft.Build.BuildEngine.Project.Load (System.String
> projectFileName) [0x00000]
>  at Mono.XBuild.CommandLine.MainClass.Execute () [0x00000]
>
> According to MSDN, MSBuild Conditions
> http://msdn.microsoft.com/en-us/library/7szfhaft.aspx conditions
> should be of the form 'stringA' != 'stringB', where "single quotes are
> not required for simple alphanumeric strings or boolean values". Is
> $(SilverlightBuild) this case? If it isn't, shouldn't it be quoted as
> '$(SilverlightBuild)'?
>
> If I manually change this condition to be quoted (I also note, all
> other uses of $(SilverlightBuild) is actually quoted in this file),
> xbuild builds this assembly successfully.
>
> Even if this is actually valid (then I argue documentation is
> confusing), I think it would be a good idea to quote this anyway.
>
> --
> Seo Sanghyeon
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081215/d8f76c2c/attachment.html>

From dinov at microsoft.com  Tue Dec 16 01:36:44 2008
From: dinov at microsoft.com (Dino Viehland)
Date: Mon, 15 Dec 2008 16:36:44 -0800
Subject: [IronPython] Compiled dlls and builtin modules
In-Reply-To: <d2155e360812151549o750621f4o4a8335189bceb6ad@mail.gmail.com>
References: <2679f6510812120827x173452fbg97ba9ac332d63e92@mail.gmail.com>
	<350E7D38B6D819428718949920EC2355564B62ECF8@NA-EXMSG-C102.redmond.corp.microsoft.com>
	<4946E265.9000802@voidspace.org.uk>
	<d2155e360812151549o750621f4o4a8335189bceb6ad@mail.gmail.com>
Message-ID: <350E7D38B6D819428718949920EC2355564B62ED93@NA-EXMSG-C102.redmond.corp.microsoft.com>

Yeah, I would like to see us more closely match the C modules as well.  Socket is particularly challenging though because socket.py depends on the ref counting semantics of CPython's garbage collector.

From: users-bounces at lists.ironpython.com [mailto:users-bounces at lists.ironpython.com] On Behalf Of Curt Hagenlocher
Sent: Monday, December 15, 2008 3:50 PM
To: Discussion of IronPython
Subject: Re: [IronPython] Compiled dlls and builtin modules

Sounds waaay too easy :P.

In the long run, I'd prefer that we implement only the "C" part of these modules and share the Python part with CPython.  But this is not only a potentially breaking change, it also would require modification to the standard library for at least the "socket" module.
On Mon, Dec 15, 2008 at 3:04 PM, Michael Foord <fuzzyman at voidspace.org.uk<mailto:fuzzyman at voidspace.org.uk>> wrote:
Dino Viehland wrote:

I think this behavior is currently by design because we're using sys.meta_path which does take precedence over the built-in modules. We could switch to using sys.path_hooks in 2.1 or 3.0 if the consensus is this behavior in undesirable (which would also give control over when pre-compiled modules take precedence over other places on sys.path). Personally I'm +0 to make the change but this would probably break the other recently reported behavior w/ combinations on disk as .py files & precompiled files J
Another suggestion is that you stop including these unneeded modules in the version of the standard library shipped with the IronPython 2 installer.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081215/7e974bfa/attachment.html>

From dinov at microsoft.com  Tue Dec 16 01:37:38 2008
From: dinov at microsoft.com (Dino Viehland)
Date: Mon, 15 Dec 2008 16:37:38 -0800
Subject: [IronPython] Compiled dlls and builtin modules
In-Reply-To: <4946E265.9000802@voidspace.org.uk>
References: <2679f6510812120827x173452fbg97ba9ac332d63e92@mail.gmail.com>
	<350E7D38B6D819428718949920EC2355564B62ECF8@NA-EXMSG-C102.redmond.corp.microsoft.com>
	<4946E265.9000802@voidspace.org.uk>
Message-ID: <350E7D38B6D819428718949920EC2355564B62ED94@NA-EXMSG-C102.redmond.corp.microsoft.com>

Yeah, it's funny, I thought we were supposed to have done that for at least some of them.

Maybe these were just missed, but I agree we should definitely do at least that.


-----Original Message-----
From: users-bounces at lists.ironpython.com [mailto:users-bounces at lists.ironpython.com] On Behalf Of Michael Foord
Sent: Monday, December 15, 2008 3:04 PM
To: Discussion of IronPython
Subject: Re: [IronPython] Compiled dlls and builtin modules

Dino Viehland wrote:
>
> I think this behavior is currently by design because we're using
> sys.meta_path which does take precedence over the built-in modules. We
> could switch to using sys.path_hooks in 2.1 or 3.0 if the consensus is
> this behavior in undesirable (which would also give control over when
> pre-compiled modules take precedence over other places on sys.path).
> Personally I'm +0 to make the change but this would probably break the
> other recently reported behavior w/ combinations on disk as .py files
> & precompiled files J
>
Another suggestion is that you stop including these unneeded modules in
the version of the standard library shipped with the IronPython 2
installer.

Michael

> *From:* users-bounces at lists.ironpython.com
> [mailto:users-bounces at lists.ironpython.com] *On Behalf Of *Glenn Jones
> *Sent:* Friday, December 12, 2008 8:27 AM
> *To:* Discussion of IronPython
> *Subject:* [IronPython] Compiled dlls and builtin modules
>
> Hi all,
>
> We are compiling the python standard library (so it can be ngend when
> it is shipped with Resolver One) and we've discovered that modules in
> the dll take precedence over builtins. We noticed this because re and
> socket are included in the standard Lib directory (installed with
> IPy2) and when we ran against the dll built from the Lib, importing
> them failed.
>
> We have removed the following files from the standard library (and we
> may have to remove some others):
> socket, copy_reg, re, sre_compile, sre_constants, sre_parse (the sre_*
> were referenced in re)
>
> A better solution would be for builtin modules to take precedence over
> modules in referenced assemblies.
>
> As an experiment, we created a datetime module that just printed that
> it was being imported. When we included it in a dll and referenced it,
> it replaced the builtin datetime.
>
> Glenn & Orestis
>
> ------------------------------------------------------------------------
>
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>


--
http://www.ironpythoninaction.com/
http://www.voidspace.org.uk/blog


_______________________________________________
Users mailing list
Users at lists.ironpython.com
http://lists.ironpython.com/listinfo.cgi/users-ironpython.com



From dinov at microsoft.com  Tue Dec 16 08:12:33 2008
From: dinov at microsoft.com (Dino Viehland)
Date: Mon, 15 Dec 2008 23:12:33 -0800
Subject: [IronPython] Issue about string.upper and string.lower
In-Reply-To: <2679f6510812120727o58a85ec4ub686aeac4dc4820b@mail.gmail.com>
References: <2679f6510812120727o58a85ec4ub686aeac4dc4820b@mail.gmail.com>
Message-ID: <350E7D38B6D819428718949920EC2355564B62EE33@NA-EXMSG-C102.redmond.corp.microsoft.com>

I've actually looked at this not too long ago and I think your proposal of calling the Invariant functions is the correct solution.  I was looking at a few things:  This bug http://bugs.python.org/issue1528802, the 3.0 decimal.py module, and also just using Turkish I at the command prompt.  If you follow the comments the bug says:


"String upper and lower conversion are locale dependent and
implemented by the underlying libc, whereas Unicode
upper/lower conversion is not and only depends on the
Unicode character database."


That's a pretty clear statement that we shouldn't be using the current locale for our upper/lower string conversions.  It wouldn't surprise me if that breaks something somewhere because we won't be doing locale dependent conversions on what someone expects to its type to be str not unicode but in this case I think it'd be better to be consistent with the Unicode side of Python as that's our future.

As for the decimal module it doesn't change from 2.x to 3.0.  So .upper() apparently doesn't have this problem when CPython switches to Unicode strings.  Or at least no one's hit it, and when they do I think the resolution would be the same as 1528802.

Finally at the command prompt I could never get CPython to do a culture-sensitive operation.  I hadn't fully convinced myself on that part though because I hadn't yet escalated to a Turkish install of the OS running IronPython.

But I'm still pretty confident we're at fault and we should change our lower/upper implementation.  Obviously the change is easy but I'll do a full test pass to see if it breaks anything.

I also think calling ToUpper to get non-Pythonic results is easy enough (I actually think it's kind of better this way - it saves typing out the framework friendly ToUpperInvariant :)).

From: users-bounces at lists.ironpython.com [mailto:users-bounces at lists.ironpython.com] On Behalf Of Glenn Jones
Sent: Friday, December 12, 2008 7:27 AM
To: Discussion of IronPython
Subject: [IronPython] Issue about string.upper and string.lower

Hello everybody,

We ran across http://www.codeplex.com/IronPython/WorkItem/View.aspx?WorkItemId=13629 (turkish collation issues) today while trying to port Resolver One to IronPython 2.0.

This is in a test that tries to import decimal while in the turkish locale (it was actually reported by a user!). It does an .upper on a string with an 'i', and that doesn't give the expected results.

This is a very big issue because all the python code out there expects string transformations to be locale-independent, and there may be strange bugs in strange places.

Is mapping .upper and .lower to ToUpperInvariant and ToLowerInvariant an acceptable solution? People that want to do locale-dependent transformations can always use the .NET specific ToUpper/ToLower.

We can work around the decimal being unimportable by hacking it, but clearly this is not a general solution. We will report other modules that might fail from this as we find them.

Thanks,
Glenn and Orestis
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081215/507cda43/attachment.html>

From hernan.martinez at ecc.es  Tue Dec 16 10:02:18 2008
From: hernan.martinez at ecc.es (=?iso-8859-1?Q?Hern=E1n_Mart=EDnez_Foffani?=)
Date: Tue, 16 Dec 2008 10:02:18 +0100
Subject: [IronPython] Help contents
In-Reply-To: <d2155e360812151050i6185f2fdk807189e5ed535d90@mail.gmail.com>
Message-ID: <EDEELHKACGEJPBNKEMDBCEKKDLAA.hernan.martinez@ecc.es>

Thanks a lot again.

> If you have only one version/culture, I believe you can just place
> the XML doc in the same directory as the DLL. 
> 
> 
>> 
>> It seems clear now, thank you.
>> 
>> Reading DocBuilder.cs:GetXPathDocument(..) I assume that if I want
>> to let our users to be able to get IP console help on our own C#
>> libraries, we can place the XML docs in a culture subdirectory under
>> the DLLs install directory, right?
>> 
>> 



From kristian.jaksch at gmail.com  Tue Dec 16 10:25:39 2008
From: kristian.jaksch at gmail.com (xkrja)
Date: Tue, 16 Dec 2008 01:25:39 -0800 (PST)
Subject: [IronPython]  Newbie: convert string to python expression??
Message-ID: <21029759.post@talk.nabble.com>


Hi,
I'm just starting out learning IronPython. How can I convert a string that
someone writes in a textbox to a python expression or statement?? Compare it
to the console input: If I write for example a=2 in the console it's not
just a string I assume? It becomes an expression in the python code or
something like that?

This is what I got so far:

import clr, sys
import Microsoft

clr.AddReference("Mapack, Version=1.0.0.0, Culture=neutral,
PublicKeyToken=null")
clr.AddReference('Microsoft.Scripting')
clr.AddReference('Microsoft.Scripting.Core')

from System.Windows import Application
from System.Windows.Controls import UserControl
from Mapack import *

class Writer(object):
    def __init__(self, element):
        self.element = element
    def write(self, text):
        self.element.Text = self.element.Text + text
        
def inputBox_KeyDown(s, e): #I want to check the key down events. If 'Enter'
is pressed I want the 
    root.message.Text = ''      #string that is written in the textbox to be
converted to an expression
    key = e.Key.value__         #or statement or whatever it is
    if key == 3:
        #If key '3' (Enter key) is pressed I want the text in "infoBox" to
be read and converted? How???
         
root = Application.Current.LoadRootVisual(UserControl(), "app.xaml")
sys.stdout = Writer(root.message)

root.inputBox.KeyDown += inputBox_KeyDown

Thanks for help!
-- 
View this message in context: http://www.nabble.com/Newbie%3A-convert-string-to-python-expression---tp21029759p21029759.html
Sent from the IronPython mailing list archive at Nabble.com.



From konryd at gmail.com  Tue Dec 16 11:21:56 2008
From: konryd at gmail.com (Konrad Delong)
Date: Tue, 16 Dec 2008 11:21:56 +0100
Subject: [IronPython] Newbie: convert string to python expression??
In-Reply-To: <21029759.post@talk.nabble.com>
References: <21029759.post@talk.nabble.com>
Message-ID: <74401640812160221o18f71aape038806a11bf88f7@mail.gmail.com>

2008/12/16 xkrja <kristian.jaksch at gmail.com>:
>
> Hi,
> I'm just starting out learning IronPython. How can I convert a string that
> someone writes in a textbox to a python expression or statement?? Compare it
> to the console input: If I write for example a=2 in the console it's not
> just a string I assume? It becomes an expression in the python code or
> something like that

If I understood well, you're looking for eval() function. You use it this way:
eval("1 + 4") # => 5

In your case:
root.message.Text = eval(root.message.Text)

eval() is one of the built-in python functions. You can read about
rest of them on
http://www.python.org/doc/2.5.2/lib/built-in-funcs.html

hope that answers your question.
Konrad


From tom.wright at resolversystems.com  Tue Dec 16 12:02:44 2008
From: tom.wright at resolversystems.com (Tom Wright)
Date: Tue, 16 Dec 2008 11:02:44 +0000
Subject: [IronPython] IronPython and file descriptors
In-Reply-To: <4946D288.5020909@voidspace.org.uk>
References: <49467FDA.5000101@resolversystems.com>	<d2155e360812150928i6e18806k309ce32ab98fde4e@mail.gmail.com>	<494695C0.9040401@resolversystems.com>	<d2155e360812150947vccb612ey9d32711bef718395@mail.gmail.com>	<49469F86.8040901@resolversystems.com>	<a623f7d10812151021u7e6dbba3k3b0971f4d42295f3@mail.gmail.com>	<350E7D38B6D819428718949920EC2355564B62EAC4@NA-EXMSG-C102.redmond.corp.microsoft.com>	<6f4025010812151252p2bb355d7p1b9d915c0b033016@mail.gmail.com>	<350E7D38B6D819428718949920EC2355564B62EC42@NA-EXMSG-C102.redmond.corp.microsoft.com>
	<4946D288.5020909@voidspace.org.uk>
Message-ID: <49478AD4.1080002@resolversystems.com>

I'm not sure about sockets. The only example I have found in the wild so 
far is with files in PIL.

The python implementation seems to have socket.fileno() returning a file 
descriptor which comes from <sys/socket.h>'s socketpair function in the 
standard C library (see socketmodule.c if you are interested)

So, again, it would be possible for people to pass this identifier as an 
integer directly to C. Whether people do this in practice is a different 
question...

Tom



Michael Foord wrote:
> Dino Viehland wrote:
>>
>> (now I?ve replied to the correct thread?)
>>
>> Can you open a feature request on CodePlex? It's certainly an 
>> interesting idea to ponder and I'm leaning towards it but there's 
>> lots of details to be gotten right.
>>
>> Do you know if this needs to work w/ sockets as well? (There's also 
>> the question of can we make it work with sockets? :))
>>
>> There'll be a bunch of places we need to update (nt, socket, file, 
>> select, etc...) so I think it'll have to wait until 2.1 instead of 
>> coming in a minor update like 2.0.1.
>>
>
> Tom will have to comment on whether it is needed for sockets as well. 
> Not yet I guess. :-)
>
> Tom created an issue on Codeplex - I've added a comment to that:
>
> http://www.codeplex.com/IronPython/WorkItem/View.aspx?WorkItemId=20242
>
> Michael
>>
>> *From:* users-bounces at lists.ironpython.com 
>> [mailto:users-bounces at lists.ironpython.com] *On Behalf Of *Michael Foord
>> *Sent:* Monday, December 15, 2008 12:53 PM
>> *To:* Discussion of IronPython
>> *Subject:* Re: [IronPython] IronPython and file descriptors
>>
>> 2008/12/15 Dino Viehland <dinov at microsoft.com 
>> <mailto:dinov at microsoft.com>>
>>
>> Presumably CPython is linking against msvcrt and is using the C 
>> abstraction rather than calling the Win32 APIs which return handles 
>> directly.
>>
>> As Curt said the biggest problem for us is that .NET does not expose 
>> C file descriptors. Therefore we could fix this by P/Invoking out to 
>> msvcrt. For users on other platforms we'd need to either add support 
>> for their platform or hope that their handles == C runtime file 
>> descriptors. For something like fileno maybe this is ok because it's 
>> an interop point only - or are there any non-interop uses of fileno 
>> in the world?
>>
>> What do people think of this? This would be the 1st place where we 
>> would add a P/Invoke so I'd want to tread lightly and make sure this 
>> is really the right thing to do.
>>
>>
>>
>> Well, some way of doing this is needed if certain Python C extensions 
>> are to work with Ironclad. Perhaps a global or per engine setting 
>> that allows file handles to be associated with a C descriptor. Users 
>> of Ironclad could switch it on if they wished and take the consequences.
>>
>> Michael Foord
>>
>>
>>     -----Original Message-----
>>     From: users-bounces at lists.ironpython.com
>>     <mailto:users-bounces at lists.ironpython.com>
>>     [mailto:users-bounces at lists.ironpython.com
>>     <mailto:users-bounces at lists.ironpython.com>] On Behalf Of Slide
>>     Sent: Monday, December 15, 2008 10:21 AM
>>     To: Discussion of IronPython
>>     Subject: Re: [IronPython] IronPython and file descriptors
>>
>>     On Mon, Dec 15, 2008 at 11:18 AM, Tom Wright
>>     <tom.wright at resolversystems.com
>>     <mailto:tom.wright at resolversystems.com>> wrote:
>>     > Agreed. It is certainly possible with some work to get a file
>>     descriptor and
>>     > pass it to C code.
>>     >
>>     > This is not the problem, however. Ironclad's aim (eventually) is
>>     to allow
>>     > *arbitrary* C extensions to work with Ironpython without
>>     changing the
>>     > extensions. Ctypes aim is to allow arbitrary C code to be run by
>>     python
>>     > (possibly in other python libraries which you really don't want
>>     to modify).
>>     >
>>     > In CPython .fileno() is a file descriptor and some modules use
>>     this fact
>>     > (specifically PIL) by passing file descriptors as integers into
>>     C code. I do
>>     > not know how one can make this code work in IronPython unless
>>     .fileno()
>>     > returns a file descriptor.
>>     >
>>     > The game here is not making a particular C library work with
>>     IronPython by
>>     > modifying this library but making as many libraries as possible
>>     work without
>>     > modification. Of course whether this is worthwhile depends on
>>     how many
>>     > libraries rely on this fact.
>>     >
>>     > Sorry - I hope this is a little clearer.
>>     >
>>     > Tom
>>
>>
>>     How does CPython get the file descriptor for fileno()? Does it just
>>     return the HANDLE on Windows?
>>
>>     slide
>>     _______________________________________________
>>     Users mailing list
>>     Users at lists.ironpython.com <mailto:Users at lists.ironpython.com>
>>     http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>>     _______________________________________________
>>     Users mailing list
>>     Users at lists.ironpython.com <mailto:Users at lists.ironpython.com>
>>     http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>>
>>
>>
>>
>> -- 
>> http://www.ironpythoninaction.com/
>>
>> ------------------------------------------------------------------------
>>
>> _______________________________________________
>> Users mailing list
>> Users at lists.ironpython.com
>> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>>   
>
>



From kristian.jaksch at gmail.com  Tue Dec 16 12:14:38 2008
From: kristian.jaksch at gmail.com (xkrja)
Date: Tue, 16 Dec 2008 03:14:38 -0800 (PST)
Subject: [IronPython] Newbie: convert string to python expression??
In-Reply-To: <74401640812160221o18f71aape038806a11bf88f7@mail.gmail.com>
References: <21029759.post@talk.nabble.com>
	<74401640812160221o18f71aape038806a11bf88f7@mail.gmail.com>
Message-ID: <21031241.post@talk.nabble.com>


Thanks for the reply. This is what I have:

def inputBox_KeyDown(s, e):
    root.message.Text = ''
    key = e.Key.value__
    if key == 3:
        result = eval(root.inputBox.Text)
        root.message.Text = str(result)

eval() seems to work sometimes. For example if the text in the textbox is
1+1 . But if the text for example is a=1 i get an error "Unexpected token
'=' " (This is in Ironpython studio). I guess it is because a=1 is not an
expression. What should I use then? What if I don't know what the user will
type in? 

Thanks again!
-- 
View this message in context: http://www.nabble.com/Newbie%3A-convert-string-to-python-expression---tp21029759p21031241.html
Sent from the IronPython mailing list archive at Nabble.com.



From fuzzyman at voidspace.org.uk  Tue Dec 16 12:23:28 2008
From: fuzzyman at voidspace.org.uk (Michael Foord)
Date: Tue, 16 Dec 2008 11:23:28 +0000
Subject: [IronPython] Newbie: convert string to python expression??
In-Reply-To: <21031241.post@talk.nabble.com>
References: <21029759.post@talk.nabble.com>	<74401640812160221o18f71aape038806a11bf88f7@mail.gmail.com>
	<21031241.post@talk.nabble.com>
Message-ID: <49478FB0.2040803@voidspace.org.uk>

xkrja wrote:
> Thanks for the reply. This is what I have:
>
> def inputBox_KeyDown(s, e):
>     root.message.Text = ''
>     key = e.Key.value__
>     if key == 3:
>         result = eval(root.inputBox.Text)
>         root.message.Text = str(result)
>
> eval() seems to work sometimes. For example if the text in the textbox is
> 1+1 . But if the text for example is a=1 i get an error "Unexpected token
> '=' " (This is in Ironpython studio). I guess it is because a=1 is not an
> expression. What should I use then? What if I don't know what the user will
> type in? 
>
> Thanks again!
>   
If you only *want* an expression then a statement is indeed invalid. As 
you are accepting arbitrary user input then you will need error handling 
around the eval for when the user enters nonsense (as they are 
guaranteed to at some point...). You probably want a way to signal to 
the user what went wrong - perhaps a system message box.

Michael

-- 
http://www.ironpythoninaction.com/



From dblank at brynmawr.edu  Tue Dec 16 13:00:23 2008
From: dblank at brynmawr.edu (Douglas Blank)
Date: Tue, 16 Dec 2008 07:00:23 -0500 (EST)
Subject: [IronPython] Newbie: convert string to python expression??
In-Reply-To: <21031241.post@talk.nabble.com>
Message-ID: <1040764474.2765171229428823660.JavaMail.root@ganesh.brynmawr.edu>

This is just a Python issue (not specific to IronPython).

exec is used for statement(s) and eval() for expressions. You might have to:

try:
   eval(root.inputBox.Text)
except:
   exec root.inputBox.Text

-Doug

----- xkrja <kristian.jaksch at gmail.com> wrote:
> 
> Thanks for the reply. This is what I have:
> 
> def inputBox_KeyDown(s, e):
>     root.message.Text = ''
>     key = e.Key.value__
>     if key == 3:
>         result = eval(root.inputBox.Text)
>         root.message.Text = str(result)
> 
> eval() seems to work sometimes. For example if the text in the textbox is
> 1+1 . But if the text for example is a=1 i get an error "Unexpected token
> '=' " (This is in Ironpython studio). I guess it is because a=1 is not an
> expression. What should I use then? What if I don't know what the user will
> type in? 
> 
> Thanks again!
> -- 
> View this message in context: http://www.nabble.com/Newbie%3A-convert-string-to-python-expression---tp21029759p21031241.html
> Sent from the IronPython mailing list archive at Nabble.com.
> 
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com



From yashgt at gmail.com  Tue Dec 16 13:16:29 2008
From: yashgt at gmail.com (Yash Ganthe)
Date: Tue, 16 Dec 2008 17:46:29 +0530
Subject: [IronPython] Modules importing each other
Message-ID: <4f428e7b0812160416j3703c6a3h3f93bb4cb898a5aa@mail.gmail.com>

I use IronPython 2.0.

In Program.py I have:
print "hello"
DEBUGMODE=0
.
.
.
import LogMessage

In LogMessage.py I have
from Program import DEBUGMODE

When I run :
ipy Program.py
I observe that "hello" is printed twice: Once when Program.py is run and
next when a symbol from Program.py is imported in LogMessage. If one module
imports another, does it execute all statements in it even if that module
has been loaded before?
What do I need to do to avoid it?

Program.py and LogMessage.py are both in the same folder.

Thanks,
Yash
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081216/8b3bd225/attachment.html>

From fuzzyman at voidspace.org.uk  Tue Dec 16 13:19:22 2008
From: fuzzyman at voidspace.org.uk (Michael Foord)
Date: Tue, 16 Dec 2008 12:19:22 +0000
Subject: [IronPython] Modules importing each other
In-Reply-To: <4f428e7b0812160416j3703c6a3h3f93bb4cb898a5aa@mail.gmail.com>
References: <4f428e7b0812160416j3703c6a3h3f93bb4cb898a5aa@mail.gmail.com>
Message-ID: <49479CCA.6080208@voidspace.org.uk>

You would find the same thing in CPython. When you execute Program.py it 
is run with the name '__main__' (the main program being executed). When 
you import it, you import it with the name 'Program' which causes it to 
be imported with a different name. I suggest you put your debug symbol 
in a separate module that both Program.py and LogMessage.py import from.

Michael Foord


Yash Ganthe wrote:
> I use IronPython 2.0.
>
> In Program.py I have:
> print "hello"
> DEBUGMODE=0
> .
> .
> .
> import LogMessage
>
> In LogMessage.py I have
> from Program import DEBUGMODE
>
> When I run :
> ipy Program.py
> I observe that "hello" is printed twice: Once when Program.py is run 
> and next when a symbol from Program.py is imported in LogMessage. If 
> one module imports another, does it execute all statements in it even 
> if that module has been loaded before?
> What do I need to do to avoid it?
>
> Program.py and LogMessage.py are both in the same folder.
>
> Thanks,
> Yash
>
> ------------------------------------------------------------------------
>
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>   


-- 
http://www.ironpythoninaction.com/



From kristian.jaksch at gmail.com  Tue Dec 16 13:28:51 2008
From: kristian.jaksch at gmail.com (xkrja)
Date: Tue, 16 Dec 2008 04:28:51 -0800 (PST)
Subject: [IronPython] Newbie: convert string to python expression??
In-Reply-To: <1040764474.2765171229428823660.JavaMail.root@ganesh.brynmawr.edu>
References: <21029759.post@talk.nabble.com>
	<74401640812160221o18f71aape038806a11bf88f7@mail.gmail.com>
	<21031241.post@talk.nabble.com>
	<1040764474.2765171229428823660.JavaMail.root@ganesh.brynmawr.edu>
Message-ID: <21032160.post@talk.nabble.com>


Thanks for the replies guys.

I tried the solution below and it works in some cases. But a strange thing
is (remember I'm a newbie) that if I first type, for example a=1 in the
"inputBox" and then hit enter, the debugger goes down to exec
root.inputBox.Text and nothing is reported so I assume the statement is
executed correctly. 

But if I next time type print a in the "inputBox" the debugger goes down to
exec root.inputBox.Text again but now an exception is raised: "name 'a' is
not defined". Why is that??? I typed a=1!

Thanks!


Douglas Blank wrote:
> 
> This is just a Python issue (not specific to IronPython).
> 
> exec is used for statement(s) and eval() for expressions. You might have
> to:
> 
> try:
>    eval(root.inputBox.Text)
> except:
>    exec root.inputBox.Text
> 
> -Doug
> 
> 
> 
> 

-- 
View this message in context: http://www.nabble.com/Newbie%3A-convert-string-to-python-expression---tp21029759p21032160.html
Sent from the IronPython mailing list archive at Nabble.com.



From fuzzyman at voidspace.org.uk  Tue Dec 16 13:32:24 2008
From: fuzzyman at voidspace.org.uk (Michael Foord)
Date: Tue, 16 Dec 2008 12:32:24 +0000
Subject: [IronPython] Newbie: convert string to python expression??
In-Reply-To: <21032160.post@talk.nabble.com>
References: <21029759.post@talk.nabble.com>	<74401640812160221o18f71aape038806a11bf88f7@mail.gmail.com>	<21031241.post@talk.nabble.com>	<1040764474.2765171229428823660.JavaMail.root@ganesh.brynmawr.edu>
	<21032160.post@talk.nabble.com>
Message-ID: <49479FD8.2030501@voidspace.org.uk>

xkrja wrote:
> Thanks for the replies guys.
>
> I tried the solution below and it works in some cases. But a strange thing
> is (remember I'm a newbie) that if I first type, for example a=1 in the
> "inputBox" and then hit enter, the debugger goes down to exec
> root.inputBox.Text and nothing is reported so I assume the statement is
> executed correctly. 
>
> But if I next time type print a in the "inputBox" the debugger goes down to
> exec root.inputBox.Text again but now an exception is raised: "name 'a' is
> not defined". Why is that??? I typed a=1!
>   

You've snipped the code so I can't see it exactly, but I'm pretty sure 
you are doing this inside a method. This will create a new 'scope' every 
time you enter the method and so you are creating a local variable that 
disappears when you exit the method.

When you exec you can provide a dictionary as a context for the 
execution to happen in. If you store this as an instance member and 
re-use the same execution context every time then changes will be 
'remembered'.

Michael

> Thanks!
>
>
> Douglas Blank wrote:
>   
>> This is just a Python issue (not specific to IronPython).
>>
>> exec is used for statement(s) and eval() for expressions. You might have
>> to:
>>
>> try:
>>    eval(root.inputBox.Text)
>> except:
>>    exec root.inputBox.Text
>>
>> -Doug
>>
>>
>>
>>
>>     
>
>   


-- 
http://www.ironpythoninaction.com/



From kristian.jaksch at gmail.com  Tue Dec 16 15:25:01 2008
From: kristian.jaksch at gmail.com (xkrja)
Date: Tue, 16 Dec 2008 06:25:01 -0800 (PST)
Subject: [IronPython] Newbie: convert string to python expression??
In-Reply-To: <49479FD8.2030501@voidspace.org.uk>
References: <21029759.post@talk.nabble.com>
	<74401640812160221o18f71aape038806a11bf88f7@mail.gmail.com>
	<21031241.post@talk.nabble.com>
	<1040764474.2765171229428823660.JavaMail.root@ganesh.brynmawr.edu>
	<21032160.post@talk.nabble.com> <49479FD8.2030501@voidspace.org.uk>
Message-ID: <21033977.post@talk.nabble.com>


Thanks for the reply.

Below is a snippet of what I've got:

def inputBox_KeyDown(s, e):
    root.message.Text = ''
    key = e.Key.value__
    result = root.inputBox.Text
    if key == 3: #If 'Enter' key is pressed
        try:
          root.message.Text = eval(result)
        except:
          exec result
         
root = Application.Current.LoadRootVisual(UserControl(), "app.xaml")
sys.stdout = Writer(root.message)

root.inputBox.KeyDown += inputBox_KeyDown

Can't I just put everthing that is evaluated or executed in some kind of
global scope so that they can be accessed just like in the console? If not:
I looked a little at dictionaries but didn't really understand how to use
them in this case? Can someone give me a short example that can be used for
my case?

Once again, thanks for all help!


Michael Foord-5 wrote:
> 
> You've snipped the code so I can't see it exactly, but I'm pretty sure 
> you are doing this inside a method. This will create a new 'scope' every 
> time you enter the method and so you are creating a local variable that 
> disappears when you exit the method.
> 
> When you exec you can provide a dictionary as a context for the 
> execution to happen in. If you store this as an instance member and 
> re-use the same execution context every time then changes will be 
> 'remembered'.
> 
> Michael
> 

-- 
View this message in context: http://www.nabble.com/Newbie%3A-convert-string-to-python-expression---tp21029759p21033977.html
Sent from the IronPython mailing list archive at Nabble.com.



From fuzzyman at voidspace.org.uk  Tue Dec 16 16:23:57 2008
From: fuzzyman at voidspace.org.uk (Michael Foord)
Date: Tue, 16 Dec 2008 15:23:57 +0000
Subject: [IronPython] Newbie: convert string to python expression??
In-Reply-To: <21033977.post@talk.nabble.com>
References: <21029759.post@talk.nabble.com>	<74401640812160221o18f71aape038806a11bf88f7@mail.gmail.com>	<21031241.post@talk.nabble.com>	<1040764474.2765171229428823660.JavaMail.root@ganesh.brynmawr.edu>	<21032160.post@talk.nabble.com>
	<49479FD8.2030501@voidspace.org.uk> <21033977.post@talk.nabble.com>
Message-ID: <4947C80D.1070504@voidspace.org.uk>


context = {}

def inputBox_KeyDown(s, e):
    root.message.Text = ''
    key = e.Key.value__
    result = root.inputBox.Text
    if key == 3: #If 'Enter' key is pressed
	try:
          try:
            root.message.Text = eval(result)
          except SyntaxError:
            exec result in context
       except Exception, e:
         print 'Unhandled exception', e

Note the "exec result in context" and my suggested changes to your 
exception handling.

Michael


xkrja wrote:
> Thanks for the reply.
>
> Below is a snippet of what I've got:
>
> def inputBox_KeyDown(s, e):
>     root.message.Text = ''
>     key = e.Key.value__
>     result = root.inputBox.Text
>     if key == 3: #If 'Enter' key is pressed
>         try:
>           root.message.Text = eval(result)
>         except:
>           exec result
>          
> root = Application.Current.LoadRootVisual(UserControl(), "app.xaml")
> sys.stdout = Writer(root.message)
>
> root.inputBox.KeyDown += inputBox_KeyDown
>
> Can't I just put everthing that is evaluated or executed in some kind of
> global scope so that they can be accessed just like in the console? If not:
> I looked a little at dictionaries but didn't really understand how to use
> them in this case? Can someone give me a short example that can be used for
> my case?
>
> Once again, thanks for all help!
>
>
> Michael Foord-5 wrote:
>   
>> You've snipped the code so I can't see it exactly, but I'm pretty sure 
>> you are doing this inside a method. This will create a new 'scope' every 
>> time you enter the method and so you are creating a local variable that 
>> disappears when you exit the method.
>>
>> When you exec you can provide a dictionary as a context for the 
>> execution to happen in. If you store this as an instance member and 
>> re-use the same execution context every time then changes will be 
>> 'remembered'.
>>
>> Michael
>>
>>     
>
>   


-- 
http://www.ironpythoninaction.com/



From fuzzyman at voidspace.org.uk  Tue Dec 16 16:28:36 2008
From: fuzzyman at voidspace.org.uk (Michael Foord)
Date: Tue, 16 Dec 2008 15:28:36 +0000
Subject: [IronPython] Newbie: convert string to python expression??
In-Reply-To: <4947C80D.1070504@voidspace.org.uk>
References: <21029759.post@talk.nabble.com>	<74401640812160221o18f71aape038806a11bf88f7@mail.gmail.com>	<21031241.post@talk.nabble.com>	<1040764474.2765171229428823660.JavaMail.root@ganesh.brynmawr.edu>	<21032160.post@talk.nabble.com>	<49479FD8.2030501@voidspace.org.uk>
	<21033977.post@talk.nabble.com> <4947C80D.1070504@voidspace.org.uk>
Message-ID: <4947C924.9030107@voidspace.org.uk>

Oh - and Windows Forms has a Keys enumeration so that you don't have to 
rely on the underlying value of the event key property.

from System.Windows.Forms import Keys

if e.Key == Keys.Enter:

(or something like that - check out the MSDN documentation for the 
enumeration.)

Michael Foord


Michael Foord wrote:
>
> context = {}
>
> def inputBox_KeyDown(s, e):
>    root.message.Text = ''
>    key = e.Key.value__
>    result = root.inputBox.Text
>    if key == 3: #If 'Enter' key is pressed
>     try:
>          try:
>            root.message.Text = eval(result)
>          except SyntaxError:
>            exec result in context
>       except Exception, e:
>         print 'Unhandled exception', e
>
> Note the "exec result in context" and my suggested changes to your 
> exception handling.
>
> Michael
>
>
> xkrja wrote:
>> Thanks for the reply.
>>
>> Below is a snippet of what I've got:
>>
>> def inputBox_KeyDown(s, e):
>>     root.message.Text = ''
>>     key = e.Key.value__
>>     result = root.inputBox.Text
>>     if key == 3: #If 'Enter' key is pressed
>>         try:
>>           root.message.Text = eval(result)
>>         except:
>>           exec result
>>          root = Application.Current.LoadRootVisual(UserControl(), 
>> "app.xaml")
>> sys.stdout = Writer(root.message)
>>
>> root.inputBox.KeyDown += inputBox_KeyDown
>>
>> Can't I just put everthing that is evaluated or executed in some kind of
>> global scope so that they can be accessed just like in the console? 
>> If not:
>> I looked a little at dictionaries but didn't really understand how to 
>> use
>> them in this case? Can someone give me a short example that can be 
>> used for
>> my case?
>>
>> Once again, thanks for all help!
>>
>>
>> Michael Foord-5 wrote:
>>  
>>> You've snipped the code so I can't see it exactly, but I'm pretty 
>>> sure you are doing this inside a method. This will create a new 
>>> 'scope' every time you enter the method and so you are creating a 
>>> local variable that disappears when you exit the method.
>>>
>>> When you exec you can provide a dictionary as a context for the 
>>> execution to happen in. If you store this as an instance member and 
>>> re-use the same execution context every time then changes will be 
>>> 'remembered'.
>>>
>>> Michael
>>>
>>>     
>>
>>   
>
>


-- 
http://www.ironpythoninaction.com/



From kristian.jaksch at gmail.com  Tue Dec 16 17:13:03 2008
From: kristian.jaksch at gmail.com (xkrja)
Date: Tue, 16 Dec 2008 08:13:03 -0800 (PST)
Subject: [IronPython] Newbie: convert string to python expression??
In-Reply-To: <4947C924.9030107@voidspace.org.uk>
References: <21029759.post@talk.nabble.com>
	<74401640812160221o18f71aape038806a11bf88f7@mail.gmail.com>
	<21031241.post@talk.nabble.com>
	<1040764474.2765171229428823660.JavaMail.root@ganesh.brynmawr.edu>
	<21032160.post@talk.nabble.com> <49479FD8.2030501@voidspace.org.uk>
	<21033977.post@talk.nabble.com> <4947C80D.1070504@voidspace.org.uk>
	<4947C924.9030107@voidspace.org.uk>
Message-ID: <21036162.post@talk.nabble.com>


Thanks!

I'm using Ironpython with Silverlight so I can't get access to Windows.Forms

Please bear with me :-) New stuff's showing up all the time. The solution
you proposed worked but there must be something I don't get with the scope.
Please look at my snippet below:


import clr, sys
clr.AddReference("Mapack, Version=1.0.0.0, Culture=neutral,
PublicKeyToken=null")
from Mapack import *
.
.
.

def inputBox_KeyDown(s, e):
  
  key = e.Key.value__
  result = root.inputBox.Text
  if key == 3: #If 'Enter' key is pressed
       try:
           try:
               root.message.Text = eval(result)
           except SyntaxError:
              exec result in context
       except Exception, e:
           print 'Unhandled exception', e
       root.inputBox.Text = ""                  
       m = Matrix(2,2)                 #NOTICE: If I hard code this
definition in the function it works
       print m                            #but if I try to type
m=Matrix(2,2) in inputBox it says:
                                            #"Matrix is not defined"


I can create an object of the Matrix()-class if I code it straight into the
function as shown in my snippet above but I can't type m=Matrix(2,2) in the
'inputBox' and then execute it. Then I get an exception: "Matrix is not
defined". 

How can I work around this?

Thanks very much!
 


Michael Foord-5 wrote:
> 
> Oh - and Windows Forms has a Keys enumeration so that you don't have to 
> rely on the underlying value of the event key property.
> 
> from System.Windows.Forms import Keys
> 
> if e.Key == Keys.Enter:
> 
> (or something like that - check out the MSDN documentation for the 
> enumeration.)
> 
> Michael Foord
> 
> 

-- 
View this message in context: http://www.nabble.com/Newbie%3A-convert-string-to-python-expression---tp21029759p21036162.html
Sent from the IronPython mailing list archive at Nabble.com.



From fuzzyman at voidspace.org.uk  Tue Dec 16 17:19:04 2008
From: fuzzyman at voidspace.org.uk (Michael Foord)
Date: Tue, 16 Dec 2008 16:19:04 +0000
Subject: [IronPython] Newbie: convert string to python expression??
In-Reply-To: <21036162.post@talk.nabble.com>
References: <21029759.post@talk.nabble.com>	<74401640812160221o18f71aape038806a11bf88f7@mail.gmail.com>	<21031241.post@talk.nabble.com>	<1040764474.2765171229428823660.JavaMail.root@ganesh.brynmawr.edu>	<21032160.post@talk.nabble.com>
	<49479FD8.2030501@voidspace.org.uk>	<21033977.post@talk.nabble.com>
	<4947C80D.1070504@voidspace.org.uk>	<4947C924.9030107@voidspace.org.uk>
	<21036162.post@talk.nabble.com>
Message-ID: <4947D4F8.5010600@voidspace.org.uk>

xkrja wrote:
> Thanks!
>
> I'm using Ironpython with Silverlight so I can't get access to Windows.Forms
>   

Ah - in which case it probably lives in System.Windows.Input
> Please bear with me :-) New stuff's showing up all the time. The solution
> you proposed worked but there must be something I don't get with the scope.
> Please look at my snippet below:
>
>   

The advantage of a specific context is that you control what objects the 
code has access to. Try setting the Matrix object into the dictionary 
with the key 'm' and your code should have access to it.

Michael

> import clr, sys
> clr.AddReference("Mapack, Version=1.0.0.0, Culture=neutral,
> PublicKeyToken=null")
> from Mapack import *
> .
> .
> .
>
> def inputBox_KeyDown(s, e):
>   
>   key = e.Key.value__
>   result = root.inputBox.Text
>   if key == 3: #If 'Enter' key is pressed
>        try:
>            try:
>                root.message.Text = eval(result)
>            except SyntaxError:
>               exec result in context
>        except Exception, e:
>            print 'Unhandled exception', e
>        root.inputBox.Text = ""                  
>        m = Matrix(2,2)                 #NOTICE: If I hard code this
> definition in the function it works
>        print m                            #but if I try to type
> m=Matrix(2,2) in inputBox it says:
>                                             #"Matrix is not defined"
>
>
> I can create an object of the Matrix()-class if I code it straight into the
> function as shown in my snippet above but I can't type m=Matrix(2,2) in the
> 'inputBox' and then execute it. Then I get an exception: "Matrix is not
> defined". 
>
> How can I work around this?
>
> Thanks very much!
>  
>
>
> Michael Foord-5 wrote:
>   
>> Oh - and Windows Forms has a Keys enumeration so that you don't have to 
>> rely on the underlying value of the event key property.
>>
>> from System.Windows.Forms import Keys
>>
>> if e.Key == Keys.Enter:
>>
>> (or something like that - check out the MSDN documentation for the 
>> enumeration.)
>>
>> Michael Foord
>>
>>
>>     
>
>   


-- 
http://www.ironpythoninaction.com/
http://www.voidspace.org.uk/blog




From kristian.jaksch at gmail.com  Tue Dec 16 18:08:07 2008
From: kristian.jaksch at gmail.com (Kristian Jaksch)
Date: Tue, 16 Dec 2008 18:08:07 +0100
Subject: [IronPython] Newbie: convert string to python expression??
In-Reply-To: <4947D4F8.5010600@voidspace.org.uk>
References: <21029759.post@talk.nabble.com> <21031241.post@talk.nabble.com>
	<1040764474.2765171229428823660.JavaMail.root@ganesh.brynmawr.edu>
	<21032160.post@talk.nabble.com> <49479FD8.2030501@voidspace.org.uk>
	<21033977.post@talk.nabble.com> <4947C80D.1070504@voidspace.org.uk>
	<4947C924.9030107@voidspace.org.uk> <21036162.post@talk.nabble.com>
	<4947D4F8.5010600@voidspace.org.uk>
Message-ID: <1dab55500812160908s7edb66acg67e61f80357ee652@mail.gmail.com>

Ok, did you mean something like below?:
*
clr.AddReference("Mapack, Version=1.0.0.0, Culture=neutral,
PublicKeyToken=null")

from Mapack import **    * #A library that contain classes for working with
algebra etc.*

*context = {}

context['m'] = Matrix(5,5)    #creating a 5 rows, 5 columns matrix object in
context


def inputBox_KeyDown(s, e):
 key = e.Key.value__
 result = root.inputBox.Text
 if key == 3: #If 'Enter' key is pressed
      try:
          try:
              root.message.Text = eval(result)
          except SyntaxError:
             exec result in context
      except Exception, e:
          print 'Unhandled exception', e
      root.inputBox.Text = "" #Clearing inputBox*

But this creates a fixed size matrix and I still get an exception if I type
for example *m=Matrix(2,2)* in the 'inputBox'. I want to make this as
general as possible. Can't I place everything that is imported from 'Mapack'
into the 'context' dictionary together with whatever the user types in
besides of that? There are several more classes that must be placed in
'context'. It would be nice to have the user type *n=m.Transpose()* or
whatever method he wants to get out from the 'Mapack' library and then let
the code be executed correctly.

Thanks very much!


2008/12/16 Michael Foord <fuzzyman at voidspace.org.uk>:
> xkrja wrote:
>>
>> Thanks!
>>
>> I'm using Ironpython with Silverlight so I can't get access to
>> Windows.Forms
>>
>
> Ah - in which case it probably lives in System.Windows.Input
>>
>> Please bear with me :-) New stuff's showing up all the time. The solution
>> you proposed worked but there must be something I don't get with the
>> scope.
>> Please look at my snippet below:
>>
>>
>
> The advantage of a specific context is that you control what objects the
> code has access to. Try setting the Matrix object into the dictionary with
> the key 'm' and your code should have access to it.
>
> Michael
>
>> import clr, sys
>> clr.AddReference("Mapack, Version=1.0.0.0, Culture=neutral,
>> PublicKeyToken=null")
>> from Mapack import *
>> .
>> .
>> .
>>
>> def inputBox_KeyDown(s, e):
>>    key = e.Key.value__
>>  result = root.inputBox.Text
>>  if key == 3: #If 'Enter' key is pressed
>>       try:
>>           try:
>>               root.message.Text = eval(result)
>>           except SyntaxError:
>>              exec result in context
>>       except Exception, e:
>>           print 'Unhandled exception', e
>>       root.inputBox.Text = ""                         m = Matrix(2,2)
>>             #NOTICE: If I hard code this
>> definition in the function it works
>>       print m                            #but if I try to type
>> m=Matrix(2,2) in inputBox it says:
>>                                            #"Matrix is not defined"
>>
>>
>> I can create an object of the Matrix()-class if I code it straight into
>> the
>> function as shown in my snippet above but I can't type m=Matrix(2,2) in
>> the
>> 'inputBox' and then execute it. Then I get an exception: "Matrix is not
>> defined".
>> How can I work around this?
>>
>> Thanks very much!
>>
>>
>> Michael Foord-5 wrote:
>>
>>>
>>> Oh - and Windows Forms has a Keys enumeration so that you don't have to
>>> rely on the underlying value of the event key property.
>>>
>>> from System.Windows.Forms import Keys
>>>
>>> if e.Key == Keys.Enter:
>>>
>>> (or something like that - check out the MSDN documentation for the
>>> enumeration.)
>>>
>>> Michael Foord
>>>
>>>
>>>
>>
>>
>
>
> --
> http://www.ironpythoninaction.com/
> http://www.voidspace.org.uk/blog
>
>
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081216/bd898f09/attachment.html>

From fuzzyman at voidspace.org.uk  Tue Dec 16 18:13:27 2008
From: fuzzyman at voidspace.org.uk (Michael Foord)
Date: Tue, 16 Dec 2008 17:13:27 +0000
Subject: [IronPython] Newbie: convert string to python expression??
In-Reply-To: <1dab55500812160908s7edb66acg67e61f80357ee652@mail.gmail.com>
References: <21029759.post@talk.nabble.com>
	<21031241.post@talk.nabble.com>	<1040764474.2765171229428823660.JavaMail.root@ganesh.brynmawr.edu>	<21032160.post@talk.nabble.com>
	<49479FD8.2030501@voidspace.org.uk>	<21033977.post@talk.nabble.com>
	<4947C80D.1070504@voidspace.org.uk>	<4947C924.9030107@voidspace.org.uk>
	<21036162.post@talk.nabble.com>	<4947D4F8.5010600@voidspace.org.uk>
	<1dab55500812160908s7edb66acg67e61f80357ee652@mail.gmail.com>
Message-ID: <4947E1B7.7@voidspace.org.uk>

Kristian Jaksch wrote:
> Ok, did you mean something like below?:
> *
> clr.AddReference("Mapack, Version=1.0.0.0, Culture=neutral, 
> PublicKeyToken=null")
>
> from Mapack import **    * #A library that contain classes for working 
> with algebra etc.*
>
> *context = {}
>
> context['m'] = Matrix(5,5)    #creating a 5 rows, 5 columns matrix 
> object in context
>
> *

Looks good.
> *
> def inputBox_KeyDown(s, e):
>  key = e.Key.value__
>  result = root.inputBox.Text
>  if key == 3: #If 'Enter' key is pressed
>       try:
>           try:
>               root.message.Text = eval(result)
>           except SyntaxError:
>              exec result in context
>       except Exception, e:
>           print 'Unhandled exception', e
>       root.inputBox.Text = "" #Clearing inputBox*
>
> But this creates a fixed size matrix and I still get an exception if I 
> type for example *m=Matrix(2,2)* in the 'inputBox'. I want to make 
> this as general as possible. Can't I place everything that is imported 
> from 'Mapack' into the 'context' dictionary together with whatever the 
> user types in besides of that?

You can. I'll give you a clue - import * makes it harder. Try importing 
Mapack and then it will be easier to put things from the module into the 
dictionary.
> There are several more classes that must be placed in 'context'. It 
> would be nice to have the user type *n=m.Transpose()* or whatever 
> method he wants to get out from the 'Mapack' library and then let the 
> code be executed correctly.

Does that not work already - it should do.

Michael
>
> Thanks very much!
>
>
> 2008/12/16 Michael Foord <fuzzyman at voidspace.org.uk 
> <mailto:fuzzyman at voidspace.org.uk>>:
> > xkrja wrote:
> >>
> >> Thanks!
> >>
> >> I'm using Ironpython with Silverlight so I can't get access to
> >> Windows.Forms
> >>  
> >
> > Ah - in which case it probably lives in System.Windows.Input
> >>
> >> Please bear with me :-) New stuff's showing up all the time. The 
> solution
> >> you proposed worked but there must be something I don't get with the
> >> scope.
> >> Please look at my snippet below:
> >>
> >>  
> >
> > The advantage of a specific context is that you control what objects the
> > code has access to. Try setting the Matrix object into the 
> dictionary with
> > the key 'm' and your code should have access to it.
> >
> > Michael
> >
> >> import clr, sys
> >> clr.AddReference("Mapack, Version=1.0.0.0, Culture=neutral,
> >> PublicKeyToken=null")
> >> from Mapack import *
> >> .
> >> .
> >> .
> >>
> >> def inputBox_KeyDown(s, e):
> >>    key = e.Key.value__
> >>  result = root.inputBox.Text
> >>  if key == 3: #If 'Enter' key is pressed
> >>       try:
> >>           try:
> >>               root.message.Text = eval(result)
> >>           except SyntaxError:
> >>              exec result in context
> >>       except Exception, e:
> >>           print 'Unhandled exception', e
> >>       root.inputBox.Text = ""                         m = 
> Matrix(2,2)    
> >>             #NOTICE: If I hard code this
> >> definition in the function it works
> >>       print m                            #but if I try to type
> >> m=Matrix(2,2) in inputBox it says:
> >>                                            #"Matrix is not defined"
> >>
> >>
> >> I can create an object of the Matrix()-class if I code it straight into
> >> the
> >> function as shown in my snippet above but I can't type m=Matrix(2,2) in
> >> the
> >> 'inputBox' and then execute it. Then I get an exception: "Matrix is not
> >> defined".
> >> How can I work around this?
> >>
> >> Thanks very much!
> >>  
> >>
> >> Michael Foord-5 wrote:
> >>  
> >>>
> >>> Oh - and Windows Forms has a Keys enumeration so that you don't 
> have to
> >>> rely on the underlying value of the event key property.
> >>>
> >>> from System.Windows.Forms import Keys
> >>>
> >>> if e.Key == Keys.Enter:
> >>>
> >>> (or something like that - check out the MSDN documentation for the
> >>> enumeration.)
> >>>
> >>> Michael Foord
> >>>
> >>>
> >>>    
> >>
> >>  
> >
> >
> > --
> > http://www.ironpythoninaction.com/
> > http://www.voidspace.org.uk/blog
> >
> >
> > _______________________________________________
> > Users mailing list
> > Users at lists.ironpython.com <mailto:Users at lists.ironpython.com>
> > http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
> >
>
> ------------------------------------------------------------------------
>
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>   


-- 
http://www.ironpythoninaction.com/
http://www.voidspace.org.uk/blog




From orestis at resolversystems.com  Tue Dec 16 19:08:55 2008
From: orestis at resolversystems.com (Orestis Markou)
Date: Tue, 16 Dec 2008 18:08:55 +0000
Subject: [IronPython] Newbie: convert string to python expression??
In-Reply-To: <4947E1B7.7@voidspace.org.uk>
References: <21029759.post@talk.nabble.com>	<21031241.post@talk.nabble.com>	<1040764474.2765171229428823660.JavaMail.root@ganesh.brynmawr.edu>	<21032160.post@talk.nabble.com>	<49479FD8.2030501@voidspace.org.uk>	<21033977.post@talk.nabble.com>	<4947C80D.1070504@voidspace.org.uk>	<4947C924.9030107@voidspace.org.uk>	<21036162.post@talk.nabble.com>	<4947D4F8.5010600@voidspace.org.uk>	<1dab55500812160908s7edb66acg67e61f80357ee652@mail.gmail.com>
	<4947E1B7.7@voidspace.org.uk>
Message-ID: <4947EEB7.4050701@resolversystems.com>

you have to pass the context to eval as well:

eval(result, context)

make sure that the context doesn't change between calls to eval and 
exec. It should be a property on your form, really, but it doesn't seem 
so. Just make sure that you don't re-create the context every time.

If you don't want to setup the context everytime, what you can do is:

eval(result, globals(), locals())
exec(result, globals(), locals())

But this is a major security concern as the user can access (and 
change!) everything that is in every scope. You probably don't want to 
do that. Try printing globals() to see what kind of objects you have 
lying around.



Michael Foord wrote:
> Kristian Jaksch wrote:
>> Ok, did you mean something like below?:
>> *
>> clr.AddReference("Mapack, Version=1.0.0.0, Culture=neutral, 
>> PublicKeyToken=null")
>>
>> from Mapack import **    * #A library that contain classes for working 
>> with algebra etc.*
>>
>> *context = {}
>>
>> context['m'] = Matrix(5,5)    #creating a 5 rows, 5 columns matrix 
>> object in context
>>
>> *
> 
> Looks good.
>> *
>> def inputBox_KeyDown(s, e):
>>  key = e.Key.value__
>>  result = root.inputBox.Text
>>  if key == 3: #If 'Enter' key is pressed
>>       try:
>>           try:
>>               root.message.Text = eval(result)
>>           except SyntaxError:
>>              exec result in context
>>       except Exception, e:
>>           print 'Unhandled exception', e
>>       root.inputBox.Text = "" #Clearing inputBox*
>>
>> But this creates a fixed size matrix and I still get an exception if I 
>> type for example *m=Matrix(2,2)* in the 'inputBox'. I want to make 
>> this as general as possible. Can't I place everything that is imported 
>> from 'Mapack' into the 'context' dictionary together with whatever the 
>> user types in besides of that?
> 
> You can. I'll give you a clue - import * makes it harder. Try importing 
> Mapack and then it will be easier to put things from the module into the 
> dictionary.
>> There are several more classes that must be placed in 'context'. It 
>> would be nice to have the user type *n=m.Transpose()* or whatever 
>> method he wants to get out from the 'Mapack' library and then let the 
>> code be executed correctly.
> 
> Does that not work already - it should do.
> 
> Michael
>>
>> Thanks very much!
>>
>>
>> 2008/12/16 Michael Foord <fuzzyman at voidspace.org.uk 
>> <mailto:fuzzyman at voidspace.org.uk>>:
>> > xkrja wrote:
>> >>
>> >> Thanks!
>> >>
>> >> I'm using Ironpython with Silverlight so I can't get access to
>> >> Windows.Forms
>> >>  >
>> > Ah - in which case it probably lives in System.Windows.Input
>> >>
>> >> Please bear with me :-) New stuff's showing up all the time. The 
>> solution
>> >> you proposed worked but there must be something I don't get with the
>> >> scope.
>> >> Please look at my snippet below:
>> >>
>> >>  >
>> > The advantage of a specific context is that you control what objects 
>> the
>> > code has access to. Try setting the Matrix object into the 
>> dictionary with
>> > the key 'm' and your code should have access to it.
>> >
>> > Michael
>> >
>> >> import clr, sys
>> >> clr.AddReference("Mapack, Version=1.0.0.0, Culture=neutral,
>> >> PublicKeyToken=null")
>> >> from Mapack import *
>> >> .
>> >> .
>> >> .
>> >>
>> >> def inputBox_KeyDown(s, e):
>> >>    key = e.Key.value__
>> >>  result = root.inputBox.Text
>> >>  if key == 3: #If 'Enter' key is pressed
>> >>       try:
>> >>           try:
>> >>               root.message.Text = eval(result)
>> >>           except SyntaxError:
>> >>              exec result in context
>> >>       except Exception, e:
>> >>           print 'Unhandled exception', e
>> >>       root.inputBox.Text = ""                         m = 
>> Matrix(2,2)    >>             #NOTICE: If I hard code this
>> >> definition in the function it works
>> >>       print m                            #but if I try to type
>> >> m=Matrix(2,2) in inputBox it says:
>> >>                                            #"Matrix is not defined"
>> >>
>> >>
>> >> I can create an object of the Matrix()-class if I code it straight 
>> into
>> >> the
>> >> function as shown in my snippet above but I can't type 
>> m=Matrix(2,2) in
>> >> the
>> >> 'inputBox' and then execute it. Then I get an exception: "Matrix is 
>> not
>> >> defined".
>> >> How can I work around this?
>> >>
>> >> Thanks very much!
>> >>  >>
>> >> Michael Foord-5 wrote:
>> >>  >>>
>> >>> Oh - and Windows Forms has a Keys enumeration so that you don't 
>> have to
>> >>> rely on the underlying value of the event key property.
>> >>>
>> >>> from System.Windows.Forms import Keys
>> >>>
>> >>> if e.Key == Keys.Enter:
>> >>>
>> >>> (or something like that - check out the MSDN documentation for the
>> >>> enumeration.)
>> >>>
>> >>> Michael Foord
>> >>>
>> >>>
>> >>>    >>
>> >>  >
>> >
>> > --
>> > http://www.ironpythoninaction.com/
>> > http://www.voidspace.org.uk/blog
>> >
>> >
>> > _______________________________________________
>> > Users mailing list
>> > Users at lists.ironpython.com <mailto:Users at lists.ironpython.com>
>> > http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>> >
>>
>> ------------------------------------------------------------------------
>>
>> _______________________________________________
>> Users mailing list
>> Users at lists.ironpython.com
>> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>>   
> 
> 

-- 
Orestis Markou
Software Engineer,
Resolver Systems Ltd.
orestis at resolversystems.com
+44 (0) 20 7253 6372



From fuzzyman at voidspace.org.uk  Tue Dec 16 19:18:27 2008
From: fuzzyman at voidspace.org.uk (Michael Foord)
Date: Tue, 16 Dec 2008 18:18:27 +0000
Subject: [IronPython] Newbie: convert string to python expression??
In-Reply-To: <4947E1B7.7@voidspace.org.uk>
References: <21029759.post@talk.nabble.com>	<21031241.post@talk.nabble.com>	<1040764474.2765171229428823660.JavaMail.root@ganesh.brynmawr.edu>	<21032160.post@talk.nabble.com>	<49479FD8.2030501@voidspace.org.uk>	<21033977.post@talk.nabble.com>	<4947C80D.1070504@voidspace.org.uk>	<4947C924.9030107@voidspace.org.uk>	<21036162.post@talk.nabble.com>	<4947D4F8.5010600@voidspace.org.uk>	<1dab55500812160908s7edb66acg67e61f80357ee652@mail.gmail.com>
	<4947E1B7.7@voidspace.org.uk>
Message-ID: <4947F0F3.6000607@voidspace.org.uk>

Michael Foord wrote:
> Kristian Jaksch wrote:
>> *
>> def inputBox_KeyDown(s, e):
>>  key = e.Key.value__
>>  result = root.inputBox.Text
>>  if key == 3: #If 'Enter' key is pressed
>>       try:
>>           try:
>>               root.message.Text = eval(result)
>>           except SyntaxError:
>>              exec result in context
>>       except Exception, e:
>>           print 'Unhandled exception', e
>>       root.inputBox.Text = "" #Clearing inputBox*
>>
>> But this creates a fixed size matrix and I still get an exception if 
>> I type for example *m=Matrix(2,2)* in the 'inputBox'. I want to make 
>> this as general as possible. Can't I place everything that is 
>> imported from 'Mapack' into the 'context' dictionary together with 
>> whatever the user types in besides of that?
>
> You can. I'll give you a clue - import * makes it harder. Try 
> importing Mapack and then it will be easier to put things from the 
> module into the dictionary.

Hmmm - actually you could exec the import * inside the context to 
populate it.

Michael
>> There are several more classes that must be placed in 'context'. It 
>> would be nice to have the user type *n=m.Transpose()* or whatever 
>> method he wants to get out from the 'Mapack' library and then let the 
>> code be executed correctly.
>
> Does that not work already - it should do.
>
> Michael
>>
>> Thanks very much!
>>
>>
>> 2008/12/16 Michael Foord <fuzzyman at voidspace.org.uk 
>> <mailto:fuzzyman at voidspace.org.uk>>:
>> > xkrja wrote:
>> >>
>> >> Thanks!
>> >>
>> >> I'm using Ironpython with Silverlight so I can't get access to
>> >> Windows.Forms
>> >>  >
>> > Ah - in which case it probably lives in System.Windows.Input
>> >>
>> >> Please bear with me :-) New stuff's showing up all the time. The 
>> solution
>> >> you proposed worked but there must be something I don't get with the
>> >> scope.
>> >> Please look at my snippet below:
>> >>
>> >>  >
>> > The advantage of a specific context is that you control what 
>> objects the
>> > code has access to. Try setting the Matrix object into the 
>> dictionary with
>> > the key 'm' and your code should have access to it.
>> >
>> > Michael
>> >
>> >> import clr, sys
>> >> clr.AddReference("Mapack, Version=1.0.0.0, Culture=neutral,
>> >> PublicKeyToken=null")
>> >> from Mapack import *
>> >> .
>> >> .
>> >> .
>> >>
>> >> def inputBox_KeyDown(s, e):
>> >>    key = e.Key.value__
>> >>  result = root.inputBox.Text
>> >>  if key == 3: #If 'Enter' key is pressed
>> >>       try:
>> >>           try:
>> >>               root.message.Text = eval(result)
>> >>           except SyntaxError:
>> >>              exec result in context
>> >>       except Exception, e:
>> >>           print 'Unhandled exception', e
>> >>       root.inputBox.Text = ""                         m = 
>> Matrix(2,2)    >>             #NOTICE: If I hard code this
>> >> definition in the function it works
>> >>       print m                            #but if I try to type
>> >> m=Matrix(2,2) in inputBox it says:
>> >>                                            #"Matrix is not defined"
>> >>
>> >>
>> >> I can create an object of the Matrix()-class if I code it straight 
>> into
>> >> the
>> >> function as shown in my snippet above but I can't type 
>> m=Matrix(2,2) in
>> >> the
>> >> 'inputBox' and then execute it. Then I get an exception: "Matrix 
>> is not
>> >> defined".
>> >> How can I work around this?
>> >>
>> >> Thanks very much!
>> >>  >>
>> >> Michael Foord-5 wrote:
>> >>  >>>
>> >>> Oh - and Windows Forms has a Keys enumeration so that you don't 
>> have to
>> >>> rely on the underlying value of the event key property.
>> >>>
>> >>> from System.Windows.Forms import Keys
>> >>>
>> >>> if e.Key == Keys.Enter:
>> >>>
>> >>> (or something like that - check out the MSDN documentation for the
>> >>> enumeration.)
>> >>>
>> >>> Michael Foord
>> >>>
>> >>>
>> >>>    >>
>> >>  >
>> >
>> > --
>> > http://www.ironpythoninaction.com/
>> > http://www.voidspace.org.uk/blog
>> >
>> >
>> > _______________________________________________
>> > Users mailing list
>> > Users at lists.ironpython.com <mailto:Users at lists.ironpython.com>
>> > http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>> >
>>
>> ------------------------------------------------------------------------
>>
>> _______________________________________________
>> Users mailing list
>> Users at lists.ironpython.com
>> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>>   
>
>


-- 
http://www.ironpythoninaction.com/
http://www.voidspace.org.uk/blog




From kristian.jaksch at gmail.com  Tue Dec 16 19:45:56 2008
From: kristian.jaksch at gmail.com (Kristian Jaksch)
Date: Tue, 16 Dec 2008 19:45:56 +0100
Subject: [IronPython] Newbie: convert string to python expression??
In-Reply-To: <4947F0F3.6000607@voidspace.org.uk>
References: <21029759.post@talk.nabble.com> <49479FD8.2030501@voidspace.org.uk>
	<21033977.post@talk.nabble.com> <4947C80D.1070504@voidspace.org.uk>
	<4947C924.9030107@voidspace.org.uk> <21036162.post@talk.nabble.com>
	<4947D4F8.5010600@voidspace.org.uk>
	<1dab55500812160908s7edb66acg67e61f80357ee652@mail.gmail.com>
	<4947E1B7.7@voidspace.org.uk> <4947F0F3.6000607@voidspace.org.uk>
Message-ID: <1dab55500812161045r175637ddyb89aa37369222c9e@mail.gmail.com>

Orestis,

passing the 'context' to 'eval' seems like the right thing to do. I tried
your solution, see below:

*
clr.AddReference("Mapack, Version=1.0.0.0, Culture=neutral,
PublicKeyToken=null")

from Mapack import **    * #A library that contain classes for working with
algebra etc.*
.
.
.
*context = {}

def inputBox_KeyDown(s, e):
  key = e.Key.value__
  result = root.inputBox.Text
  if key == 3: #If 'Enter' key is pressed
       try:
           try:
               root.message.Text = eval(result, globals(), locals())
           except SyntaxError:
              exec(result, globals(), locals())
       except Exception, e:
           print 'Unhandled exception', e
       root.inputBox.Text = "" #Clearing inputBox

*If I then define my matrix (as always): I type *m=Matrix(2,2)* in
'inputBox' and I don't receive any exceptions so I assume the definition is
ok. But if I type *print m* I get an exception: "m is not defined". Why is
that?


Michael,

What do you mean exec the import * inside the context to poulate it?
Something like:

*context['Mapack'] = exec "from Mapack import *"   #Doesn't work

*Hmmm... think I'm getting tired.

Anyway thanks for help!



2008/12/16 Michael Foord <fuzzyman at voidspace.org.uk>

> Michael Foord wrote:
>
>> Kristian Jaksch wrote:
>>
>>> *
>>> def inputBox_KeyDown(s, e):
>>>  key = e.Key.value__
>>>  result = root.inputBox.Text
>>>  if key == 3: #If 'Enter' key is pressed
>>>      try:
>>>          try:
>>>              root.message.Text = eval(result)
>>>          except SyntaxError:
>>>             exec result in context
>>>      except Exception, e:
>>>          print 'Unhandled exception', e
>>>      root.inputBox.Text = "" #Clearing inputBox*
>>>
>>> But this creates a fixed size matrix and I still get an exception if I
>>> type for example *m=Matrix(2,2)* in the 'inputBox'. I want to make this as
>>> general as possible. Can't I place everything that is imported from 'Mapack'
>>> into the 'context' dictionary together with whatever the user types in
>>> besides of that?
>>>
>>
>> You can. I'll give you a clue - import * makes it harder. Try importing
>> Mapack and then it will be easier to put things from the module into the
>> dictionary.
>>
>
> Hmmm - actually you could exec the import * inside the context to populate
> it.
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081216/689e3654/attachment.html>

From kamil at dworakowski.name  Tue Dec 16 20:07:40 2008
From: kamil at dworakowski.name (Kamil Dworakowski)
Date: Tue, 16 Dec 2008 19:07:40 +0000
Subject: [IronPython] weird performance issue
Message-ID: <93fc29d60812161107i691af398w514a20bf11fe2844@mail.gmail.com>

We see a very strange side effect of running the follwing code on the
performance of recalculations in Resolver One.

def Dumbcorator(fn):
    def _inner(self, *args):
        return fn(self, *args)
    return _inner

class Locker(object):
    @property
    @Dumbcorator
    def thing(self):
        return 2

l = Locker()
l.thing

Specifically, one of the performance test reports results 4 to 8 times
worse with this snippet in than without (the clock is turned on after
this snippet executes). If I comment out this snippet from the code,
it suddenly runs 4 to 8 times faster. The fact that I can comment it
out means that it is not used in the test, so it is bizzare that it
slows it down.

We can't reproduce the slow down in a clean example though. I hope
something clicks for somebody and gives me a hint as to what is going
on. Tomorrow I'll continue on trying to minimize the code that is
being slowed down, and hopefully come up with some decent repro.


From kamil at dworakowski.name  Tue Dec 16 20:07:55 2008
From: kamil at dworakowski.name (Kamil Dworakowski)
Date: Tue, 16 Dec 2008 19:07:55 +0000
Subject: [IronPython] weird performance issue
Message-ID: <93fc29d60812161107y79acc3dqbd99391e42f24eed@mail.gmail.com>

We see a very strange side effect of running the follwing code on the
performance of recalculations in Resolver One.

def Dumbcorator(fn):
    def _inner(self, *args):
        return fn(self, *args)
    return _inner

class Locker(object):
    @property
    @Dumbcorator
    def thing(self):
        return 2

l = Locker()
l.thing

Specifically, one of the performance test reports results 4 to 8 times
worse with this snippet in than without (the clock is turned on after
this snippet executes). If I comment out this snippet from the code,
it suddenly runs 4 to 8 times faster. The fact that I can comment it
out means that it is not used in the test, so it is bizzare that it
slows it down.

We can't reproduce the slow down in a clean example though. I hope
something clicks for somebody and gives me a hint as to what is going
on. Tomorrow I'll continue on trying to minimize the code that is
being slowed down, and hopefully come up with some decent repro.


From fuzzyman at voidspace.org.uk  Tue Dec 16 20:14:17 2008
From: fuzzyman at voidspace.org.uk (Michael Foord)
Date: Tue, 16 Dec 2008 19:14:17 +0000
Subject: [IronPython] Newbie: convert string to python expression??
In-Reply-To: <1dab55500812161045r175637ddyb89aa37369222c9e@mail.gmail.com>
References: <21029759.post@talk.nabble.com>
	<49479FD8.2030501@voidspace.org.uk>	<21033977.post@talk.nabble.com>
	<4947C80D.1070504@voidspace.org.uk>	<4947C924.9030107@voidspace.org.uk>
	<21036162.post@talk.nabble.com>	<4947D4F8.5010600@voidspace.org.uk>	<1dab55500812160908s7edb66acg67e61f80357ee652@mail.gmail.com>	<4947E1B7.7@voidspace.org.uk>
	<4947F0F3.6000607@voidspace.org.uk>
	<1dab55500812161045r175637ddyb89aa37369222c9e@mail.gmail.com>
Message-ID: <4947FE09.107@voidspace.org.uk>

Kristian Jaksch wrote:
>
> Michael,
>
> What do you mean exec the import * inside the context to poulate it? 
> Something like:
>
> *context['Mapack'] = exec "from Mapack import *"   #Doesn't work
>
> *Hmmm... think I'm getting tired.
>
exec doesn't return anything.

Try:

*exec "from Mapack import *" in context*

It looked like from your response to Orestis that you were moving away 
from using an explicit context - which I think will get you into 
confusion with local variables created by your user code.

Michael
> Anyway thanks for help!
>
>
>
> 2008/12/16 Michael Foord <fuzzyman at voidspace.org.uk 
> <mailto:fuzzyman at voidspace.org.uk>>
>
>     Michael Foord wrote:
>
>         Kristian Jaksch wrote:
>
>             *
>             def inputBox_KeyDown(s, e):
>              key = e.Key.value__
>              result = root.inputBox.Text
>              if key == 3: #If 'Enter' key is pressed
>                  try:
>                      try:
>                          root.message.Text = eval(result)
>                      except SyntaxError:
>                         exec result in context
>                  except Exception, e:
>                      print 'Unhandled exception', e
>                  root.inputBox.Text = "" #Clearing inputBox*
>
>             But this creates a fixed size matrix and I still get an
>             exception if I type for example *m=Matrix(2,2)* in the
>             'inputBox'. I want to make this as general as possible.
>             Can't I place everything that is imported from 'Mapack'
>             into the 'context' dictionary together with whatever the
>             user types in besides of that?
>
>
>         You can. I'll give you a clue - import * makes it harder. Try
>         importing Mapack and then it will be easier to put things from
>         the module into the dictionary.
>
>
>     Hmmm - actually you could exec the import * inside the context to
>     populate it.
>
> ------------------------------------------------------------------------
>
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>   


-- 
http://www.ironpythoninaction.com/
http://www.voidspace.org.uk/blog




From kristian.jaksch at gmail.com  Tue Dec 16 22:17:24 2008
From: kristian.jaksch at gmail.com (Kristian Jaksch)
Date: Tue, 16 Dec 2008 22:17:24 +0100
Subject: [IronPython] Newbie: convert string to python expression??
In-Reply-To: <4947FE09.107@voidspace.org.uk>
References: <21029759.post@talk.nabble.com> <4947C80D.1070504@voidspace.org.uk>
	<4947C924.9030107@voidspace.org.uk> <21036162.post@talk.nabble.com>
	<4947D4F8.5010600@voidspace.org.uk>
	<1dab55500812160908s7edb66acg67e61f80357ee652@mail.gmail.com>
	<4947E1B7.7@voidspace.org.uk> <4947F0F3.6000607@voidspace.org.uk>
	<1dab55500812161045r175637ddyb89aa37369222c9e@mail.gmail.com>
	<4947FE09.107@voidspace.org.uk>
Message-ID: <1dab55500812161317h19d1899et278e43d2288d6b1c@mail.gmail.com>

Michael,

Thank you for your help! It finally works. Now there is lots of other things
left but this has really helped a lot.



2008/12/16 Michael Foord <fuzzyman at voidspace.org.uk>

> Kristian Jaksch wrote:
>
>>
>> Michael,
>>
>> What do you mean exec the import * inside the context to poulate it?
>> Something like:
>>
>> *context['Mapack'] = exec "from Mapack import *"   #Doesn't work
>>
>> *Hmmm... think I'm getting tired.
>>
>>  exec doesn't return anything.
>
> Try:
>
> *exec "from Mapack import *" in context*
>
> It looked like from your response to Orestis that you were moving away from
> using an explicit context - which I think will get you into confusion with
> local variables created by your user code.
>
> Michael
>
>> Anyway thanks for help!
>>
>>
>>
>> 2008/12/16 Michael Foord <fuzzyman at voidspace.org.uk <mailto:
>> fuzzyman at voidspace.org.uk>>
>>
>>
>>    Michael Foord wrote:
>>
>>        Kristian Jaksch wrote:
>>
>>            *
>>            def inputBox_KeyDown(s, e):
>>             key = e.Key.value__
>>             result = root.inputBox.Text
>>             if key == 3: #If 'Enter' key is pressed
>>                 try:
>>                     try:
>>                         root.message.Text = eval(result)
>>                     except SyntaxError:
>>                        exec result in context
>>                 except Exception, e:
>>                     print 'Unhandled exception', e
>>                 root.inputBox.Text = "" #Clearing inputBox*
>>
>>            But this creates a fixed size matrix and I still get an
>>            exception if I type for example *m=Matrix(2,2)* in the
>>            'inputBox'. I want to make this as general as possible.
>>            Can't I place everything that is imported from 'Mapack'
>>            into the 'context' dictionary together with whatever the
>>            user types in besides of that?
>>
>>
>>        You can. I'll give you a clue - import * makes it harder. Try
>>        importing Mapack and then it will be easier to put things from
>>        the module into the dictionary.
>>
>>
>>    Hmmm - actually you could exec the import * inside the context to
>>    populate it.
>>
>> ------------------------------------------------------------------------
>>
>> _______________________________________________
>> Users mailing list
>> Users at lists.ironpython.com
>> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>>
>>
>
>
> --
> http://www.ironpythoninaction.com/
> http://www.voidspace.org.uk/blog
>
>
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081216/09705386/attachment.html>

From dinov at microsoft.com  Wed Dec 17 04:21:17 2008
From: dinov at microsoft.com (Dino Viehland)
Date: Tue, 16 Dec 2008 19:21:17 -0800
Subject: [IronPython] weird performance issue
In-Reply-To: <93fc29d60812161107y79acc3dqbd99391e42f24eed@mail.gmail.com>
References: <93fc29d60812161107y79acc3dqbd99391e42f24eed@mail.gmail.com>
Message-ID: <350E7D38B6D819428718949920EC2355564BF25378@NA-EXMSG-C102.redmond.corp.microsoft.com>

My guess would be that we're pushing the PythonContext._callSplatSite outside of the sweet spot in DLR site caches.

You could check this by putting a breakpoint in PythonContext.Call(object, params object[]).  Then look and see if _callSplatSite._rules is an instance of EmptyRuleSet<T> with _supportAdding == false.

If this is the cause it should be pretty easy to fix - property's should probably get their own call site object so it'll be specialized to the specific property (maybe we could share a common one of the function is a PythonFunction as these generate highly shareable calls between multiple function instances).  We're also doing a params call today which we could stop doing and get a nice perf boost on properties.

> -----Original Message-----
> From: users-bounces at lists.ironpython.com [mailto:users-
> bounces at lists.ironpython.com] On Behalf Of Kamil Dworakowski
> Sent: Tuesday, December 16, 2008 11:08 AM
> To: users at lists.ironpython.com
> Subject: [IronPython] weird performance issue
>
> We see a very strange side effect of running the follwing code on the
> performance of recalculations in Resolver One.
>
> def Dumbcorator(fn):
>     def _inner(self, *args):
>         return fn(self, *args)
>     return _inner
>
> class Locker(object):
>     @property
>     @Dumbcorator
>     def thing(self):
>         return 2
>
> l = Locker()
> l.thing
>
> Specifically, one of the performance test reports results 4 to 8 times
> worse with this snippet in than without (the clock is turned on after
> this snippet executes). If I comment out this snippet from the code,
> it suddenly runs 4 to 8 times faster. The fact that I can comment it
> out means that it is not used in the test, so it is bizzare that it
> slows it down.
>
> We can't reproduce the slow down in a clean example though. I hope
> something clicks for somebody and gives me a hint as to what is going
> on. Tomorrow I'll continue on trying to minimize the code that is
> being slowed down, and hopefully come up with some decent repro.
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com


From sanxiyn at gmail.com  Wed Dec 17 09:04:52 2008
From: sanxiyn at gmail.com (Seo Sanghyeon)
Date: Wed, 17 Dec 2008 17:04:52 +0900
Subject: [IronPython] Invalid construct in DLR's MSBuild file?
In-Reply-To: <d2155e360812151629v55c27776r3b3548977a07dbd1@mail.gmail.com>
References: <5b0248170812100239i2abe78fby4c6c20a7fef6daef@mail.gmail.com>
	<d2155e360812151629v55c27776r3b3548977a07dbd1@mail.gmail.com>
Message-ID: <5b0248170812170004y17a50132j90f09a691f338445@mail.gmail.com>

2008/12/16 Curt Hagenlocher <curt at hagenlocher.org>:
> This looks like a problem in quite a few of our csproj files.  Could you
> open up a bug report on CodePlex for this?

Done.
http://www.codeplex.com/IronPython/WorkItem/View.aspx?WorkItemId=20262

-- 
Seo Sanghyeon


From orestis at resolversystems.com  Wed Dec 17 11:14:41 2008
From: orestis at resolversystems.com (Orestis Markou)
Date: Wed, 17 Dec 2008 10:14:41 +0000
Subject: [IronPython] Newbie: convert string to python expression??
In-Reply-To: <1dab55500812161317h19d1899et278e43d2288d6b1c@mail.gmail.com>
References: <21029759.post@talk.nabble.com>
	<4947C80D.1070504@voidspace.org.uk>	<4947C924.9030107@voidspace.org.uk>
	<21036162.post@talk.nabble.com>	<4947D4F8.5010600@voidspace.org.uk>	<1dab55500812160908s7edb66acg67e61f80357ee652@mail.gmail.com>	<4947E1B7.7@voidspace.org.uk>
	<4947F0F3.6000607@voidspace.org.uk>	<1dab55500812161045r175637ddyb89aa37369222c9e@mail.gmail.com>	<4947FE09.107@voidspace.org.uk>
	<1dab55500812161317h19d1899et278e43d2288d6b1c@mail.gmail.com>
Message-ID: <4948D111.4030203@resolversystems.com>

You should definitely use a separate context, or you will get bitten by 
it later on.

To debug similar things, use the tried and true print: print the context 
before you execute something, and after. In your case:

print 'before', context
print eval(result, context)
print 'after', context

Also, in case Silverlight is difficult to debug, you can always rip out 
snippets and use them in either a text file or in the interactive 
interpreter.

Have fun!

Kristian Jaksch wrote:
> Michael,
> 
> Thank you for your help! It finally works. Now there is lots of other 
> things left but this has really helped a lot.
> 
> 
> 
> 2008/12/16 Michael Foord <fuzzyman at voidspace.org.uk 
> <mailto:fuzzyman at voidspace.org.uk>>
> 
>     Kristian Jaksch wrote:
> 
> 
>         Michael,
> 
>         What do you mean exec the import * inside the context to poulate
>         it? Something like:
> 
>         *context['Mapack'] = exec "from Mapack import *"   #Doesn't work
> 
>         *Hmmm... think I'm getting tired.
> 
>     exec doesn't return anything.
> 
>     Try:
> 
>     *exec "from Mapack import *" in context*
> 
>     It looked like from your response to Orestis that you were moving
>     away from using an explicit context - which I think will get you
>     into confusion with local variables created by your user code.
> 
>     Michael
> 
>         Anyway thanks for help!
> 
> 
> 
>         2008/12/16 Michael Foord <fuzzyman at voidspace.org.uk
>         <mailto:fuzzyman at voidspace.org.uk>
>         <mailto:fuzzyman at voidspace.org.uk
>         <mailto:fuzzyman at voidspace.org.uk>>>
> 
> 
>            Michael Foord wrote:
> 
>                Kristian Jaksch wrote:
> 
>                    *
>                    def inputBox_KeyDown(s, e):
>                     key = e.Key.value__
>                     result = root.inputBox.Text
>                     if key == 3: #If 'Enter' key is pressed
>                         try:
>                             try:
>                                 root.message.Text = eval(result)
>                             except SyntaxError:
>                                exec result in context
>                         except Exception, e:
>                             print 'Unhandled exception', e
>                         root.inputBox.Text = "" #Clearing inputBox*
> 
>                    But this creates a fixed size matrix and I still get an
>                    exception if I type for example *m=Matrix(2,2)* in the
>                    'inputBox'. I want to make this as general as possible.
>                    Can't I place everything that is imported from 'Mapack'
>                    into the 'context' dictionary together with whatever the
>                    user types in besides of that?
> 
> 
>                You can. I'll give you a clue - import * makes it harder. Try
>                importing Mapack and then it will be easier to put things
>         from
>                the module into the dictionary.
> 
> 
>            Hmmm - actually you could exec the import * inside the context to
>            populate it.
> 
>         ------------------------------------------------------------------------
> 
>         _______________________________________________
>         Users mailing list
>         Users at lists.ironpython.com <mailto:Users at lists.ironpython.com>
>         http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>          
> 
> 
> 
>     -- 
>     http://www.ironpythoninaction.com/
>     http://www.voidspace.org.uk/blog
> 
> 
>     _______________________________________________
>     Users mailing list
>     Users at lists.ironpython.com <mailto:Users at lists.ironpython.com>
>     http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
> 
> 
> 
> ------------------------------------------------------------------------
> 
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com

-- 
Orestis Markou
Software Engineer,
Resolver Systems Ltd.
orestis at resolversystems.com
+44 (0) 20 7253 6372



From william at resolversystems.com  Wed Dec 17 17:12:29 2008
From: william at resolversystems.com (William Reade)
Date: Wed, 17 Dec 2008 16:12:29 +0000
Subject: [IronPython] IronPython and file descriptors
In-Reply-To: <49478AD4.1080002@resolversystems.com>
References: <49467FDA.5000101@resolversystems.com>	<d2155e360812150928i6e18806k309ce32ab98fde4e@mail.gmail.com>	<494695C0.9040401@resolversystems.com>	<d2155e360812150947vccb612ey9d32711bef718395@mail.gmail.com>	<49469F86.8040901@resolversystems.com>	<a623f7d10812151021u7e6dbba3k3b0971f4d42295f3@mail.gmail.com>	<350E7D38B6D819428718949920EC2355564B62EAC4@NA-EXMSG-C102.redmond.corp.microsoft.com>	<6f4025010812151252p2bb355d7p1b9d915c0b033016@mail.gmail.com>	<350E7D38B6D819428718949920EC2355564B62EC42@NA-EXMSG-C102.redmond.corp.microsoft.com>	<4946D288.5020909@voidspace.org.uk>
	<49478AD4.1080002@resolversystems.com>
Message-ID: <494924ED.30604@resolversystems.com>

Just a quick note: you can't necessarily just P/invoke out to msvcrt, 
because CPython was built with msvcr71. Sorry :(.

Tom Wright wrote:
> I'm not sure about sockets. The only example I have found in the wild 
> so far is with files in PIL.
>
> The python implementation seems to have socket.fileno() returning a 
> file descriptor which comes from <sys/socket.h>'s socketpair function 
> in the standard C library (see socketmodule.c if you are interested)
>
> So, again, it would be possible for people to pass this identifier as 
> an integer directly to C. Whether people do this in practice is a 
> different question...
>
> Tom
>
>
>
> Michael Foord wrote:
>> Dino Viehland wrote:
>>>
>>> (now I?ve replied to the correct thread?)
>>>
>>> Can you open a feature request on CodePlex? It's certainly an 
>>> interesting idea to ponder and I'm leaning towards it but there's 
>>> lots of details to be gotten right.
>>>
>>> Do you know if this needs to work w/ sockets as well? (There's also 
>>> the question of can we make it work with sockets? :))
>>>
>>> There'll be a bunch of places we need to update (nt, socket, file, 
>>> select, etc...) so I think it'll have to wait until 2.1 instead of 
>>> coming in a minor update like 2.0.1.
>>>
>>
>> Tom will have to comment on whether it is needed for sockets as well. 
>> Not yet I guess. :-)
>>
>> Tom created an issue on Codeplex - I've added a comment to that:
>>
>> http://www.codeplex.com/IronPython/WorkItem/View.aspx?WorkItemId=20242
>>
>> Michael
>>>
>>> *From:* users-bounces at lists.ironpython.com 
>>> [mailto:users-bounces at lists.ironpython.com] *On Behalf Of *Michael 
>>> Foord
>>> *Sent:* Monday, December 15, 2008 12:53 PM
>>> *To:* Discussion of IronPython
>>> *Subject:* Re: [IronPython] IronPython and file descriptors
>>>
>>> 2008/12/15 Dino Viehland <dinov at microsoft.com 
>>> <mailto:dinov at microsoft.com>>
>>>
>>> Presumably CPython is linking against msvcrt and is using the C 
>>> abstraction rather than calling the Win32 APIs which return handles 
>>> directly.
>>>
>>> As Curt said the biggest problem for us is that .NET does not expose 
>>> C file descriptors. Therefore we could fix this by P/Invoking out to 
>>> msvcrt. For users on other platforms we'd need to either add support 
>>> for their platform or hope that their handles == C runtime file 
>>> descriptors. For something like fileno maybe this is ok because it's 
>>> an interop point only - or are there any non-interop uses of fileno 
>>> in the world?
>>>
>>> What do people think of this? This would be the 1st place where we 
>>> would add a P/Invoke so I'd want to tread lightly and make sure this 
>>> is really the right thing to do.
>>>
>>>
>>>
>>> Well, some way of doing this is needed if certain Python C 
>>> extensions are to work with Ironclad. Perhaps a global or per engine 
>>> setting that allows file handles to be associated with a C 
>>> descriptor. Users of Ironclad could switch it on if they wished and 
>>> take the consequences.
>>>
>>> Michael Foord
>>>
>>>
>>>     -----Original Message-----
>>>     From: users-bounces at lists.ironpython.com
>>>     <mailto:users-bounces at lists.ironpython.com>
>>>     [mailto:users-bounces at lists.ironpython.com
>>>     <mailto:users-bounces at lists.ironpython.com>] On Behalf Of Slide
>>>     Sent: Monday, December 15, 2008 10:21 AM
>>>     To: Discussion of IronPython
>>>     Subject: Re: [IronPython] IronPython and file descriptors
>>>
>>>     On Mon, Dec 15, 2008 at 11:18 AM, Tom Wright
>>>     <tom.wright at resolversystems.com
>>>     <mailto:tom.wright at resolversystems.com>> wrote:
>>>     > Agreed. It is certainly possible with some work to get a file
>>>     descriptor and
>>>     > pass it to C code.
>>>     >
>>>     > This is not the problem, however. Ironclad's aim (eventually) is
>>>     to allow
>>>     > *arbitrary* C extensions to work with Ironpython without
>>>     changing the
>>>     > extensions. Ctypes aim is to allow arbitrary C code to be run by
>>>     python
>>>     > (possibly in other python libraries which you really don't want
>>>     to modify).
>>>     >
>>>     > In CPython .fileno() is a file descriptor and some modules use
>>>     this fact
>>>     > (specifically PIL) by passing file descriptors as integers into
>>>     C code. I do
>>>     > not know how one can make this code work in IronPython unless
>>>     .fileno()
>>>     > returns a file descriptor.
>>>     >
>>>     > The game here is not making a particular C library work with
>>>     IronPython by
>>>     > modifying this library but making as many libraries as possible
>>>     work without
>>>     > modification. Of course whether this is worthwhile depends on
>>>     how many
>>>     > libraries rely on this fact.
>>>     >
>>>     > Sorry - I hope this is a little clearer.
>>>     >
>>>     > Tom
>>>
>>>
>>>     How does CPython get the file descriptor for fileno()? Does it just
>>>     return the HANDLE on Windows?
>>>
>>>     slide
>>>     _______________________________________________
>>>     Users mailing list
>>>     Users at lists.ironpython.com <mailto:Users at lists.ironpython.com>
>>>     http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>>>     _______________________________________________
>>>     Users mailing list
>>>     Users at lists.ironpython.com <mailto:Users at lists.ironpython.com>
>>>     http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>>>
>>>
>>>
>>>
>>> -- 
>>> http://www.ironpythoninaction.com/
>>>
>>> ------------------------------------------------------------------------ 
>>>
>>>
>>> _______________________________________________
>>> Users mailing list
>>> Users at lists.ironpython.com
>>> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>>>   
>>
>>
>
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>



From glenn.k.jones+ipy at gmail.com  Wed Dec 17 18:37:41 2008
From: glenn.k.jones+ipy at gmail.com (Glenn Jones)
Date: Wed, 17 Dec 2008 17:37:41 +0000
Subject: [IronPython] Change in stack traces
Message-ID: <2679f6510812170937wde3d409l8bdddde520166189@mail.gmail.com>

Hi all,

Another observation from the Resolver One upgrade:
When we run the following:
        engine = Python.CreateEngine()
        scope = engine.CreateScope()
        script = """
def Fn():
    raise Exception
Fn()
"""
        source = engine.CreateScriptSourceFromString(script, 'aname',
SourceCodeKind.Statements)
        code = source.Compile()
        code.Execute(scope)

We get a stacktrace that looks like this:
  File "stacktrace.py", line 10, in testStackTrace
    code.Execute(scope)
  File "aname", line 4, in aname
  File "aname", line 3, in Fn

The equivalent code in IPy1 and Python 2.5, has
  File "aname", line 4, in <module>
instead of
  File "aname", line 4, in aname


When we don't pass the name into the call to CreateScriptSourceFromString,
we get:
  File "stacktrace.py", line 10, in testStackTrace
    code.Execute(scope)
  File "<string>", line 4, in <module>
  File "<string>", line 3, in Fn

Is there a way for us to get behaviour like IPy1 and Python 2.5? Is this
possibly a bug in the traceback generation of IPy2?

Thanks
Glenn
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081217/92de877f/attachment.html>

From dinov at microsoft.com  Wed Dec 17 18:51:05 2008
From: dinov at microsoft.com (Dino Viehland)
Date: Wed, 17 Dec 2008 09:51:05 -0800
Subject: [IronPython] Change in stack traces
In-Reply-To: <2679f6510812170937wde3d409l8bdddde520166189@mail.gmail.com>
References: <2679f6510812170937wde3d409l8bdddde520166189@mail.gmail.com>
Message-ID: <350E7D38B6D819428718949920EC2355564BF254E7@NA-EXMSG-C102.redmond.corp.microsoft.com>

I'm guessing the comparison to CPython here is a comparison to code that has been exec'd or eval'd.  You can pass a PythonCompilerOptions to scriptSource.Compile with ModuleOptions.ExecOrEvalCode and you should then get the 1.x/CPython behavior.

From: users-bounces at lists.ironpython.com [mailto:users-bounces at lists.ironpython.com] On Behalf Of Glenn Jones
Sent: Wednesday, December 17, 2008 9:38 AM
To: Discussion of IronPython
Subject: [IronPython] Change in stack traces

Hi all,

Another observation from the Resolver One upgrade:
When we run the following:
        engine = Python.CreateEngine()
        scope = engine.CreateScope()
        script = """
def Fn():
    raise Exception
Fn()
"""
        source = engine.CreateScriptSourceFromString(script, 'aname', SourceCodeKind.Statements)
        code = source.Compile()
        code.Execute(scope)

We get a stacktrace that looks like this:
  File "stacktrace.py", line 10, in testStackTrace
    code.Execute(scope)
  File "aname", line 4, in aname
  File "aname", line 3, in Fn

The equivalent code in IPy1 and Python 2.5, has
  File "aname", line 4, in <module>
instead of
  File "aname", line 4, in aname


When we don't pass the name into the call to CreateScriptSourceFromString, we get:
  File "stacktrace.py", line 10, in testStackTrace
    code.Execute(scope)
  File "<string>", line 4, in <module>
  File "<string>", line 3, in Fn

Is there a way for us to get behaviour like IPy1 and Python 2.5? Is this possibly a bug in the traceback generation of IPy2?

Thanks
Glenn
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081217/d5fdfd5b/attachment.html>

From kamil at dworakowski.name  Wed Dec 17 19:02:59 2008
From: kamil at dworakowski.name (Kamil Dworakowski)
Date: Wed, 17 Dec 2008 18:02:59 +0000
Subject: [IronPython] weird performance issue
In-Reply-To: <350E7D38B6D819428718949920EC2355564BF25378@NA-EXMSG-C102.redmond.corp.microsoft.com>
References: <93fc29d60812161107y79acc3dqbd99391e42f24eed@mail.gmail.com>
	<350E7D38B6D819428718949920EC2355564BF25378@NA-EXMSG-C102.redmond.corp.microsoft.com>
Message-ID: <93fc29d60812171002r1ff25939v67fbae80be9dee7a@mail.gmail.com>

Thanks Dino, it looks like you are right. I print whenever
_supportAdding == false.
It happens a lot with the snippet executed, whereas it doesn't happen
at all with
the snippet commented out.

We also discovered that the slow down only occurs when we compile code
to dlls (no _supportAdding == false if runned from sources).
Still can't come up with a repro though. If you provide us with a patch, we can
test if it solves the problem in our code.

As a short term solution we have removed all the properties from
decorated functions. This fixed one performance test. We moved to
other performance tests and we still see that some functions are
slower when executed from dlls, so the same slow down might be
triggered by sth else.

On Wed, Dec 17, 2008 at 3:21 AM, Dino Viehland <dinov at microsoft.com> wrote:
> My guess would be that we're pushing the PythonContext._callSplatSite outside of the sweet spot in DLR site caches.
>
> You could check this by putting a breakpoint in PythonContext.Call(object, params object[]).  Then look and see if _callSplatSite._rules is an instance of EmptyRuleSet<T> with _supportAdding == false.
>
> If this is the cause it should be pretty easy to fix - property's should probably get their own call site object so it'll be specialized to the specific property (maybe we could share a common one of the function is a PythonFunction as these generate highly shareable calls between multiple function instances).  We're also doing a params call today which we could stop doing and get a nice perf boost on properties.
>
>> -----Original Message-----
>> From: users-bounces at lists.ironpython.com [mailto:users-
>> bounces at lists.ironpython.com] On Behalf Of Kamil Dworakowski
>> Sent: Tuesday, December 16, 2008 11:08 AM
>> To: users at lists.ironpython.com
>> Subject: [IronPython] weird performance issue
>>
>> We see a very strange side effect of running the follwing code on the
>> performance of recalculations in Resolver One.
>>
>> def Dumbcorator(fn):
>>     def _inner(self, *args):
>>         return fn(self, *args)
>>     return _inner
>>
>> class Locker(object):
>>     @property
>>     @Dumbcorator
>>     def thing(self):
>>         return 2
>>
>> l = Locker()
>> l.thing
>>
>> Specifically, one of the performance test reports results 4 to 8 times
>> worse with this snippet in than without (the clock is turned on after
>> this snippet executes). If I comment out this snippet from the code,
>> it suddenly runs 4 to 8 times faster. The fact that I can comment it
>> out means that it is not used in the test, so it is bizzare that it
>> slows it down.
>>
>> We can't reproduce the slow down in a clean example though. I hope
>> something clicks for somebody and gives me a hint as to what is going
>> on. Tomorrow I'll continue on trying to minimize the code that is
>> being slowed down, and hopefully come up with some decent repro.
>> _______________________________________________
>> Users mailing list
>> Users at lists.ironpython.com
>> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>


From orestis at resolversystems.com  Wed Dec 17 19:15:52 2008
From: orestis at resolversystems.com (Orestis Markou)
Date: Wed, 17 Dec 2008 18:15:52 +0000
Subject: [IronPython] weird performance issue
In-Reply-To: <93fc29d60812171002r1ff25939v67fbae80be9dee7a@mail.gmail.com>
References: <93fc29d60812161107y79acc3dqbd99391e42f24eed@mail.gmail.com>	<350E7D38B6D819428718949920EC2355564BF25378@NA-EXMSG-C102.redmond.corp.microsoft.com>
	<93fc29d60812171002r1ff25939v67fbae80be9dee7a@mail.gmail.com>
Message-ID: <494941D8.4090709@resolversystems.com>

Also: It seems that even normal properties are slower than methods. This 
became apparent when we had property access in tight loops. The weird 
thing is that this also happens only when using compiled dlls.

Is there something inherently different with compiled python files? 
Perhaps some optimization not kicking in?

Kamil Dworakowski wrote:
> Thanks Dino, it looks like you are right. I print whenever
> _supportAdding == false.
> It happens a lot with the snippet executed, whereas it doesn't happen
> at all with
> the snippet commented out.
> 
> We also discovered that the slow down only occurs when we compile code
> to dlls (no _supportAdding == false if runned from sources).
> Still can't come up with a repro though. If you provide us with a patch, we can
> test if it solves the problem in our code.
> 
> As a short term solution we have removed all the properties from
> decorated functions. This fixed one performance test. We moved to
> other performance tests and we still see that some functions are
> slower when executed from dlls, so the same slow down might be
> triggered by sth else.
> 
> On Wed, Dec 17, 2008 at 3:21 AM, Dino Viehland <dinov at microsoft.com> wrote:
>> My guess would be that we're pushing the PythonContext._callSplatSite outside of the sweet spot in DLR site caches.
>>
>> You could check this by putting a breakpoint in PythonContext.Call(object, params object[]).  Then look and see if _callSplatSite._rules is an instance of EmptyRuleSet<T> with _supportAdding == false.
>>
>> If this is the cause it should be pretty easy to fix - property's should probably get their own call site object so it'll be specialized to the specific property (maybe we could share a common one of the function is a PythonFunction as these generate highly shareable calls between multiple function instances).  We're also doing a params call today which we could stop doing and get a nice perf boost on properties.
>>
>>> -----Original Message-----
>>> From: users-bounces at lists.ironpython.com [mailto:users-
>>> bounces at lists.ironpython.com] On Behalf Of Kamil Dworakowski
>>> Sent: Tuesday, December 16, 2008 11:08 AM
>>> To: users at lists.ironpython.com
>>> Subject: [IronPython] weird performance issue
>>>
>>> We see a very strange side effect of running the follwing code on the
>>> performance of recalculations in Resolver One.
>>>
>>> def Dumbcorator(fn):
>>>     def _inner(self, *args):
>>>         return fn(self, *args)
>>>     return _inner
>>>
>>> class Locker(object):
>>>     @property
>>>     @Dumbcorator
>>>     def thing(self):
>>>         return 2
>>>
>>> l = Locker()
>>> l.thing
>>>
>>> Specifically, one of the performance test reports results 4 to 8 times
>>> worse with this snippet in than without (the clock is turned on after
>>> this snippet executes). If I comment out this snippet from the code,
>>> it suddenly runs 4 to 8 times faster. The fact that I can comment it
>>> out means that it is not used in the test, so it is bizzare that it
>>> slows it down.
>>>
>>> We can't reproduce the slow down in a clean example though. I hope
>>> something clicks for somebody and gives me a hint as to what is going
>>> on. Tomorrow I'll continue on trying to minimize the code that is
>>> being slowed down, and hopefully come up with some decent repro.
>>> _______________________________________________
>>> Users mailing list
>>> Users at lists.ironpython.com
>>> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>> _______________________________________________
>> Users mailing list
>> Users at lists.ironpython.com
>> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>>
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com

-- 
Orestis Markou
Software Engineer,
Resolver Systems Ltd.
orestis at resolversystems.com
+44 (0) 20 7253 6372



From orestis at resolversystems.com  Wed Dec 17 19:19:09 2008
From: orestis at resolversystems.com (Orestis Markou)
Date: Wed, 17 Dec 2008 18:19:09 +0000
Subject: [IronPython] DLR spec document
Message-ID: <4949429D.1070001@resolversystems.com>

Hi all,

the Hosting API spec from 
http://blogs.msdn.com/ironpython/archive/2008/03/16/dlr-resources.aspx 
seems to contain insinuations that it is not up-to-date.

I was wondering if this actually stands and how much we should depend on it.

Thanks very much,
Orestis
-- 
Orestis Markou
Software Engineer,
Resolver Systems Ltd.
orestis at resolversystems.com
+44 (0) 20 7253 6372



From dinov at microsoft.com  Wed Dec 17 19:32:24 2008
From: dinov at microsoft.com (Dino Viehland)
Date: Wed, 17 Dec 2008 10:32:24 -0800
Subject: [IronPython] weird performance issue
In-Reply-To: <494941D8.4090709@resolversystems.com>
References: <93fc29d60812161107y79acc3dqbd99391e42f24eed@mail.gmail.com>
	<350E7D38B6D819428718949920EC2355564BF25378@NA-EXMSG-C102.redmond.corp.microsoft.com>
	<93fc29d60812171002r1ff25939v67fbae80be9dee7a@mail.gmail.com>
	<494941D8.4090709@resolversystems.com>
Message-ID: <350E7D38B6D819428718949920EC2355564BF25527@NA-EXMSG-C102.redmond.corp.microsoft.com>

There is one difference between pre-compiled & compiled at runtime: Global accesses will be a little bit slower with compiled files.  That's because the files need to support being loaded into multiple ScriptRuntime's where if we compile it they don't.  Other than that they should be basically the same.

I would expect properties to be a little slower than methods.  The reason is that we first dispatch to the property descriptor, and then it dispatches to the function.  But I think the perf fix will actually improve this somewhat as well.

> -----Original Message-----
> From: users-bounces at lists.ironpython.com [mailto:users-
> bounces at lists.ironpython.com] On Behalf Of Orestis Markou
> Sent: Wednesday, December 17, 2008 10:16 AM
> To: Discussion of IronPython
> Subject: Re: [IronPython] weird performance issue
>
> Also: It seems that even normal properties are slower than methods.
> This
> became apparent when we had property access in tight loops. The weird
> thing is that this also happens only when using compiled dlls.
>
> Is there something inherently different with compiled python files?
> Perhaps some optimization not kicking in?
>
> Kamil Dworakowski wrote:
> > Thanks Dino, it looks like you are right. I print whenever
> > _supportAdding == false.
> > It happens a lot with the snippet executed, whereas it doesn't happen
> > at all with
> > the snippet commented out.
> >
> > We also discovered that the slow down only occurs when we compile
> code
> > to dlls (no _supportAdding == false if runned from sources).
> > Still can't come up with a repro though. If you provide us with a
> patch, we can
> > test if it solves the problem in our code.
> >
> > As a short term solution we have removed all the properties from
> > decorated functions. This fixed one performance test. We moved to
> > other performance tests and we still see that some functions are
> > slower when executed from dlls, so the same slow down might be
> > triggered by sth else.
> >
> > On Wed, Dec 17, 2008 at 3:21 AM, Dino Viehland <dinov at microsoft.com>
> wrote:
> >> My guess would be that we're pushing the
> PythonContext._callSplatSite outside of the sweet spot in DLR site
> caches.
> >>
> >> You could check this by putting a breakpoint in
> PythonContext.Call(object, params object[]).  Then look and see if
> _callSplatSite._rules is an instance of EmptyRuleSet<T> with
> _supportAdding == false.
> >>
> >> If this is the cause it should be pretty easy to fix - property's
> should probably get their own call site object so it'll be specialized
> to the specific property (maybe we could share a common one of the
> function is a PythonFunction as these generate highly shareable calls
> between multiple function instances).  We're also doing a params call
> today which we could stop doing and get a nice perf boost on
> properties.
> >>
> >>> -----Original Message-----
> >>> From: users-bounces at lists.ironpython.com [mailto:users-
> >>> bounces at lists.ironpython.com] On Behalf Of Kamil Dworakowski
> >>> Sent: Tuesday, December 16, 2008 11:08 AM
> >>> To: users at lists.ironpython.com
> >>> Subject: [IronPython] weird performance issue
> >>>
> >>> We see a very strange side effect of running the follwing code on
> the
> >>> performance of recalculations in Resolver One.
> >>>
> >>> def Dumbcorator(fn):
> >>>     def _inner(self, *args):
> >>>         return fn(self, *args)
> >>>     return _inner
> >>>
> >>> class Locker(object):
> >>>     @property
> >>>     @Dumbcorator
> >>>     def thing(self):
> >>>         return 2
> >>>
> >>> l = Locker()
> >>> l.thing
> >>>
> >>> Specifically, one of the performance test reports results 4 to 8
> times
> >>> worse with this snippet in than without (the clock is turned on
> after
> >>> this snippet executes). If I comment out this snippet from the
> code,
> >>> it suddenly runs 4 to 8 times faster. The fact that I can comment
> it
> >>> out means that it is not used in the test, so it is bizzare that it
> >>> slows it down.
> >>>
> >>> We can't reproduce the slow down in a clean example though. I hope
> >>> something clicks for somebody and gives me a hint as to what is
> going
> >>> on. Tomorrow I'll continue on trying to minimize the code that is
> >>> being slowed down, and hopefully come up with some decent repro.
> >>> _______________________________________________
> >>> Users mailing list
> >>> Users at lists.ironpython.com
> >>> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
> >> _______________________________________________
> >> Users mailing list
> >> Users at lists.ironpython.com
> >> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
> >>
> > _______________________________________________
> > Users mailing list
> > Users at lists.ironpython.com
> > http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>
> --
> Orestis Markou
> Software Engineer,
> Resolver Systems Ltd.
> orestis at resolversystems.com
> +44 (0) 20 7253 6372
>
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com



From seshapv at microsoft.com  Wed Dec 17 19:21:55 2008
From: seshapv at microsoft.com (Seshadri Pillailokam Vijayaraghavan)
Date: Wed, 17 Dec 2008 10:21:55 -0800
Subject: [IronPython] DLR spec document
In-Reply-To: <4949429D.1070001@resolversystems.com>
References: <4949429D.1070001@resolversystems.com>
Message-ID: <4F06AAB8EC03C349A5D6D0A797C4881C6980B04E6C@NA-EXMSG-C112.redmond.corp.microsoft.com>

We recently created a separate codeplex project for the DLR and that will host the latest DLR documents (including the Hosting spec). Please see this - http://www.codeplex.com/dlr/Wiki/View.aspx?title=Docs%20and%20specs&referringTitle=Home

This page is guaranteed to contain the latest DLR documents from now on.

Thanks
sesh
-----Original Message-----
From: users-bounces at lists.ironpython.com [mailto:users-bounces at lists.ironpython.com] On Behalf Of Orestis Markou
Sent: Wednesday, December 17, 2008 10:19 AM
To: Discussion of IronPython
Subject: [IronPython] DLR spec document

Hi all,

the Hosting API spec from
http://blogs.msdn.com/ironpython/archive/2008/03/16/dlr-resources.aspx
seems to contain insinuations that it is not up-to-date.

I was wondering if this actually stands and how much we should depend on it.

Thanks very much,
Orestis
--
Orestis Markou
Software Engineer,
Resolver Systems Ltd.
orestis at resolversystems.com
+44 (0) 20 7253 6372

_______________________________________________
Users mailing list
Users at lists.ironpython.com
http://lists.ironpython.com/listinfo.cgi/users-ironpython.com



From dinov at microsoft.com  Wed Dec 17 20:02:18 2008
From: dinov at microsoft.com (Dino Viehland)
Date: Wed, 17 Dec 2008 11:02:18 -0800
Subject: [IronPython] weird performance issue
In-Reply-To: <93fc29d60812171002r1ff25939v67fbae80be9dee7a@mail.gmail.com>
References: <93fc29d60812161107y79acc3dqbd99391e42f24eed@mail.gmail.com>
	<350E7D38B6D819428718949920EC2355564BF25378@NA-EXMSG-C102.redmond.corp.microsoft.com>
	<93fc29d60812171002r1ff25939v67fbae80be9dee7a@mail.gmail.com>
Message-ID: <350E7D38B6D819428718949920EC2355564BF25555@NA-EXMSG-C102.redmond.corp.microsoft.com>

Here's the simplest version of the fix, it's just updating one file so properties get their own call sites.

I'll work on a more involved fix that I'll get into 2.1 soon and probably into 2.0.1 a few days later.

Note this diff is from 2.1 but I've updated the names of things that have changed since 2.0.

File: Descriptors.cs
===================================================================
--- Descriptors.cs;C633889  (server)    12/17/2008 10:43 AM
+++ Descriptors.cs  (local)    12/17/2008 10:43 AM
@@ -21,6 +21,10 @@

 using IronPython.Runtime.Operations;
 using IronPython.Runtime.Types;
+using IronPython.Runtime.Binding;
+using Microsoft.Scripting.Actions;

 namespace IronPython.Runtime {
     [PythonType]
@@ -92,6 +96,7 @@
     [PythonType]
     public class PythonProperty : PythonTypeSlot {
         private object _fget, _fset, _fdel, _doc;
+        private CallSite<Func<CallSite, CodeContext, object, object, object>> _getSite;

         public PythonProperty() {
         }
@@ -177,7 +182,15 @@

         public new object __get__(CodeContext/*!*/ context, object instance, object owner) {
             if (instance == null) return this;
-            if (fget != null) return PythonCalls.Call(context, fget, instance);
+            if (_getSite == null) {
+                _getSite = CallSite<Func<CallSite, CodeContext, object, object, object>>.Create(
+                    new InvokeBinder(
+                        PythonContext.GetContext(context).DefaultBinderState,
+                        new CallSignature(1)
+                    )
+                );
+            }
+            if (fget != null) return _getSite.Target(_getSite, context, fget, instance);
             throw PythonOps.AttributeError("unreadable attribute");
         }



> -----Original Message-----
> From: users-bounces at lists.ironpython.com [mailto:users-
> bounces at lists.ironpython.com] On Behalf Of Kamil Dworakowski
> Sent: Wednesday, December 17, 2008 10:03 AM
> To: Discussion of IronPython
> Subject: Re: [IronPython] weird performance issue
>
> Thanks Dino, it looks like you are right. I print whenever
> _supportAdding == false.
> It happens a lot with the snippet executed, whereas it doesn't happen
> at all with
> the snippet commented out.
>
> We also discovered that the slow down only occurs when we compile code
> to dlls (no _supportAdding == false if runned from sources).
> Still can't come up with a repro though. If you provide us with a
> patch, we can
> test if it solves the problem in our code.
>
> As a short term solution we have removed all the properties from
> decorated functions. This fixed one performance test. We moved to
> other performance tests and we still see that some functions are
> slower when executed from dlls, so the same slow down might be
> triggered by sth else.
>
> On Wed, Dec 17, 2008 at 3:21 AM, Dino Viehland <dinov at microsoft.com>
> wrote:
> > My guess would be that we're pushing the PythonContext._callSplatSite
> outside of the sweet spot in DLR site caches.
> >
> > You could check this by putting a breakpoint in
> PythonContext.Call(object, params object[]).  Then look and see if
> _callSplatSite._rules is an instance of EmptyRuleSet<T> with
> _supportAdding == false.
> >
> > If this is the cause it should be pretty easy to fix - property's
> should probably get their own call site object so it'll be specialized
> to the specific property (maybe we could share a common one of the
> function is a PythonFunction as these generate highly shareable calls
> between multiple function instances).  We're also doing a params call
> today which we could stop doing and get a nice perf boost on
> properties.
> >
> >> -----Original Message-----
> >> From: users-bounces at lists.ironpython.com [mailto:users-
> >> bounces at lists.ironpython.com] On Behalf Of Kamil Dworakowski
> >> Sent: Tuesday, December 16, 2008 11:08 AM
> >> To: users at lists.ironpython.com
> >> Subject: [IronPython] weird performance issue
> >>
> >> We see a very strange side effect of running the follwing code on
> the
> >> performance of recalculations in Resolver One.
> >>
> >> def Dumbcorator(fn):
> >>     def _inner(self, *args):
> >>         return fn(self, *args)
> >>     return _inner
> >>
> >> class Locker(object):
> >>     @property
> >>     @Dumbcorator
> >>     def thing(self):
> >>         return 2
> >>
> >> l = Locker()
> >> l.thing
> >>
> >> Specifically, one of the performance test reports results 4 to 8
> times
> >> worse with this snippet in than without (the clock is turned on
> after
> >> this snippet executes). If I comment out this snippet from the code,
> >> it suddenly runs 4 to 8 times faster. The fact that I can comment it
> >> out means that it is not used in the test, so it is bizzare that it
> >> slows it down.
> >>
> >> We can't reproduce the slow down in a clean example though. I hope
> >> something clicks for somebody and gives me a hint as to what is
> going
> >> on. Tomorrow I'll continue on trying to minimize the code that is
> >> being slowed down, and hopefully come up with some decent repro.
> >> _______________________________________________
> >> Users mailing list
> >> Users at lists.ironpython.com
> >> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
> > _______________________________________________
> > Users mailing list
> > Users at lists.ironpython.com
> > http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
> >
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com


From dinov at microsoft.com  Wed Dec 17 20:04:33 2008
From: dinov at microsoft.com (Dino Viehland)
Date: Wed, 17 Dec 2008 11:04:33 -0800
Subject: [IronPython] weird performance issue
In-Reply-To: <93fc29d60812171002r1ff25939v67fbae80be9dee7a@mail.gmail.com>
References: <93fc29d60812161107y79acc3dqbd99391e42f24eed@mail.gmail.com>
	<350E7D38B6D819428718949920EC2355564BF25378@NA-EXMSG-C102.redmond.corp.microsoft.com>
	<93fc29d60812171002r1ff25939v67fbae80be9dee7a@mail.gmail.com>
Message-ID: <350E7D38B6D819428718949920EC2355564BF25556@NA-EXMSG-C102.redmond.corp.microsoft.com>

Oh, if you also want to get a list of call stacks to PythonContext.Call and send those our way that'd be useful as well.

In general we only want to hit that when the overall cost of whatever we're doing will dwarf any overhead of that one call.  And we certainly have more calls to it then we should :(

> -----Original Message-----
> From: users-bounces at lists.ironpython.com [mailto:users-
> bounces at lists.ironpython.com] On Behalf Of Kamil Dworakowski
> Sent: Wednesday, December 17, 2008 10:03 AM
> To: Discussion of IronPython
> Subject: Re: [IronPython] weird performance issue
>
> Thanks Dino, it looks like you are right. I print whenever
> _supportAdding == false.
> It happens a lot with the snippet executed, whereas it doesn't happen
> at all with
> the snippet commented out.
>
> We also discovered that the slow down only occurs when we compile code
> to dlls (no _supportAdding == false if runned from sources).
> Still can't come up with a repro though. If you provide us with a
> patch, we can
> test if it solves the problem in our code.
>
> As a short term solution we have removed all the properties from
> decorated functions. This fixed one performance test. We moved to
> other performance tests and we still see that some functions are
> slower when executed from dlls, so the same slow down might be
> triggered by sth else.
>
> On Wed, Dec 17, 2008 at 3:21 AM, Dino Viehland <dinov at microsoft.com>
> wrote:
> > My guess would be that we're pushing the PythonContext._callSplatSite
> outside of the sweet spot in DLR site caches.
> >
> > You could check this by putting a breakpoint in
> PythonContext.Call(object, params object[]).  Then look and see if
> _callSplatSite._rules is an instance of EmptyRuleSet<T> with
> _supportAdding == false.
> >
> > If this is the cause it should be pretty easy to fix - property's
> should probably get their own call site object so it'll be specialized
> to the specific property (maybe we could share a common one of the
> function is a PythonFunction as these generate highly shareable calls
> between multiple function instances).  We're also doing a params call
> today which we could stop doing and get a nice perf boost on
> properties.
> >
> >> -----Original Message-----
> >> From: users-bounces at lists.ironpython.com [mailto:users-
> >> bounces at lists.ironpython.com] On Behalf Of Kamil Dworakowski
> >> Sent: Tuesday, December 16, 2008 11:08 AM
> >> To: users at lists.ironpython.com
> >> Subject: [IronPython] weird performance issue
> >>
> >> We see a very strange side effect of running the follwing code on
> the
> >> performance of recalculations in Resolver One.
> >>
> >> def Dumbcorator(fn):
> >>     def _inner(self, *args):
> >>         return fn(self, *args)
> >>     return _inner
> >>
> >> class Locker(object):
> >>     @property
> >>     @Dumbcorator
> >>     def thing(self):
> >>         return 2
> >>
> >> l = Locker()
> >> l.thing
> >>
> >> Specifically, one of the performance test reports results 4 to 8
> times
> >> worse with this snippet in than without (the clock is turned on
> after
> >> this snippet executes). If I comment out this snippet from the code,
> >> it suddenly runs 4 to 8 times faster. The fact that I can comment it
> >> out means that it is not used in the test, so it is bizzare that it
> >> slows it down.
> >>
> >> We can't reproduce the slow down in a clean example though. I hope
> >> something clicks for somebody and gives me a hint as to what is
> going
> >> on. Tomorrow I'll continue on trying to minimize the code that is
> >> being slowed down, and hopefully come up with some decent repro.
> >> _______________________________________________
> >> Users mailing list
> >> Users at lists.ironpython.com
> >> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
> > _______________________________________________
> > Users mailing list
> > Users at lists.ironpython.com
> > http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
> >
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com


From empirebuilder at gmail.com  Wed Dec 17 20:59:33 2008
From: empirebuilder at gmail.com (Dody Gunawinata)
Date: Wed, 17 Dec 2008 21:59:33 +0200
Subject: [IronPython] ASP.Net IronPython Refresh
Message-ID: <8cd017b80812171159l55653792ve4ba9fb5754f24eb@mail.gmail.com>

Does anybody know when the IronPython for ASP.Net gets updated for the final
version of IronPython 2.0?

-- 
nomadlife.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081217/63fc410a/attachment.html>

From jslutter at reactorzero.com  Wed Dec 17 21:19:41 2008
From: jslutter at reactorzero.com (Jeff Slutter)
Date: Wed, 17 Dec 2008 15:19:41 -0500
Subject: [IronPython] Loading IronPython.dll from memory
Message-ID: <49495EDD.1010408@reactorzero.com>

[I can work around this, but I'm curious if there is a solution]

Due to some crazy requirements, I'm trying to load my assemblies from
memory. I have setup an Assembly Resolver with:

AppDomain.CurrentDomain.AssemblyResolve += new
ResolveEventHandler(ResolveInternalAssembly);

And it gets called. Inside the ResolveInternalAssembly I do the following:

1) check my Dictionary to see if I have already loaded the assembly, if
so I return it, otherwise...
2) I, currently, open the file, read in the file to byte[] and then call
3) AppDomain.CurrentDomain.Load(asmbytes) to load the Assembly, store it
to the dictionary and return it.

(see: http://weblogs.asp.net/justin_rogers/archive/2004/06/07/149964.aspx)

This works well and all until I get to the following code within one of
my loaded assemblies:
m_engine = m_runtime.GetEngine("Python");

It throws an exception saying:
Microsoft.Scripting.InvalidImplementationException
Type 'IronPython.Runtime.PythonContext' doesn't provide a suitable
public constructor or its implementation is faulty: The type initializer
for 'IronPython.Runtime.SysModule' threw an exception.


I assume this is some sort of security/permissions problem because that
is the main thing that is different when loading an assembly from memory
than file.

My work around is to pre-load the IronPython.dll assembly before hooking
up my AssemblyResolve (its the only one I have to 'preload')
dynamicAssemblyHash["IronPython"] =
Assembly.LoadFile(System.IO.Path.Combine(assemblyLocation,
"IronPython.dll"));

Any ideas?
-Jeff





From fuzzyman at voidspace.org.uk  Wed Dec 17 22:37:37 2008
From: fuzzyman at voidspace.org.uk (Michael Foord)
Date: Wed, 17 Dec 2008 21:37:37 +0000
Subject: [IronPython] Loading IronPython.dll from memory
In-Reply-To: <49495EDD.1010408@reactorzero.com>
References: <49495EDD.1010408@reactorzero.com>
Message-ID: <49497121.9040100@voidspace.org.uk>

Jeff Slutter wrote:
> [I can work around this, but I'm curious if there is a solution]
>
> Due to some crazy requirements, I'm trying to load my assemblies from
> memory. I have setup an Assembly Resolver with:
>
> AppDomain.CurrentDomain.AssemblyResolve += new
> ResolveEventHandler(ResolveInternalAssembly);
>
> And it gets called. Inside the ResolveInternalAssembly I do the following:
>
> 1) check my Dictionary to see if I have already loaded the assembly, if
> so I return it, otherwise...
> 2) I, currently, open the file, read in the file to byte[] and then call
> 3) AppDomain.CurrentDomain.Load(asmbytes) to load the Assembly, store it
> to the dictionary and return it.
>
> (see: http://weblogs.asp.net/justin_rogers/archive/2004/06/07/149964.aspx)
>
> This works well and all until I get to the following code within one of
> my loaded assemblies:
> m_engine = m_runtime.GetEngine("Python");
>   

Is this using the IronPython hosting API? That doesn't look like code 
that would have worked with any recent version of IronPython.

Try:

engine = Python.CreateEngine();

Michael



> It throws an exception saying:
> Microsoft.Scripting.InvalidImplementationException
> Type 'IronPython.Runtime.PythonContext' doesn't provide a suitable
> public constructor or its implementation is faulty: The type initializer
> for 'IronPython.Runtime.SysModule' threw an exception.
>
>
> I assume this is some sort of security/permissions problem because that
> is the main thing that is different when loading an assembly from memory
> than file.
>
> My work around is to pre-load the IronPython.dll assembly before hooking
> up my AssemblyResolve (its the only one I have to 'preload')
> dynamicAssemblyHash["IronPython"] =
> Assembly.LoadFile(System.IO.Path.Combine(assemblyLocation,
> "IronPython.dll"));
>
> Any ideas?
> -Jeff
>
>
>
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>   


-- 
http://www.ironpythoninaction.com/
http://www.voidspace.org.uk/blog




From jslutter at reactorzero.com  Thu Dec 18 03:17:10 2008
From: jslutter at reactorzero.com (Jeff Slutter)
Date: Wed, 17 Dec 2008 21:17:10 -0500
Subject: [IronPython] Loading IronPython.dll from memory
In-Reply-To: <49497121.9040100@voidspace.org.uk>
References: <49495EDD.1010408@reactorzero.com>
	<49497121.9040100@voidspace.org.uk>
Message-ID: <4949B2A6.4050400@reactorzero.com>

I snipped a little extra - my code is really:

Dictionary<string, object> runtimeOptions = new Dictionary<string,
object>();
runtimeOptions["Debug"] = true;
m_runtime = IronPython.Hosting.Python.CreateRuntime(runtimeOptions);
m_runtime.LoadAssembly(typeof(String).Assembly); // Add reference to System
m_runtime.LoadAssembly(typeof(Uri).Assembly); // Add reference to mscorlib
m_runtime.LoadAssembly(System.Reflection.Assembly.GetCallingAssembly());
m_engine = m_runtime.GetEngine("Python");

I've also tried replacing the above with:
m_engine = Python.CreateEngine();

and I get the same exception in the call to Python.CreateEngine()

(Note: the exception exists without all the extra LoadAssemblys and
stripped to the bone).



Michael Foord wrote:
> Jeff Slutter wrote:
>> [I can work around this, but I'm curious if there is a solution]
>>
>> Due to some crazy requirements, I'm trying to load my assemblies from
>> memory. I have setup an Assembly Resolver with:
>>
>> AppDomain.CurrentDomain.AssemblyResolve += new
>> ResolveEventHandler(ResolveInternalAssembly);
>>
>> And it gets called. Inside the ResolveInternalAssembly I do the
>> following:
>>
>> 1) check my Dictionary to see if I have already loaded the assembly, if
>> so I return it, otherwise...
>> 2) I, currently, open the file, read in the file to byte[] and then call
>> 3) AppDomain.CurrentDomain.Load(asmbytes) to load the Assembly, store it
>> to the dictionary and return it.
>>
>> (see:
>> http://weblogs.asp.net/justin_rogers/archive/2004/06/07/149964.aspx)
>>
>> This works well and all until I get to the following code within one of
>> my loaded assemblies:
>> m_engine = m_runtime.GetEngine("Python");
>>   
> 
> Is this using the IronPython hosting API? That doesn't look like code
> that would have worked with any recent version of IronPython.
> 
> Try:
> 
> engine = Python.CreateEngine();
> 
> Michael
> 
> 
> 
>> It throws an exception saying:
>> Microsoft.Scripting.InvalidImplementationException
>> Type 'IronPython.Runtime.PythonContext' doesn't provide a suitable
>> public constructor or its implementation is faulty: The type initializer
>> for 'IronPython.Runtime.SysModule' threw an exception.
>>
>>
>> I assume this is some sort of security/permissions problem because that
>> is the main thing that is different when loading an assembly from memory
>> than file.
>>
>> My work around is to pre-load the IronPython.dll assembly before hooking
>> up my AssemblyResolve (its the only one I have to 'preload')
>> dynamicAssemblyHash["IronPython"] =
>> Assembly.LoadFile(System.IO.Path.Combine(assemblyLocation,
>> "IronPython.dll"));
>>
>> Any ideas?
>> -Jeff
>>
>>
>>
>> _______________________________________________
>> Users mailing list
>> Users at lists.ironpython.com
>> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>>   
> 
> 



From dinov at microsoft.com  Thu Dec 18 03:17:18 2008
From: dinov at microsoft.com (Dino Viehland)
Date: Wed, 17 Dec 2008 18:17:18 -0800
Subject: [IronPython] test...
Message-ID: <350E7D38B6D819428718949920EC2355564BF257C2@NA-EXMSG-C102.redmond.corp.microsoft.com>

This is just a test... there were some issues with the mailing list today, just verifying that it should be back now.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081217/9e71ebb7/attachment.html>

From dinov at microsoft.com  Thu Dec 18 05:15:08 2008
From: dinov at microsoft.com (Dino Viehland)
Date: Wed, 17 Dec 2008 20:15:08 -0800
Subject: [IronPython] Issue about string.upper and string.lower
In-Reply-To: <350E7D38B6D819428718949920EC2355564B62EE33@NA-EXMSG-C102.redmond.corp.microsoft.com>
References: <2679f6510812120727o58a85ec4ub686aeac4dc4820b@mail.gmail.com>
	<350E7D38B6D819428718949920EC2355564B62EE33@NA-EXMSG-C102.redmond.corp.microsoft.com>
Message-ID: <350E7D38B6D819428718949920EC2355564BF257E1@NA-EXMSG-C102.redmond.corp.microsoft.com>

There were no issues with making this change revealed by our tests.  The change is in our 2.1 branch internally so should so up externally really soon.  I suspect we'll discuss backporting it to 2.0.1 at one of our team meetings but unless we hear about some issues this causes then I personally would be pro-backporting.

From: users-bounces at lists.ironpython.com [mailto:users-bounces at lists.ironpython.com] On Behalf Of Dino Viehland
Sent: Monday, December 15, 2008 11:13 PM
To: glenn.k.jones+ipy at gmail.com; Discussion of IronPython
Subject: Re: [IronPython] Issue about string.upper and string.lower

I've actually looked at this not too long ago and I think your proposal of calling the Invariant functions is the correct solution.  I was looking at a few things:  This bug http://bugs.python.org/issue1528802, the 3.0 decimal.py module, and also just using Turkish I at the command prompt.  If you follow the comments the bug says:


"String upper and lower conversion are locale dependent and
implemented by the underlying libc, whereas Unicode
upper/lower conversion is not and only depends on the
Unicode character database."


That's a pretty clear statement that we shouldn't be using the current locale for our upper/lower string conversions.  It wouldn't surprise me if that breaks something somewhere because we won't be doing locale dependent conversions on what someone expects to its type to be str not unicode but in this case I think it'd be better to be consistent with the Unicode side of Python as that's our future.

As for the decimal module it doesn't change from 2.x to 3.0.  So .upper() apparently doesn't have this problem when CPython switches to Unicode strings.  Or at least no one's hit it, and when they do I think the resolution would be the same as 1528802.

Finally at the command prompt I could never get CPython to do a culture-sensitive operation.  I hadn't fully convinced myself on that part though because I hadn't yet escalated to a Turkish install of the OS running IronPython.

But I'm still pretty confident we're at fault and we should change our lower/upper implementation.  Obviously the change is easy but I'll do a full test pass to see if it breaks anything.

I also think calling ToUpper to get non-Pythonic results is easy enough (I actually think it's kind of better this way - it saves typing out the framework friendly ToUpperInvariant :)).

From: users-bounces at lists.ironpython.com [mailto:users-bounces at lists.ironpython.com] On Behalf Of Glenn Jones
Sent: Friday, December 12, 2008 7:27 AM
To: Discussion of IronPython
Subject: [IronPython] Issue about string.upper and string.lower

Hello everybody,

We ran across http://www.codeplex.com/IronPython/WorkItem/View.aspx?WorkItemId=13629 (turkish collation issues) today while trying to port Resolver One to IronPython 2.0.

This is in a test that tries to import decimal while in the turkish locale (it was actually reported by a user!). It does an .upper on a string with an 'i', and that doesn't give the expected results.

This is a very big issue because all the python code out there expects string transformations to be locale-independent, and there may be strange bugs in strange places.

Is mapping .upper and .lower to ToUpperInvariant and ToLowerInvariant an acceptable solution? People that want to do locale-dependent transformations can always use the .NET specific ToUpper/ToLower.

We can work around the decimal being unimportable by hacking it, but clearly this is not a general solution. We will report other modules that might fail from this as we find them.

Thanks,
Glenn and Orestis
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081217/b090cf4b/attachment.html>

From tom.wright at resolversystems.com  Thu Dec 18 13:00:10 2008
From: tom.wright at resolversystems.com (Tom Wright)
Date: Thu, 18 Dec 2008 12:00:10 +0000
Subject: [IronPython] IronPython and file descriptors
In-Reply-To: <494924ED.30604@resolversystems.com>
References: <49467FDA.5000101@resolversystems.com>	<d2155e360812150928i6e18806k309ce32ab98fde4e@mail.gmail.com>	<494695C0.9040401@resolversystems.com>	<d2155e360812150947vccb612ey9d32711bef718395@mail.gmail.com>	<49469F86.8040901@resolversystems.com>	<a623f7d10812151021u7e6dbba3k3b0971f4d42295f3@mail.gmail.com>	<350E7D38B6D819428718949920EC2355564B62EAC4@NA-EXMSG-C102.redmond.corp.microsoft.com>	<6f4025010812151252p2bb355d7p1b9d915c0b033016@mail.gmail.com>	<350E7D38B6D819428718949920EC2355564B62EC42@NA-EXMSG-C102.redmond.corp.microsoft.com>	<4946D288.5020909@voidspace.org.uk>	<49478AD4.1080002@resolversystems.com>
	<494924ED.30604@resolversystems.com>
Message-ID: <494A3B4A.4030707@resolversystems.com>

Before playing with this, it's probably worth knowing that the following 
can cause exceptions or no exceptions and lost output from the final write.

1 Writing to a PythonFile
2 Getting the PythonFile's stream's file handle
3 Converting the handle to a file descriptor
4 Writing to this file descriptor directly in C using write ---- Fine up 
to this point
5 Writing to the PythonFile again

Since code that does steps 1-4 is likely to also perform 5 it may not be 
worthwhile having file.fileno() return a valid file descriptor unless 
step 5 can be made to work. Though I appreciate that this is problematic 
and that there are good reasons for having PythonFiles wrap FileStreams.

Tom


William Reade wrote:
> Just a quick note: you can't necessarily just P/invoke out to msvcrt, 
> because CPython was built with msvcr71. Sorry :(.
>
> Tom Wright wrote:
>> I'm not sure about sockets. The only example I have found in the wild 
>> so far is with files in PIL.
>>
>> The python implementation seems to have socket.fileno() returning a 
>> file descriptor which comes from <sys/socket.h>'s socketpair function 
>> in the standard C library (see socketmodule.c if you are interested)
>>
>> So, again, it would be possible for people to pass this identifier as 
>> an integer directly to C. Whether people do this in practice is a 
>> different question...
>>
>> Tom
>>
>>
>>
>> Michael Foord wrote:
>>> Dino Viehland wrote:
>>>>
>>>> (now I?ve replied to the correct thread?)
>>>>
>>>> Can you open a feature request on CodePlex? It's certainly an 
>>>> interesting idea to ponder and I'm leaning towards it but there's 
>>>> lots of details to be gotten right.
>>>>
>>>> Do you know if this needs to work w/ sockets as well? (There's also 
>>>> the question of can we make it work with sockets? :))
>>>>
>>>> There'll be a bunch of places we need to update (nt, socket, file, 
>>>> select, etc...) so I think it'll have to wait until 2.1 instead of 
>>>> coming in a minor update like 2.0.1.
>>>>
>>>
>>> Tom will have to comment on whether it is needed for sockets as 
>>> well. Not yet I guess. :-)
>>>
>>> Tom created an issue on Codeplex - I've added a comment to that:
>>>
>>> http://www.codeplex.com/IronPython/WorkItem/View.aspx?WorkItemId=20242
>>>
>>> Michael
>>>>
>>>> *From:* users-bounces at lists.ironpython.com 
>>>> [mailto:users-bounces at lists.ironpython.com] *On Behalf Of *Michael 
>>>> Foord
>>>> *Sent:* Monday, December 15, 2008 12:53 PM
>>>> *To:* Discussion of IronPython
>>>> *Subject:* Re: [IronPython] IronPython and file descriptors
>>>>
>>>> 2008/12/15 Dino Viehland <dinov at microsoft.com 
>>>> <mailto:dinov at microsoft.com>>
>>>>
>>>> Presumably CPython is linking against msvcrt and is using the C 
>>>> abstraction rather than calling the Win32 APIs which return handles 
>>>> directly.
>>>>
>>>> As Curt said the biggest problem for us is that .NET does not 
>>>> expose C file descriptors. Therefore we could fix this by 
>>>> P/Invoking out to msvcrt. For users on other platforms we'd need to 
>>>> either add support for their platform or hope that their handles == 
>>>> C runtime file descriptors. For something like fileno maybe this is 
>>>> ok because it's an interop point only - or are there any 
>>>> non-interop uses of fileno in the world?
>>>>
>>>> What do people think of this? This would be the 1st place where we 
>>>> would add a P/Invoke so I'd want to tread lightly and make sure 
>>>> this is really the right thing to do.
>>>>
>>>>
>>>>
>>>> Well, some way of doing this is needed if certain Python C 
>>>> extensions are to work with Ironclad. Perhaps a global or per 
>>>> engine setting that allows file handles to be associated with a C 
>>>> descriptor. Users of Ironclad could switch it on if they wished and 
>>>> take the consequences.
>>>>
>>>> Michael Foord
>>>>
>>>>
>>>> -----Original Message-----
>>>> From: users-bounces at lists.ironpython.com
>>>> <mailto:users-bounces at lists.ironpython.com>
>>>> [mailto:users-bounces at lists.ironpython.com
>>>> <mailto:users-bounces at lists.ironpython.com>] On Behalf Of Slide
>>>> Sent: Monday, December 15, 2008 10:21 AM
>>>> To: Discussion of IronPython
>>>> Subject: Re: [IronPython] IronPython and file descriptors
>>>>
>>>> On Mon, Dec 15, 2008 at 11:18 AM, Tom Wright
>>>> <tom.wright at resolversystems.com
>>>> <mailto:tom.wright at resolversystems.com>> wrote:
>>>> > Agreed. It is certainly possible with some work to get a file
>>>> descriptor and
>>>> > pass it to C code.
>>>> >
>>>> > This is not the problem, however. Ironclad's aim (eventually) is
>>>> to allow
>>>> > *arbitrary* C extensions to work with Ironpython without
>>>> changing the
>>>> > extensions. Ctypes aim is to allow arbitrary C code to be run by
>>>> python
>>>> > (possibly in other python libraries which you really don't want
>>>> to modify).
>>>> >
>>>> > In CPython .fileno() is a file descriptor and some modules use
>>>> this fact
>>>> > (specifically PIL) by passing file descriptors as integers into
>>>> C code. I do
>>>> > not know how one can make this code work in IronPython unless
>>>> .fileno()
>>>> > returns a file descriptor.
>>>> >
>>>> > The game here is not making a particular C library work with
>>>> IronPython by
>>>> > modifying this library but making as many libraries as possible
>>>> work without
>>>> > modification. Of course whether this is worthwhile depends on
>>>> how many
>>>> > libraries rely on this fact.
>>>> >
>>>> > Sorry - I hope this is a little clearer.
>>>> >
>>>> > Tom
>>>>
>>>>
>>>> How does CPython get the file descriptor for fileno()? Does it just
>>>> return the HANDLE on Windows?
>>>>
>>>> slide
>>>> _______________________________________________
>>>> Users mailing list
>>>> Users at lists.ironpython.com <mailto:Users at lists.ironpython.com>
>>>> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>>>> _______________________________________________
>>>> Users mailing list
>>>> Users at lists.ironpython.com <mailto:Users at lists.ironpython.com>
>>>> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>>>>
>>>>
>>>>
>>>>
>>>> -- 
>>>> http://www.ironpythoninaction.com/
>>>>
>>>> ------------------------------------------------------------------------ 
>>>>
>>>>
>>>> _______________________________________________
>>>> Users mailing list
>>>> Users at lists.ironpython.com
>>>> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>>>
>>>
>>
>> _______________________________________________
>> Users mailing list
>> Users at lists.ironpython.com
>> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>>
>
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com



From yashgt at gmail.com  Thu Dec 18 15:57:54 2008
From: yashgt at gmail.com (Yash Ganthe)
Date: Thu, 18 Dec 2008 20:27:54 +0530
Subject: [IronPython] Descriptive reports in unittest module
Message-ID: <4f428e7b0812180657v50f7523flb54365b19c7766a4@mail.gmail.com>

I am using the unittest module in IronPython 2.0.

My TestClass has a function named testLoginWithInvalidUsername. I would like
to attach a string to this function so that if the test fails, the string
would be printed as:
Failed : Testing the login functionality with invalid user name
I am not asking for something that will automatically figure out the string
to be printed. I am looking for a way for me to attach an attribute to the
test function so that the string in the attribute gets printed.

Is there something readily avaliable for doing this?

Currently my test report comes out as:
testDateDiffWeek (PI.GetMessages.GetMessagesTestClass) ... ok
testGetMessageGUIDnull (PI.GetMessages.GetMessagesTestClass) ... FAIL

Thanks,
Yash
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081218/1a3d9f22/attachment.html>

From fuzzyman at gmail.com  Thu Dec 18 16:09:52 2008
From: fuzzyman at gmail.com (Michael Foord)
Date: Thu, 18 Dec 2008 15:09:52 +0000
Subject: [IronPython] Descriptive reports in unittest module
In-Reply-To: <4f428e7b0812180657v50f7523flb54365b19c7766a4@mail.gmail.com>
References: <4f428e7b0812180657v50f7523flb54365b19c7766a4@mail.gmail.com>
Message-ID: <6f4025010812180709qb99239dycbe2a2061aa65a26@mail.gmail.com>

2008/12/18 Yash Ganthe <yashgt at gmail.com>

> I am using the unittest module in IronPython 2.0.
>
> My TestClass has a function named testLoginWithInvalidUsername. I would
> like to attach a string to this function so that if the test fails, the
> string would be printed as:
> Failed : Testing the login functionality with invalid user name
> I am not asking for something that will automatically figure out the string
> to be printed. I am looking for a way for me to attach an attribute to the
> test function so that the string in the attribute gets printed.
>
> Is there something readily avaliable for doing this?



Most unittest assert methods take an optional message parameter - where the
message is the failure message. e.g.

self.assertEquals('foo', 'bar', 'foo is not equal to bar')

Michael



>
>
> Currently my test report comes out as:
> testDateDiffWeek (PI.GetMessages.GetMessagesTestClass) ... ok
> testGetMessageGUIDnull (PI.GetMessages.GetMessagesTestClass) ... FAIL
>
> Thanks,
> Yash
>
>
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>
>


-- 
http://www.ironpythoninaction.com/
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081218/c2ae106f/attachment.html>

From michael.foord at resolversystems.com  Thu Dec 18 18:27:49 2008
From: michael.foord at resolversystems.com (Michael Foord)
Date: Thu, 18 Dec 2008 17:27:49 +0000
Subject: [IronPython] The Resolver One Spreadsheet Challenge
Message-ID: <494A8815.6090504@resolversystems.com>

Resolver One is the Python powered spreadsheet created by Resolver Systems.

Resolver One is a highly programmable spreadsheet program built with 
IronPython. It is capable of creating powerful spreadsheet systems, but 
is easy to program with Python and .NET libraries.

We?re convinced that Resolver One allows people to create astonishing 
things that simply aren?t possible in a traditional spreadsheet 
environment. And we want to prove it. Enter the Resolver One Spreadsheet 
Challenge:

http://www.resolversystems.com/competition/

The Resolver One Challenge

We're so confident about the revolutionary potential of Resolver One 
that we've set up the $25,000 Resolver One Challenge. Every month 
between now and May, we will be giving away $2,000 for the best 
spreadsheet we receive. And in late May, we'll be handing over $15,000 
for the best of the best.
Let your imagination run wild

Build a blogging engine directly in Resolver One. Hook Resolver One up 
to existing .NET or Python libraries in unusual ways. Build the game of 
life, or a Mandelbrot viewer directly into the grid. Get Infocom 
adventure games running inside a spreadsheet; or for that matter, have a 
conversation with Eliza. Make a music player that does visualisations in 
the cells.
Or something more businesslike?

Use the sophisticated web integration to pull of stock prices, or 
integrate your spreadsheet with Google Maps. (Perhaps you could build a 
spreadsheet that plots a map, showing in which part of the country stock 
or house prices are rising or falling the most.) Build an election 
predictor (and use a combination of Monte Carlo analysis and the web 
front end to make it really special).

In other words: Resolver One gives you the tools, you just need to use 
your imagination, and your Python and spreadsheet skills!

Resolver One is free to try and for non-commercial and Open Source uses. 
You can download it from:

http://www.resolversystems.com/download/

To get you started with Resolver One we have a new tutorial. It takes 
you through all the major features, with examples to try:

http://www.resolversystems.com/documentation/index.php/Tutorial.html

Michael Foord
Software Developer
Resolver Systems

-- 

Michael Foord
Senior Software Engineer, Resolver Systems Ltd.
michael.foord at resolversystems.com
+44 (0) 20 7253 6372

Try out Resolver One! <http://www.resolversystems.com/get-it/>

17a Clerkenwell Road, London EC1M 5RD, UK
VAT No.: GB 893 5643 79 Registered in England and Wales as company number 5467329.
Registered address: 843 Finchley Road, London NW11 8NA, UK 



From kristian.jaksch at gmail.com  Fri Dec 19 10:38:04 2008
From: kristian.jaksch at gmail.com (xkrja)
Date: Fri, 19 Dec 2008 01:38:04 -0800 (PST)
Subject: [IronPython]  Online IronPython interpreter and DLRconsole.
Message-ID: <21088369.post@talk.nabble.com>


I've seen Michael Foord's "Interactive Interpreter" made in Silverlight 2 and
the DLRconsole (I assume made by Jimmy Schementi) and I like them a lot.

But my question is: What controls have been used to mimic the console
"feeling"? I've tried to read the source code but can't find how it's built
up. I would like something similar but can't figure out how to do it. 

Of course I can go with a textblock and textbox but that wouldn't give the
real feeling.

Thanks for any help!
-- 
View this message in context: http://www.nabble.com/Online-IronPython-interpreter-and-DLRconsole.-tp21088369p21088369.html
Sent from the IronPython mailing list archive at Nabble.com.



From fuzzyman at voidspace.org.uk  Fri Dec 19 13:10:45 2008
From: fuzzyman at voidspace.org.uk (Michael Foord)
Date: Fri, 19 Dec 2008 12:10:45 +0000
Subject: [IronPython] Online IronPython interpreter and DLRconsole.
In-Reply-To: <21088369.post@talk.nabble.com>
References: <21088369.post@talk.nabble.com>
Message-ID: <494B8F45.1070900@voidspace.org.uk>

xkrja wrote:
> I've seen Michael Foord's "Interactive Interpreter" made in Silverlight 2 and
> the DLRconsole (I assume made by Jimmy Schementi) and I like them a lot.
>
> But my question is: What controls have been used to mimic the console
> "feeling"? I've tried to read the source code but can't find how it's built
> up. I would like something similar but can't figure out how to do it. 
>
> Of course I can go with a textblock and textbox but that wouldn't give the
> real feeling.
>
> Thanks for any help!
>   
My project ('Python in the Browser') doesn't use any controls. It works 
with an HTML textarea and Javascript. When I get any time to work on it 
I will probably change this - but still use HTML components rather than 
Silverlight controls.

You might like to also check out Dan Eloff's very slick Silvershell:

http://code.google.com/p/silvershell/

Michael Foord

-- 
http://www.ironpythoninaction.com/



From Jimmy.Schementi at microsoft.com  Sat Dec 20 01:43:38 2008
From: Jimmy.Schementi at microsoft.com (Jimmy Schementi)
Date: Fri, 19 Dec 2008 16:43:38 -0800
Subject: [IronPython] ASP.Net IronPython Refresh
In-Reply-To: <8cd017b80812171159l55653792ve4ba9fb5754f24eb@mail.gmail.com>
References: <8cd017b80812171159l55653792ve4ba9fb5754f24eb@mail.gmail.com>
Message-ID: <5283CA0A4168DF4FBBD71AE9ECA5A3284BDEA69D14@NA-EXMSG-C116.redmond.corp.microsoft.com>

Hopefully getting a build out soon ... I'll keep this list posted.

From: users-bounces at lists.ironpython.com [mailto:users-bounces at lists.ironpython.com] On Behalf Of Dody Gunawinata
Sent: Wednesday, December 17, 2008 12:00 PM
To: Discussion of IronPython
Subject: [IronPython] ASP.Net IronPython Refresh

Does anybody know when the IronPython for ASP.Net gets updated for the final version of IronPython 2.0?

--
nomadlife.org<http://nomadlife.org>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081219/92eb18f1/attachment.html>

From Jimmy.Schementi at microsoft.com  Sat Dec 20 01:56:04 2008
From: Jimmy.Schementi at microsoft.com (Jimmy Schementi)
Date: Fri, 19 Dec 2008 16:56:04 -0800
Subject: [IronPython] Online IronPython interpreter and DLRconsole.
In-Reply-To: <494B8F45.1070900@voidspace.org.uk>
References: <21088369.post@talk.nabble.com> <494B8F45.1070900@voidspace.org.uk>
Message-ID: <5283CA0A4168DF4FBBD71AE9ECA5A3284BDEA69D1C@NA-EXMSG-C116.redmond.corp.microsoft.com>

DLRConsole isn't really reusable today, which is why I've been rewriting it in C# and adding it to Microsoft.Scripting.Silverlight: (http://github.com/jschementi/agdlr). It can be enabled by putting console=true in initParams (http://blog.jimmy.schementi.com/2008/11/repls-in-silverlight.html). It's a HTML based console, with a <input type="text" ... /> for input. I tried a <textarea> for a while and it just complicated things. I stayed away from using Silverlight controls and such, because it wouldn't work for a HTML-only Silverlight app. HTML is just fine for this type of thing, but it would be nice to have a Silverlight UI implementation. Maybe combine my work and SilverShell? =)

(P.S. Jim Hugunin wrote the initial version of DLRConsole ... I've only kept it working =))

Michael Foord wrote:
> 
> xkrja wrote:
> > I've seen Michael Foord's "Interactive Interpreter" made in
> Silverlight 2 and
> > the DLRconsole (I assume made by Jimmy Schementi) and I like them a
> lot.
> >
> > But my question is: What controls have been used to mimic the console
> > "feeling"? I've tried to read the source code but can't find how it's
> built
> > up. I would like something similar but can't figure out how to do it.
> >
> > Of course I can go with a textblock and textbox but that wouldn't
> give the
> > real feeling.
> >
> > Thanks for any help!
> >
> My project ('Python in the Browser') doesn't use any controls. It works
> with an HTML textarea and Javascript. When I get any time to work on it
> I will probably change this - but still use HTML components rather than
> Silverlight controls.
> 
> You might like to also check out Dan Eloff's very slick Silvershell:
> 
> http://code.google.com/p/silvershell/
> 
> Michael Foord


From kristian.jaksch at gmail.com  Sat Dec 20 11:09:44 2008
From: kristian.jaksch at gmail.com (Kristian Jaksch)
Date: Sat, 20 Dec 2008 11:09:44 +0100
Subject: [IronPython] Online IronPython interpreter and DLRconsole.
In-Reply-To: <5283CA0A4168DF4FBBD71AE9ECA5A3284BDEA69D1C@NA-EXMSG-C116.redmond.corp.microsoft.com>
References: <21088369.post@talk.nabble.com> <494B8F45.1070900@voidspace.org.uk>
	<5283CA0A4168DF4FBBD71AE9ECA5A3284BDEA69D1C@NA-EXMSG-C116.redmond.corp.microsoft.com>
Message-ID: <1dab55500812200209y1af333e9tf907ad238ba19c6f@mail.gmail.com>

Thanks for the replies.

Ok, so it's HTML and JavaScript only. Hmm... I think I try to cheat in some
way to get a similar feeling but with a couple of Silverlight controls.

2008/12/20 Jimmy Schementi <Jimmy.Schementi at microsoft.com>

> DLRConsole isn't really reusable today, which is why I've been rewriting it
> in C# and adding it to Microsoft.Scripting.Silverlight: (
> http://github.com/jschementi/agdlr). It can be enabled by putting
> console=true in initParams (
> http://blog.jimmy.schementi.com/2008/11/repls-in-silverlight.html). It's a
> HTML based console, with a <input type="text" ... /> for input. I tried a
> <textarea> for a while and it just complicated things. I stayed away from
> using Silverlight controls and such, because it wouldn't work for a
> HTML-only Silverlight app. HTML is just fine for this type of thing, but it
> would be nice to have a Silverlight UI implementation. Maybe combine my work
> and SilverShell? =)
>
> (P.S. Jim Hugunin wrote the initial version of DLRConsole ... I've only
> kept it working =))
>
> Michael Foord wrote:
> >
> > xkrja wrote:
> > > I've seen Michael Foord's "Interactive Interpreter" made in
> > Silverlight 2 and
> > > the DLRconsole (I assume made by Jimmy Schementi) and I like them a
> > lot.
> > >
> > > But my question is: What controls have been used to mimic the console
> > > "feeling"? I've tried to read the source code but can't find how it's
> > built
> > > up. I would like something similar but can't figure out how to do it.
> > >
> > > Of course I can go with a textblock and textbox but that wouldn't
> > give the
> > > real feeling.
> > >
> > > Thanks for any help!
> > >
> > My project ('Python in the Browser') doesn't use any controls. It works
> > with an HTML textarea and Javascript. When I get any time to work on it
> > I will probably change this - but still use HTML components rather than
> > Silverlight controls.
> >
> > You might like to also check out Dan Eloff's very slick Silvershell:
> >
> > http://code.google.com/p/silvershell/
> >
> > Michael Foord
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081220/7c33a3c1/attachment.html>

From daftspaniel at gmail.com  Sat Dec 20 12:22:42 2008
From: daftspaniel at gmail.com (Davy Mitchell)
Date: Sat, 20 Dec 2008 11:22:42 +0000
Subject: [IronPython] ANN : Davy's IronPython Editor
Message-ID: <20253b0c0812200322w4734bf90u6947528ce97e92c8@mail.gmail.com>

Hi Folks,
I've made a small start on an IronPython Editor in WinForms using
SyntaxBox.  The project intentions is to create an editor along the lines of
SPE or DrPython.

If you want to take a look there are a couple of screen shots - please
visit:
http://code.google.com/p/davysironpythoneditor/

Code in SVN only at the moment. DIE does benefit from being compiled. Maybe
once I stop sniffing and coughing so much it will get done and uploaded !
:-)

It's not production ready yet so please, please be careful with your
precious source code. I've not tried it on Mono yet.

Feedback and ideas appreciated.

Thanks & Merry Christmas,
Davy Mitchell
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081220/f99576d8/attachment.html>

From kamil at dworakowski.name  Sat Dec 20 15:41:05 2008
From: kamil at dworakowski.name (Kamil Dworakowski)
Date: Sat, 20 Dec 2008 14:41:05 +0000
Subject: [IronPython] Descriptive reports in unittest module
In-Reply-To: <4f428e7b0812180657v50f7523flb54365b19c7766a4@mail.gmail.com>
References: <4f428e7b0812180657v50f7523flb54365b19c7766a4@mail.gmail.com>
Message-ID: <93fc29d60812200641t21b89d0dl72049cbb01cdab14@mail.gmail.com>

Do you just want better test names? Maybe this will help:

http://blog.kamil.dworakowski.name/2008/04/camelcasesplit-t-i-n-g.html

On Thu, Dec 18, 2008 at 2:57 PM, Yash Ganthe <yashgt at gmail.com> wrote:
> I am using the unittest module in IronPython 2.0.
>
> My TestClass has a function named testLoginWithInvalidUsername. I would like
> to attach a string to this function so that if the test fails, the string
> would be printed as:
> Failed : Testing the login functionality with invalid user name
> I am not asking for something that will automatically figure out the string
> to be printed. I am looking for a way for me to attach an attribute to the
> test function so that the string in the attribute gets printed.
>
> Is there something readily avaliable for doing this?
>
> Currently my test report comes out as:
> testDateDiffWeek (PI.GetMessages.GetMessagesTestClass) ... ok
> testGetMessageGUIDnull (PI.GetMessages.GetMessagesTestClass) ... FAIL
>
> Thanks,
> Yash
>
>
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>
>


From brian at merrells.org  Sat Dec 20 22:56:44 2008
From: brian at merrells.org (Brian Merrell)
Date: Sat, 20 Dec 2008 16:56:44 -0500
Subject: [IronPython] Problems with hashlib and threading
Message-ID: <4a3ced360812201356i4914639gefa19fcb8d326d4f@mail.gmail.com>

I get the following error periodically:

Exception in thread Thread-2:
Traceback (most recent call last):
  File "C:\Program Files\IronPython 2.0\Lib\threading.py", line 486, in
_Thread__bootstrap_inner
    self.run()
  File "initial_ingest.py", line 33, in run
    md5 = hashlib.md5(contents).hexdigest()
SystemError: Hash not valid for use in specified state.

I am not sharing a md5() object between threads and have tried to avoid any
shared memory issues:


class Convert(threading.Thread):
    def __init__(self, queue):
        threading.Thread.__init__(self)
        self.queue = queue
    def run(self):
        import hashlib
        while True:
            abs_filename = self.queue.get()
            f = open(abs_filename, "rb")
            contents = f.read()
            f.close()
            md5 = hashlib.md5(contents).hexdigest()
            print md5, abs_filename
            self.queue.task_done()

Any ideas why this is failing?  Thanks,

-brian
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081220/2baa0c33/attachment.html>

From kristian.jaksch at gmail.com  Sun Dec 21 16:31:46 2008
From: kristian.jaksch at gmail.com (xkrja)
Date: Sun, 21 Dec 2008 07:31:46 -0800 (PST)
Subject: [IronPython] MouseButtonDownEvent doesn't fire + which tool to work
 in?
Message-ID: <21115998.post@talk.nabble.com>


I've the following code:

def inputBox_MouseLeftButtonDown(s, e):
    if isFirst == True:
	    root.inputBox.Text = ""
	    isFirst = False

root.inputBox.MouseLeftButtonDown += inputBox_MouseLeftButtonDown

Problem is the event doesn't fire when I click the inputBox (which is a
textbox). The debugger doesn't say anything. I have another event handler
and that one works fine:

root.inputBox.KeyDown += inputBox_KeyDown

Anyone has any idea what's wrong with the MouseLeftButtonDown event?


Another question: Which tool do you guys say is the best for developing
IronPython + Silverlight apps? I find it hard in Visual Studio since the
debugger only lets you step through the code but you can't see any variable
values.
-- 
View this message in context: http://www.nabble.com/MouseButtonDownEvent-doesn%27t-fire-%2B-which-tool-to-work-in--tp21115998p21115998.html
Sent from the IronPython mailing list archive at Nabble.com.



From curt at hagenlocher.org  Sun Dec 21 16:35:05 2008
From: curt at hagenlocher.org (Curt Hagenlocher)
Date: Sun, 21 Dec 2008 07:35:05 -0800
Subject: [IronPython] MouseButtonDownEvent doesn't fire + which tool to
	work in?
In-Reply-To: <21115998.post@talk.nabble.com>
References: <21115998.post@talk.nabble.com>
Message-ID: <d2155e360812210735t69d0a99dj228e102857d1c1c9@mail.gmail.com>

How are you confirming that this doesn't fire?  If this is your exact code,
then it seems likely that it's actually throwing an exception -- isFirst
will not be defined when the first line executes.

On Sun, Dec 21, 2008 at 7:31 AM, xkrja <kristian.jaksch at gmail.com> wrote:

>
> I've the following code:
>
> def inputBox_MouseLeftButtonDown(s, e):
>    if isFirst == True:
>            root.inputBox.Text = ""
>            isFirst = False
>
> root.inputBox.MouseLeftButtonDown += inputBox_MouseLeftButtonDown
>
> Problem is the event doesn't fire when I click the inputBox (which is a
> textbox). The debugger doesn't say anything. I have another event handler
> and that one works fine:
>
> root.inputBox.KeyDown += inputBox_KeyDown
>
> Anyone has any idea what's wrong with the MouseLeftButtonDown event?
>
>
> Another question: Which tool do you guys say is the best for developing
> IronPython + Silverlight apps? I find it hard in Visual Studio since the
> debugger only lets you step through the code but you can't see any variable
> values.
> --
> View this message in context:
> http://www.nabble.com/MouseButtonDownEvent-doesn%27t-fire-%2B-which-tool-to-work-in--tp21115998p21115998.html
> Sent from the IronPython mailing list archive at Nabble.com.
>
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081221/d4868f1f/attachment.html>

From kristian.jaksch at gmail.com  Sun Dec 21 19:31:21 2008
From: kristian.jaksch at gmail.com (Kristian Jaksch)
Date: Sun, 21 Dec 2008 19:31:21 +0100
Subject: [IronPython] MouseButtonDownEvent doesn't fire + which tool to
	work in?
In-Reply-To: <d2155e360812210735t69d0a99dj228e102857d1c1c9@mail.gmail.com>
References: <21115998.post@talk.nabble.com>
	<d2155e360812210735t69d0a99dj228e102857d1c1c9@mail.gmail.com>
Message-ID: <1dab55500812211031j28e27494ia0b9a1f068c0ac25@mail.gmail.com>

This is no all the code but even if I skip the 'if'-statement and have it
like below:

*def inputBox_MouseLeftButtonDown(s, e):
       root.inputBox.Text = ""*

it still doesn't work. And if I put a breakpoint at the "*def
inputBox_MouseLeftButtonDown(s, e):*"-line it isn't fired.


2008/12/21 Curt Hagenlocher <curt at hagenlocher.org>

> How are you confirming that this doesn't fire?  If this is your exact code,
> then it seems likely that it's actually throwing an exception -- isFirst
> will not be defined when the first line executes.
>
>
> On Sun, Dec 21, 2008 at 7:31 AM, xkrja <kristian.jaksch at gmail.com> wrote:
>
>>
>> I've the following code:
>>
>> def inputBox_MouseLeftButtonDown(s, e):
>>    if isFirst == True:
>>            root.inputBox.Text = ""
>>            isFirst = False
>>
>> root.inputBox.MouseLeftButtonDown += inputBox_MouseLeftButtonDown
>>
>> Problem is the event doesn't fire when I click the inputBox (which is a
>> textbox). The debugger doesn't say anything. I have another event handler
>> and that one works fine:
>>
>> root.inputBox.KeyDown += inputBox_KeyDown
>>
>> Anyone has any idea what's wrong with the MouseLeftButtonDown event?
>>
>>
>> Another question: Which tool do you guys say is the best for developing
>> IronPython + Silverlight apps? I find it hard in Visual Studio since the
>> debugger only lets you step through the code but you can't see any
>> variable
>> values.
>> --
>> View this message in context:
>> http://www.nabble.com/MouseButtonDownEvent-doesn%27t-fire-%2B-which-tool-to-work-in--tp21115998p21115998.html
>> Sent from the IronPython mailing list archive at Nabble.com.
>>
>> _______________________________________________
>> Users mailing list
>> Users at lists.ironpython.com
>> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>>
>
>
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081221/1ceee6b8/attachment.html>

From jdhardy at gmail.com  Sun Dec 21 21:30:10 2008
From: jdhardy at gmail.com (Jeff Hardy)
Date: Sun, 21 Dec 2008 13:30:10 -0700
Subject: [IronPython] Zlib module
Message-ID: <ea89d0e50812211230kb49d0f1x698d532b4390407a@mail.gmail.com>

Hello,
I've posted a mostly-complete zlib implementation at
http://cid-414fa1a9bd174b4b.skydrive.live.com/self.aspx/Public/IronPython.Zlib.zip.
It uses zlib.net from ComponentAce
(http://www.componentace.com/zlib_.NET.htm) which is an open source,
pure .NET implementation of zlib. This was necessary as I could not
figure out a way to emulate (de)compressobj using .NET DeflateStream.

It passes most of the zlib tests (there are a few that still fail),
and all of the gzip tests. The zipfile test are currently failing
because binascii.crc32 is not implemented. There are probably other
bugs I haven't found.

The source code is available at
http://www.bitbucket.org/jdhardy/code/src/tip/IronPython.Zlib/. You'll
need Mercurial to access it.

-Jeff


From dan.eloff at gmail.com  Mon Dec 22 01:15:05 2008
From: dan.eloff at gmail.com (Dan Eloff)
Date: Sun, 21 Dec 2008 19:15:05 -0500
Subject: [IronPython] MouseButtonDownEvent doesn't fire + which tool to
	work in?
In-Reply-To: <1dab55500812211031j28e27494ia0b9a1f068c0ac25@mail.gmail.com>
References: <21115998.post@talk.nabble.com>
	<d2155e360812210735t69d0a99dj228e102857d1c1c9@mail.gmail.com>
	<1dab55500812211031j28e27494ia0b9a1f068c0ac25@mail.gmail.com>
Message-ID: <4817b6fc0812211615l53900f95xe93df0a375b9ed56@mail.gmail.com>

MouseLeftButtonDown is handled by the TextBox control, you cannot
handle the event unless you subclass TextBox and override
OnMouseLeftButtonDown.

-Dan

On Sun, Dec 21, 2008 at 1:31 PM, Kristian Jaksch
<kristian.jaksch at gmail.com> wrote:
> This is no all the code but even if I skip the 'if'-statement and have it
> like below:
>
> def inputBox_MouseLeftButtonDown(s, e):
>        root.inputBox.Text = ""
>
> it still doesn't work. And if I put a breakpoint at the "def
> inputBox_MouseLeftButtonDown(s, e):"-line it isn't fired.
>
>
> 2008/12/21 Curt Hagenlocher <curt at hagenlocher.org>
>>
>> How are you confirming that this doesn't fire?  If this is your exact
>> code, then it seems likely that it's actually throwing an exception --
>> isFirst will not be defined when the first line executes.
>>
>> On Sun, Dec 21, 2008 at 7:31 AM, xkrja <kristian.jaksch at gmail.com> wrote:
>>>
>>> I've the following code:
>>>
>>> def inputBox_MouseLeftButtonDown(s, e):
>>>    if isFirst == True:
>>>            root.inputBox.Text = ""
>>>            isFirst = False
>>>
>>> root.inputBox.MouseLeftButtonDown += inputBox_MouseLeftButtonDown
>>>
>>> Problem is the event doesn't fire when I click the inputBox (which is a
>>> textbox). The debugger doesn't say anything. I have another event handler
>>> and that one works fine:
>>>
>>> root.inputBox.KeyDown += inputBox_KeyDown
>>>
>>> Anyone has any idea what's wrong with the MouseLeftButtonDown event?
>>>
>>>
>>> Another question: Which tool do you guys say is the best for developing
>>> IronPython + Silverlight apps? I find it hard in Visual Studio since the
>>> debugger only lets you step through the code but you can't see any
>>> variable
>>> values.
>>> --
>>> View this message in context:
>>> http://www.nabble.com/MouseButtonDownEvent-doesn%27t-fire-%2B-which-tool-to-work-in--tp21115998p21115998.html
>>> Sent from the IronPython mailing list archive at Nabble.com.
>>>
>>> _______________________________________________
>>> Users mailing list
>>> Users at lists.ironpython.com
>>> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>>
>>
>> _______________________________________________
>> Users mailing list
>> Users at lists.ironpython.com
>> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>>
>
>
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>
>


From andyjian430074 at gmail.com  Mon Dec 22 08:52:07 2008
From: andyjian430074 at gmail.com (Jankins)
Date: Mon, 22 Dec 2008 01:52:07 -0600
Subject: [IronPython] Why can't I use 'import os' in Ironpython Studio? It
	always says "No module named os".
Message-ID: <afeca0fb0812212352x5864cf8do41e6a26996f58151@mail.gmail.com>

Hi,

I have been using Ironpython studio for about two days. And I found a
strange problem. I can't use "import os" or  " import copy" (until now I
just found these two) in my program, because when I debug my program, it
says "No module named os". But I can use "import sys" , "import time" or
"import math". I don't know why.

I reinstall Ironpython Studio, but it still doesn't work. And I tried to
find the lib location used by Iropython studio, but I can't find it.

Please help me about it. This problem really drives me crazy.

Thanks so much.

Jankins
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081222/5c47d590/attachment.html>

From empirebuilder at gmail.com  Mon Dec 22 17:43:29 2008
From: empirebuilder at gmail.com (Dody Gunawinata)
Date: Mon, 22 Dec 2008 17:43:29 +0100
Subject: [IronPython] ASP.Net IronPython Refresh
In-Reply-To: <5283CA0A4168DF4FBBD71AE9ECA5A3284BDEA69D14@NA-EXMSG-C116.redmond.corp.microsoft.com>
References: <8cd017b80812171159l55653792ve4ba9fb5754f24eb@mail.gmail.com>
	<5283CA0A4168DF4FBBD71AE9ECA5A3284BDEA69D14@NA-EXMSG-C116.redmond.corp.microsoft.com>
Message-ID: <8cd017b80812220843r332ca088m95f6c1c387bec9db@mail.gmail.com>

Great. I'm really looking forward to it. My CMS has been running on
Frankestein mode with IP Alpha for ASP.Net and IP 2.0 "custom build" for the
templating engine for a year now. Running two different version of
IronPython in a single web application can be a harrowing experience.
Dody G.

On Sat, Dec 20, 2008 at 1:43 AM, Jimmy Schementi <
Jimmy.Schementi at microsoft.com> wrote:

>  Hopefully getting a build out soon ? I'll keep this list posted.
>
>
>
> *From:* users-bounces at lists.ironpython.com [mailto:
> users-bounces at lists.ironpython.com] *On Behalf Of *Dody Gunawinata
> *Sent:* Wednesday, December 17, 2008 12:00 PM
> *To:* Discussion of IronPython
> *Subject:* [IronPython] ASP.Net IronPython Refresh
>
>
>
> Does anybody know when the IronPython for ASP.Net gets updated for the
> final version of IronPython 2.0?
>
> --
> nomadlife.org
>



-- 
nomadlife.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081222/fc8d8239/attachment.html>

From kfarmer at thuban.org  Tue Dec 23 02:02:30 2008
From: kfarmer at thuban.org (Keith J. Farmer)
Date: Mon, 22 Dec 2008 17:02:30 -0800
Subject: [IronPython] Announcing IronPython 2.0
In-Reply-To: <EECFB552375FDE49A23A13CE85EEC2FD1E9834A4D5@NA-EXMSG-C112.redmond.corp.microsoft.com>
References: <EECFB552375FDE49A23A13CE85EEC2FD1E9834A4D5@NA-EXMSG-C112.redmond.corp.microsoft.com>
Message-ID: <EEDCD954DD4A0D488BBDBDDCC260754466FB95@meta-exchange.circle-cross.home>

Since there?s an MSI now, could we hope for updates/new-major-versions via Windows Update?  Like online banking, I?ve come to rely on things updating themselves, rather than having to keep track of various support download sites.  Not to be exclusive ? I would expect downloads at CodePlex to remain available -- but for the ease of installation and maintainance, it?s hard to beat built-in, unified updaters.

 

If side-by-side versioning is required, I could see a current-version environment variable that could be added to the path.

 

From: users-bounces at lists.ironpython.com [mailto:users-bounces at lists.ironpython.com] On Behalf Of Dave Fugate
Sent: Wednesday, December 10, 2008 1:19 PM
To: Discussion of IronPython
Cc: python-announce-list at python.org
Subject: [IronPython] Announcing IronPython 2.0

 

Hello Python Community,

 

The IronPython and Dynamic Language Runtime teams are proud to announce the release of IronPython 2.0 final.  IronPython 2.0 is the culmination of nearly two years worth of work resulting in a CPython 2.5 compatible release on .NET 2.0 SP1.  By far, the biggest change to 2.0 is that our 1.1 codebase was refactored to run on top of the Dynamic Language Runtime <http://www.codeplex.com/dlr> .  With this we automatically get improvements in many feature areas such as better .NET interop support and hosting Python from managed code.  There have been many other major improvements as well.  The most notable are:

?         An MSI installer for Windows platforms which includes parts of the CPython 2.5 standard library

?         IronPython assemblies targeting Silverlight and tools such as Chiron to improve the Silverlight dynamic development experience

?         The addition of more C-based standard modules such as cmath and _winreg

?         Significant improvements in importing compatibility and features

?         Distribution of IronPython under the Microsoft Public License which has been approved by OSI

?         Performance improvements.  On that note, a new Wiki page has been created for IronPython performance reports - see http://www.codeplex.com/IronPython/Wiki/View.aspx?title=IronPython%20Performance

?         Over 500 bugs have been closed in 2.0.  453 of these were reported on CodePlex

?         Support for precompilation of Python source files into a single dll

 

This seems like an opportune time to remind everyone that we fix bugs based on the number of votes they have on CodePlex.  As we?re planning on releasing IronPython 2.0.1 fairly soon, please vote for your favorite bugs at http://www.codeplex.com/IronPython/WorkItem/AdvancedList.aspx to help ensure they get fixed in time for the next release.

 

We?d like to extend our gratitude to everyone in the IronPython community who reported bugs thereby making this a better release:  milind, romank, chadaustin, sjmachin, davidfraser, TimothyFitz, drewid, sanxiyn, bashmohandes, pobrien, perhaps, haypo, Undebtedly, ayarrow, tscottw_cp, rope, arman0, eshaish, nivaldo, fuzzyman, CurtHagenlocher, Eloff, brucec, py_sunil, jacobg23, mziller, beaugunderson, gbraad, Oceanborn, tarlano, jbevain, glchapman, anthonybaxter, jdhardy, jjlee, haibo, doubleyewdee, jackeyoo, whit537, sdahlbac, PeteHufnagel, jtenney, nriley, junfeng, grizlupo, rridge, lewisle, JoelBondurant, johnplatt, lthompson, debackerl, googen, tscottw, VoteFw, leppie, Qvin, heyssion2, CriGoT, baxeno, sbergman, Laurion, luntain, oldman, christmas, 05031972, kevgu, wilberforce, Korbinian, lclj, sorokerr, Eriol, tatwright, ais, TraumaPony, pelikhan, asafk, felixpollan, srid, atifaziz, vernondcole, fwereade, zpy, yanne, facorreia, Daneel, zvikag, psykotic, Cavingdeep, BEaton, sborde, orbital56, fbourgeois, antont, krosavcheg, ktc1, awilkins, ben2004uk, paulfelix, axl, JeffreySax, Lawouach, and KKI.

 

You can download IronPython 2.0 at:  http://www.codeplex.com/IronPython/Release/ProjectReleases.aspx?ReleaseId=8365

 

The IronPython Team

 

 

BTW

Updates to the IronPython 1.0 samples will be released shortly.  Stay tuned to the IronPython mailing list for details.

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081222/2a074ff4/attachment.html>

From jdhardy at gmail.com  Tue Dec 23 03:10:43 2008
From: jdhardy at gmail.com (Jeff Hardy)
Date: Mon, 22 Dec 2008 19:10:43 -0700
Subject: [IronPython] Detecting IronPython Installation Directory
Message-ID: <ea89d0e50812221810y5086f53ay8104c7c98e03781e@mail.gmail.com>

Hi,
Does the IronPython 2.0 installer write the installation directory to
a registry key (besides the WIndows Installer keys)? If not, could
that be added to a future release? Or, is there another way to get the
installation directory?

-Jeff


From inigomaringarcia at gmail.com  Tue Dec 23 08:23:35 2008
From: inigomaringarcia at gmail.com (Inigo Marin Garcia)
Date: Tue, 23 Dec 2008 08:23:35 +0100
Subject: [IronPython] unsuscribe
Message-ID: <49bd9cb0812222323h7f128a41wa25564cc7a75658@mail.gmail.com>

-- 
Saludos

I?igo Marin Garcia
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081223/316ad4da/attachment.html>

From fuzzyman at voidspace.org.uk  Tue Dec 23 11:18:19 2008
From: fuzzyman at voidspace.org.uk (Michael Foord)
Date: Tue, 23 Dec 2008 10:18:19 +0000
Subject: [IronPython] Detecting IronPython Installation Directory
In-Reply-To: <ea89d0e50812221810y5086f53ay8104c7c98e03781e@mail.gmail.com>
References: <ea89d0e50812221810y5086f53ay8104c7c98e03781e@mail.gmail.com>
Message-ID: <4950BAEB.5030206@voidspace.org.uk>

Jeff Hardy wrote:
> Hi,
> Does the IronPython 2.0 installer write the installation directory to
> a registry key (besides the WIndows Installer keys)? If not, could
> that be added to a future release? Or, is there another way to get the
> installation directory?
>
> -Jeff
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>   
If you are doing it *from* IronPython then you can always use the 
location of the executing assembly - but I guess that isn't what you are 
doing. :-)

Michael

-- 
http://www.ironpythoninaction.com/



From fuzzyman at voidspace.org.uk  Tue Dec 23 12:40:15 2008
From: fuzzyman at voidspace.org.uk (Michael Foord)
Date: Tue, 23 Dec 2008 11:40:15 +0000
Subject: [IronPython] ANN : Davy's IronPython Editor
In-Reply-To: <20253b0c0812200322w4734bf90u6947528ce97e92c8@mail.gmail.com>
References: <20253b0c0812200322w4734bf90u6947528ce97e92c8@mail.gmail.com>
Message-ID: <4950CE1F.7060308@voidspace.org.uk>

Davy Mitchell wrote:
> Hi Folks,
>
> I've made a small start on an IronPython Editor in WinForms using 
> SyntaxBox.  The project intentions is to create an editor along the 
> lines of SPE or DrPython.
>
> If you want to take a look there are a couple of screen shots - please 
> visit:
> http://code.google.com/p/davysironpythoneditor/
>
> Code in SVN only at the moment. DIE does benefit from being compiled. 
> Maybe once I stop sniffing and coughing so much it will get done and 
> uploaded ! :-)
>
> It's not production ready yet so please, please be careful with your 
> precious source code. I've not tried it on Mono yet.
>
> Feedback and ideas appreciated.
>

Awesome! You should look at PySmell for autocomplete. It won't run under 
IronPython but is intended to be run with CPython called as a subprocess 
anyway. I'm sure Orestis will be happy to give you some pointers if you 
go down this route.

All the best,

Michael Foord


> Thanks & Merry Christmas,
> Davy Mitchell
> ------------------------------------------------------------------------
>
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>   


-- 
http://www.ironpythoninaction.com/



From fuzzyman at voidspace.org.uk  Tue Dec 23 12:40:52 2008
From: fuzzyman at voidspace.org.uk (Michael Foord)
Date: Tue, 23 Dec 2008 11:40:52 +0000
Subject: [IronPython] Zlib module
In-Reply-To: <ea89d0e50812211230kb49d0f1x698d532b4390407a@mail.gmail.com>
References: <ea89d0e50812211230kb49d0f1x698d532b4390407a@mail.gmail.com>
Message-ID: <4950CE44.7000204@voidspace.org.uk>

Jeff Hardy wrote:
> Hello,
> I've posted a mostly-complete zlib implementation at
> http://cid-414fa1a9bd174b4b.skydrive.live.com/self.aspx/Public/IronPython.Zlib.zip.
> It uses zlib.net from ComponentAce
> (http://www.componentace.com/zlib_.NET.htm) which is an open source,
> pure .NET implementation of zlib. This was necessary as I could not
> figure out a way to emulate (de)compressobj using .NET DeflateStream.
>
> It passes most of the zlib tests (there are a few that still fail),
> and all of the gzip tests. The zipfile test are currently failing
> because binascii.crc32 is not implemented. There are probably other
> bugs I haven't found.
>
> The source code is available at
> http://www.bitbucket.org/jdhardy/code/src/tip/IronPython.Zlib/. You'll
> need Mercurial to access it.
>   
That's great news Jeff - thanks for your efforts.

Michael Foord


> -Jeff
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>   


-- 
http://www.ironpythoninaction.com/



From fuzzyman at voidspace.org.uk  Tue Dec 23 12:43:48 2008
From: fuzzyman at voidspace.org.uk (Michael Foord)
Date: Tue, 23 Dec 2008 11:43:48 +0000
Subject: [IronPython] Why can't I use 'import os' in Ironpython Studio?
 It	always says "No module named os".
In-Reply-To: <afeca0fb0812212352x5864cf8do41e6a26996f58151@mail.gmail.com>
References: <afeca0fb0812212352x5864cf8do41e6a26996f58151@mail.gmail.com>
Message-ID: <4950CEF4.4010206@voidspace.org.uk>

Jankins wrote:
> Hi,
>
> I have been using Ironpython studio for about two days. And I found a 
> strange problem. I can't use "import os" or  " import copy" (until now 
> I just found these two) in my program, because when I debug my 
> program, it says "No module named os". But I can use "import sys" , 
> "import time" or "import math". I don't know why.
>
> I reinstall Ironpython Studio, but it still doesn't work. And I tried 
> to find the lib location used by Iropython studio, but I can't find it.
>
> Please help me about it. This problem really drives me crazy.
>

Mainly because IronPython Studio, despite being a very impressive proof 
of concept, is a steaming heap of dung for actually developing with.

If you need the winforms / wpf designer then I recommend Visual Studio 
(Express) generating C# that you then use from IronPython. For 
developing Python projects you can use a real Python IDE like Wing, 
Komodo or SPE - or even a text editor like Vim (*ouch*) or Emacs. Any of 
these will suit you much better than IronPython Studio I'm afraid.

All the best,

Michael Foord

> Thanks so much.
>
> Jankins
> ------------------------------------------------------------------------
>
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>   


-- 
http://www.ironpythoninaction.com/



From curt at hagenlocher.org  Tue Dec 23 18:28:28 2008
From: curt at hagenlocher.org (Curt Hagenlocher)
Date: Tue, 23 Dec 2008 09:28:28 -0800
Subject: [IronPython] Detecting IronPython Installation Directory
In-Reply-To: <ea89d0e50812221810y5086f53ay8104c7c98e03781e@mail.gmail.com>
References: <ea89d0e50812221810y5086f53ay8104c7c98e03781e@mail.gmail.com>
Message-ID: <d2155e360812230928v116fcc4eua7564c7bd5e1af99@mail.gmail.com>

Looking at the WiX source (is that included in the source distribution?), it
doesn't appear that we're deliberately writing this information into the
registry.  But the MsiProductInfo function should be able to retrieve it
from the sekrit Windows Installer stash.
http://www.pinvoke.net/default.aspx/msi/MsiGetProductInfo.html

Alas, I don't imagine this can be easily accessed from within IronPython
itself, but as Michael suggests, there are simpler ways of doing that :).

On Mon, Dec 22, 2008 at 6:10 PM, Jeff Hardy <jdhardy at gmail.com> wrote:

> Hi,
> Does the IronPython 2.0 installer write the installation directory to
> a registry key (besides the WIndows Installer keys)? If not, could
> that be added to a future release? Or, is there another way to get the
> installation directory?
>
> -Jeff
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081223/5f751df9/attachment.html>

From olegtk at microsoft.com  Tue Dec 23 19:17:58 2008
From: olegtk at microsoft.com (Oleg Tkachenko)
Date: Tue, 23 Dec 2008 10:17:58 -0800
Subject: [IronPython] Detecting IronPython Installation Directory
In-Reply-To: <d2155e360812230928v116fcc4eua7564c7bd5e1af99@mail.gmail.com>
References: <ea89d0e50812221810y5086f53ay8104c7c98e03781e@mail.gmail.com>
	<d2155e360812230928v116fcc4eua7564c7bd5e1af99@mail.gmail.com>
Message-ID: <A5DE25409C1E51458D103EDB73EA953A79B0B9D4CF@NA-EXMSG-C104.redmond.corp.microsoft.com>

Actually we do store install location in the registry. It can be found somewhere deep in HKLM\SOFTWARE\Microsoft\Windows\CurrentVersion\Installer subtree.
It also can be queried using System.Management.Instrumentation:

            ManagementObjectSearcher query1 = new ManagementObjectSearcher(
                "SELECT * FROM Win32_Product where InstallState = 5 and Caption = 'IronPython 2.0'");
            ManagementObjectCollection queryCollection1 = query1.Get();

            foreach (ManagementObject mo in queryCollection1) {
                Console.WriteLine(mo["InstallLocation"]);
            }

--
Oleg

From: users-bounces at lists.ironpython.com [mailto:users-bounces at lists.ironpython.com] On Behalf Of Curt Hagenlocher
Sent: Tuesday, December 23, 2008 9:28 AM
To: Discussion of IronPython
Subject: Re: [IronPython] Detecting IronPython Installation Directory

Looking at the WiX source (is that included in the source distribution?), it doesn't appear that we're deliberately writing this information into the registry.  But the MsiProductInfo function should be able to retrieve it from the sekrit Windows Installer stash.

http://www.pinvoke.net/default.aspx/msi/MsiGetProductInfo.html

Alas, I don't imagine this can be easily accessed from within IronPython itself, but as Michael suggests, there are simpler ways of doing that :).
On Mon, Dec 22, 2008 at 6:10 PM, Jeff Hardy <jdhardy at gmail.com<mailto:jdhardy at gmail.com>> wrote:
Hi,
Does the IronPython 2.0 installer write the installation directory to
a registry key (besides the WIndows Installer keys)? If not, could
that be added to a future release? Or, is there another way to get the
installation directory?

-Jeff
_______________________________________________
Users mailing list
Users at lists.ironpython.com<mailto:Users at lists.ironpython.com>
http://lists.ironpython.com/listinfo.cgi/users-ironpython.com

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081223/7b3f38ff/attachment.html>

From curt at hagenlocher.org  Tue Dec 23 19:21:00 2008
From: curt at hagenlocher.org (Curt Hagenlocher)
Date: Tue, 23 Dec 2008 10:21:00 -0800
Subject: [IronPython] Detecting IronPython Installation Directory
In-Reply-To: <A5DE25409C1E51458D103EDB73EA953A79B0B9D4CF@NA-EXMSG-C104.redmond.corp.microsoft.com>
References: <ea89d0e50812221810y5086f53ay8104c7c98e03781e@mail.gmail.com>
	<d2155e360812230928v116fcc4eua7564c7bd5e1af99@mail.gmail.com>
	<A5DE25409C1E51458D103EDB73EA953A79B0B9D4CF@NA-EXMSG-C104.redmond.corp.microsoft.com>
Message-ID: <d2155e360812231021j58254485ob7b3ec5861fea9e2@mail.gmail.com>

I think that's just the WMI way of doing an MsiProductInfo call :).  Though
it sure is a lot more managed- and Python- friendly than having to do a
P/Invoke.

On Tue, Dec 23, 2008 at 10:17 AM, Oleg Tkachenko <olegtk at microsoft.com>wrote:

>  Actually we do store install location in the registry. It can be found
> somewhere deep in HKLM\SOFTWARE\Microsoft\Windows\CurrentVersion\Installer
> subtree.
>
> It also can be queried using System.Management.Instrumentation:
>
>
>
>             ManagementObjectSearcher query1 = new ManagementObjectSearcher(
>
>                 "SELECT * FROM Win32_Product where InstallState = 5 and
> Caption = 'IronPython 2.0'");
>
>             ManagementObjectCollection queryCollection1 = query1.Get();
>
>
>
>             foreach (ManagementObject mo in queryCollection1) {
>
>                 Console.WriteLine(mo["InstallLocation"]);
>
>             }
>
>
>
> --
>
> Oleg
>
>
>
> *From:* users-bounces at lists.ironpython.com [mailto:
> users-bounces at lists.ironpython.com] *On Behalf Of *Curt Hagenlocher
> *Sent:* Tuesday, December 23, 2008 9:28 AM
> *To:* Discussion of IronPython
> *Subject:* Re: [IronPython] Detecting IronPython Installation Directory
>
>
>
> Looking at the WiX source (is that included in the source distribution?),
> it doesn't appear that we're deliberately writing this information into the
> registry.  But the MsiProductInfo function should be able to retrieve it
> from the sekrit Windows Installer stash.
>
>
>
> http://www.pinvoke.net/default.aspx/msi/MsiGetProductInfo.html
>
>
>
> Alas, I don't imagine this can be easily accessed from within IronPython
> itself, but as Michael suggests, there are simpler ways of doing that :).
>
> On Mon, Dec 22, 2008 at 6:10 PM, Jeff Hardy <jdhardy at gmail.com> wrote:
>
> Hi,
> Does the IronPython 2.0 installer write the installation directory to
> a registry key (besides the WIndows Installer keys)? If not, could
> that be added to a future release? Or, is there another way to get the
> installation directory?
>
> -Jeff
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>
>
>
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081223/0362e56a/attachment.html>

From jdhardy at gmail.com  Tue Dec 23 20:40:51 2008
From: jdhardy at gmail.com (Jeff Hardy)
Date: Tue, 23 Dec 2008 12:40:51 -0700
Subject: [IronPython] Detecting IronPython Installation Directory
In-Reply-To: <d2155e360812231021j58254485ob7b3ec5861fea9e2@mail.gmail.com>
References: <ea89d0e50812221810y5086f53ay8104c7c98e03781e@mail.gmail.com>
	<d2155e360812230928v116fcc4eua7564c7bd5e1af99@mail.gmail.com>
	<A5DE25409C1E51458D103EDB73EA953A79B0B9D4CF@NA-EXMSG-C104.redmond.corp.microsoft.com>
	<d2155e360812231021j58254485ob7b3ec5861fea9e2@mail.gmail.com>
Message-ID: <ea89d0e50812231140j592875d2u3eb2618bea397a8f@mail.gmail.com>

Hi,
Thanks for the tips! To add a little bit more info, I'm working in
Powershell for various reasons, so P/Invoke is out of the question.
Using WMI is easy enough, but I'd prefer not to be mucking around
inside Windows Installer if I can help it.

Is there any reason to not include an entry (e.g.
HKLM:\SOFTWARE\Microsoft\IronPython\InstallPath) that stores the
installation path?. It's easy enough to do with WiX.

- Jeff

On Tue, Dec 23, 2008 at 11:21 AM, Curt Hagenlocher <curt at hagenlocher.org> wrote:
> I think that's just the WMI way of doing an MsiProductInfo call :).  Though
> it sure is a lot more managed- and Python- friendly than having to do a
> P/Invoke.
>
> On Tue, Dec 23, 2008 at 10:17 AM, Oleg Tkachenko <olegtk at microsoft.com>
> wrote:
>>
>> Actually we do store install location in the registry. It can be found
>> somewhere deep in HKLM\SOFTWARE\Microsoft\Windows\CurrentVersion\Installer
>> subtree.
>>
>> It also can be queried using System.Management.Instrumentation:
>>
>>
>>
>>             ManagementObjectSearcher query1 = new
>> ManagementObjectSearcher(
>>
>>                 "SELECT * FROM Win32_Product where InstallState = 5 and
>> Caption = 'IronPython 2.0'");
>>
>>             ManagementObjectCollection queryCollection1 = query1.Get();
>>
>>
>>
>>             foreach (ManagementObject mo in queryCollection1) {
>>
>>                 Console.WriteLine(mo["InstallLocation"]);
>>
>>             }
>>
>>
>>
>> --
>>
>> Oleg


From curt at hagenlocher.org  Tue Dec 23 20:53:42 2008
From: curt at hagenlocher.org (Curt Hagenlocher)
Date: Tue, 23 Dec 2008 11:53:42 -0800
Subject: [IronPython] Detecting IronPython Installation Directory
In-Reply-To: <ea89d0e50812231140j592875d2u3eb2618bea397a8f@mail.gmail.com>
References: <ea89d0e50812221810y5086f53ay8104c7c98e03781e@mail.gmail.com>
	<d2155e360812230928v116fcc4eua7564c7bd5e1af99@mail.gmail.com>
	<A5DE25409C1E51458D103EDB73EA953A79B0B9D4CF@NA-EXMSG-C104.redmond.corp.microsoft.com>
	<d2155e360812231021j58254485ob7b3ec5861fea9e2@mail.gmail.com>
	<ea89d0e50812231140j592875d2u3eb2618bea397a8f@mail.gmail.com>
Message-ID: <d2155e360812231153y6ba3a0fw7cd152cd0f17cb5f@mail.gmail.com>

Given that there are clean APIs for pulling the data out, what are your
objections to using the installer data?
On Tue, Dec 23, 2008 at 11:40 AM, Jeff Hardy <jdhardy at gmail.com> wrote:

> Hi,
> Thanks for the tips! To add a little bit more info, I'm working in
> Powershell for various reasons, so P/Invoke is out of the question.
> Using WMI is easy enough, but I'd prefer not to be mucking around
> inside Windows Installer if I can help it.
>
> Is there any reason to not include an entry (e.g.
> HKLM:\SOFTWARE\Microsoft\IronPython\InstallPath) that stores the
> installation path?. It's easy enough to do with WiX.
>
> - Jeff
>
> On Tue, Dec 23, 2008 at 11:21 AM, Curt Hagenlocher <curt at hagenlocher.org>
> wrote:
> > I think that's just the WMI way of doing an MsiProductInfo call :).
>  Though
> > it sure is a lot more managed- and Python- friendly than having to do a
> > P/Invoke.
> >
> > On Tue, Dec 23, 2008 at 10:17 AM, Oleg Tkachenko <olegtk at microsoft.com>
> > wrote:
> >>
> >> Actually we do store install location in the registry. It can be found
> >> somewhere deep in
> HKLM\SOFTWARE\Microsoft\Windows\CurrentVersion\Installer
> >> subtree.
> >>
> >> It also can be queried using System.Management.Instrumentation:
> >>
> >>
> >>
> >>             ManagementObjectSearcher query1 = new
> >> ManagementObjectSearcher(
> >>
> >>                 "SELECT * FROM Win32_Product where InstallState = 5 and
> >> Caption = 'IronPython 2.0'");
> >>
> >>             ManagementObjectCollection queryCollection1 = query1.Get();
> >>
> >>
> >>
> >>             foreach (ManagementObject mo in queryCollection1) {
> >>
> >>                 Console.WriteLine(mo["InstallLocation"]);
> >>
> >>             }
> >>
> >>
> >>
> >> --
> >>
> >> Oleg
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081223/5698dfe8/attachment.html>

From curt at hagenlocher.org  Tue Dec 23 20:57:12 2008
From: curt at hagenlocher.org (Curt Hagenlocher)
Date: Tue, 23 Dec 2008 11:57:12 -0800
Subject: [IronPython] Announcing IronPython 2.0
In-Reply-To: <EEDCD954DD4A0D488BBDBDDCC260754466FB95@meta-exchange.circle-cross.home>
References: <EECFB552375FDE49A23A13CE85EEC2FD1E9834A4D5@NA-EXMSG-C112.redmond.corp.microsoft.com>
	<EEDCD954DD4A0D488BBDBDDCC260754466FB95@meta-exchange.circle-cross.home>
Message-ID: <d2155e360812231157r1a20350bt91448ee135f90574@mail.gmail.com>

As with all feature requests, please add it to CodePlex. :)

On Mon, Dec 22, 2008 at 5:02 PM, Keith J. Farmer <kfarmer at thuban.org> wrote:

>  Since there's an MSI now, could we hope for updates/new-major-versions
> via Windows Update?  Like online banking, I've come to rely on things
> updating themselves, rather than having to keep track of various support
> download sites.  Not to be exclusive ? I would expect downloads at CodePlex
> to remain available -- but for the ease of installation and maintainance,
> it's hard to beat built-in, unified updaters.
>
>
>
> If side-by-side versioning is required, I could see a current-version
> environment variable that could be added to the path.
>
>
>
> *From:* users-bounces at lists.ironpython.com [mailto:
> users-bounces at lists.ironpython.com] *On Behalf Of *Dave Fugate
> *Sent:* Wednesday, December 10, 2008 1:19 PM
> *To:* Discussion of IronPython
> *Cc:* python-announce-list at python.org
> *Subject:* [IronPython] Announcing IronPython 2.0
>
>
>
> Hello Python Community,
>
>
>
> The IronPython and Dynamic Language Runtime teams are proud to announce the
> release of IronPython 2.0 final.  IronPython 2.0 is the culmination of
> nearly two years worth of work resulting in a CPython 2.5 compatible release
> on .NET 2.0 SP1.  By far, the biggest change to 2.0 is that our 1.1 codebase
> was refactored to run on top of the Dynamic Language Runtime<http://www.codeplex.com/dlr>.
> With this we automatically get improvements in many feature areas such as
> better .NET interop support and hosting Python from managed code.  There
> have been many other major improvements as well.  The most notable are:
>
> ?         An MSI installer for Windows platforms which includes parts of
> the CPython 2.5 standard library
>
> ?         IronPython assemblies targeting Silverlight and tools such as
> Chiron to improve the Silverlight dynamic development experience
>
> ?         The addition of more C-based standard modules such as *cmath*and
> *_winreg*
>
> ?         Significant improvements in importing compatibility and features
>
> ?         Distribution of IronPython under the Microsoft Public License
> which has been approved by OSI
>
> ?         Performance improvements.  On that note, a new Wiki page has
> been created for IronPython performance reports - see
> http://www.codeplex.com/IronPython/Wiki/View.aspx?title=IronPython%20Performance
>
> ?         Over 500 bugs have been closed in 2.0.  453 of these were
> reported on CodePlex
>
> ?         Support for precompilation of Python source files into a single
> dll
>
>
>
> This seems like an opportune time to remind everyone that we fix bugs based
> on the number of votes they have on CodePlex.  As we're planning on
> releasing IronPython 2.0.1 fairly soon, please vote for your favorite bugs
> at http://www.codeplex.com/IronPython/WorkItem/AdvancedList.aspx to help
> ensure they get fixed in time for the next release.
>
>
>
> We'd like to extend our gratitude to everyone in the IronPython community
> who reported bugs thereby making this a better release:  milind, romank,
> chadaustin, sjmachin, davidfraser, TimothyFitz, drewid, sanxiyn,
> bashmohandes, pobrien, perhaps, haypo, Undebtedly, ayarrow, tscottw_cp,
> rope, arman0, eshaish, nivaldo, fuzzyman, CurtHagenlocher, Eloff, brucec,
> py_sunil, jacobg23, mziller, beaugunderson, gbraad, Oceanborn, tarlano,
> jbevain, glchapman, anthonybaxter, jdhardy, jjlee, haibo, doubleyewdee,
> jackeyoo, whit537, sdahlbac, PeteHufnagel, jtenney, nriley, junfeng,
> grizlupo, rridge, lewisle, JoelBondurant, johnplatt, lthompson, debackerl,
> googen, tscottw, VoteFw, leppie, Qvin, heyssion2, CriGoT, baxeno, sbergman,
> Laurion, luntain, oldman, christmas, 05031972, kevgu, wilberforce,
> Korbinian, lclj, sorokerr, Eriol, tatwright, ais, TraumaPony, pelikhan,
> asafk, felixpollan, srid, atifaziz, vernondcole, fwereade, zpy, yanne,
> facorreia, Daneel, zvikag, psykotic, Cavingdeep, BEaton, sborde, orbital56,
> fbourgeois, antont, krosavcheg, ktc1, awilkins, ben2004uk, paulfelix, axl,
> JeffreySax, Lawouach, and KKI.
>
>
>
> You can download IronPython 2.0 at:
> http://www.codeplex.com/IronPython/Release/ProjectReleases.aspx?ReleaseId=8365
>
>
>
> The IronPython Team
>
>
>
>
>
> BTW
>
> Updates to the IronPython 1.0 samples will be released shortly.  Stay tuned
> to the IronPython mailing list for details.
>
>
>
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081223/0c0b0410/attachment.html>

From kfarmer at thuban.org  Tue Dec 23 21:25:12 2008
From: kfarmer at thuban.org (Keith J. Farmer)
Date: Tue, 23 Dec 2008 12:25:12 -0800
Subject: [IronPython] Announcing IronPython 2.0
References: <EECFB552375FDE49A23A13CE85EEC2FD1E9834A4D5@NA-EXMSG-C112.redmond.corp.microsoft.com><EEDCD954DD4A0D488BBDBDDCC260754466FB95@meta-exchange.circle-cross.home>
	<d2155e360812231157r1a20350bt91448ee135f90574@mail.gmail.com>
Message-ID: <EEDCD954DD4A0D488BBDBDDCC26075442CC94D@meta-exchange.circle-cross.home>

I recall this was a matter of some debate in the past, so I wasn't quite sure if it was *worth* adding :)
 
Of course, MSI was a big debate in the past as well...

________________________________

From: users-bounces at lists.ironpython.com on behalf of Curt Hagenlocher
Sent: Tue 12/23/2008 11:57 AM
To: Discussion of IronPython
Subject: Re: [IronPython] Announcing IronPython 2.0


As with all feature requests, please add it to CodePlex. :)


On Mon, Dec 22, 2008 at 5:02 PM, Keith J. Farmer <kfarmer at thuban.org> wrote:


	Since there's an MSI now, could we hope for updates/new-major-versions via Windows Update?  Like online banking, I've come to rely on things updating themselves, rather than having to keep track of various support download sites.  Not to be exclusive - I would expect downloads at CodePlex to remain available -- but for the ease of installation and maintainance, it's hard to beat built-in, unified updaters.

	 

	If side-by-side versioning is required, I could see a current-version environment variable that could be added to the path.

	 

	From: users-bounces at lists.ironpython.com [mailto:users-bounces at lists.ironpython.com] On Behalf Of Dave Fugate
	Sent: Wednesday, December 10, 2008 1:19 PM
	To: Discussion of IronPython
	Cc: python-announce-list at python.org
	Subject: [IronPython] Announcing IronPython 2.0

	 

	Hello Python Community,

	 

	The IronPython and Dynamic Language Runtime teams are proud to announce the release of IronPython 2.0 final.  IronPython 2.0 is the culmination of nearly two years worth of work resulting in a CPython 2.5 compatible release on .NET 2.0 SP1.  By far, the biggest change to 2.0 is that our 1.1 codebase was refactored to run on top of the Dynamic Language Runtime <http://www.codeplex.com/dlr> .  With this we automatically get improvements in many feature areas such as better .NET interop support and hosting Python from managed code.  There have been many other major improvements as well.  The most notable are:

	?         An MSI installer for Windows platforms which includes parts of the CPython 2.5 standard library

	?         IronPython assemblies targeting Silverlight and tools such as Chiron to improve the Silverlight dynamic development experience

	?         The addition of more C-based standard modules such as cmath and _winreg

	?         Significant improvements in importing compatibility and features

	?         Distribution of IronPython under the Microsoft Public License which has been approved by OSI

	?         Performance improvements.  On that note, a new Wiki page has been created for IronPython performance reports - see http://www.codeplex.com/IronPython/Wiki/View.aspx?title=IronPython%20Performance

	?         Over 500 bugs have been closed in 2.0.  453 of these were reported on CodePlex

	?         Support for precompilation of Python source files into a single dll

	 

	This seems like an opportune time to remind everyone that we fix bugs based on the number of votes they have on CodePlex.  As we're planning on releasing IronPython 2.0.1 fairly soon, please vote for your favorite bugs at http://www.codeplex.com/IronPython/WorkItem/AdvancedList.aspx to help ensure they get fixed in time for the next release.

	 

	We'd like to extend our gratitude to everyone in the IronPython community who reported bugs thereby making this a better release:  milind, romank, chadaustin, sjmachin, davidfraser, TimothyFitz, drewid, sanxiyn, bashmohandes, pobrien, perhaps, haypo, Undebtedly, ayarrow, tscottw_cp, rope, arman0, eshaish, nivaldo, fuzzyman, CurtHagenlocher, Eloff, brucec, py_sunil, jacobg23, mziller, beaugunderson, gbraad, Oceanborn, tarlano, jbevain, glchapman, anthonybaxter, jdhardy, jjlee, haibo, doubleyewdee, jackeyoo, whit537, sdahlbac, PeteHufnagel, jtenney, nriley, junfeng, grizlupo, rridge, lewisle, JoelBondurant, johnplatt, lthompson, debackerl, googen, tscottw, VoteFw, leppie, Qvin, heyssion2, CriGoT, baxeno, sbergman, Laurion, luntain, oldman, christmas, 05031972, kevgu, wilberforce, Korbinian, lclj, sorokerr, Eriol, tatwright, ais, TraumaPony, pelikhan, asafk, felixpollan, srid, atifaziz, vernondcole, fwereade, zpy, yanne, facorreia, Daneel, zvikag, psykotic, Cavingdeep, BEaton, sborde, orbital56, fbourgeois, antont, krosavcheg, ktc1, awilkins, ben2004uk, paulfelix, axl, JeffreySax, Lawouach, and KKI.

	 

	You can download IronPython 2.0 at:  http://www.codeplex.com/IronPython/Release/ProjectReleases.aspx?ReleaseId=8365

	 

	The IronPython Team

	 

	 

	BTW

	Updates to the IronPython 1.0 samples will be released shortly.  Stay tuned to the IronPython mailing list for details.

	 


	_______________________________________________
	Users mailing list
	Users at lists.ironpython.com
	http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
	
	


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081223/7d6835d8/attachment.html>

From russomf at hotmail.com  Tue Dec 23 21:55:01 2008
From: russomf at hotmail.com (Mark Russo)
Date: Tue, 23 Dec 2008 20:55:01 +0000
Subject: [IronPython] Is this a bug in IP 2.0?
Message-ID: <BAY106-W43B149C83E051C333BC137ACEC0@phx.gbl>


In IronPython 2.0, when I replace a Point object in a list using assignment, a copy of the Point object is used to replace the original list element rather than the original Point itself. This does not happen with user-defined objects. It also does not happen with Point objects in IronPython 1.1.

IronPython 2.0 (2.0.0.0) on .NET 2.0.50727.3053
Type "help", "copyright", "credits" or "license" for more information.
>>> import clr
>>> clr.AddReference("System.Drawing")
>>> from System.Drawing import *
>>>
>>> class C(object): pass
...
>>> def test1():
...     p1 = C()                # Create two objects
...     p2 = C()
...     l = [p1]                # Put one in a list
...     print id(l[-1]), id(p2) # Print ids of object in list (p1) and p2
...     l[-1] = p2              # Replace p1 with p2 in list
...     print id(l[-1]), id(p2) # Print ids of object in list and p2
...
>>> def test2():
...     p1 = Point(1,2)         # Create two Points
...     p2 = Point(3,4)
...     l = [p1]                # Put one in a list
...     print id(l[-1]), id(p2) # Print ids of Point in list (p1) and p2
...     l[-1] = p2              # Replace p1 with p2 in list
...     print id(l[-1]), id(p2) # Print ids of Point in list and p2
...
>>> test1()
43 44
44 44
>>> test2()
45 46
47 46   # The id of the Point before and after adding to the list is different
>>>

Here is the same program in IronPython 1.1

IronPython 1.1 (1.1) on .NET 2.0.50727.3053
Copyright (c) Microsoft Corporation. All rights reserved.

<snip>

>>> test1()
43 44
44 44
>>> test2()
45 46
46 46   # The id of the Point before and after adding to the list is the same
>>>

... or am I missing something.


_________________________________________________________________
Send e-mail faster without improving your typing skills.
http://windowslive.com/online/hotmail?ocid=TXT_TAGLM_WL_hotmail_acq_speed_122008
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081223/adc553c5/attachment.html>

From fuzzyman at voidspace.org.uk  Tue Dec 23 22:01:21 2008
From: fuzzyman at voidspace.org.uk (Michael Foord)
Date: Tue, 23 Dec 2008 21:01:21 +0000
Subject: [IronPython] Is this a bug in IP 2.0?
In-Reply-To: <BAY106-W43B149C83E051C333BC137ACEC0@phx.gbl>
References: <BAY106-W43B149C83E051C333BC137ACEC0@phx.gbl>
Message-ID: <495151A1.4000405@voidspace.org.uk>

Mark Russo wrote:
> In IronPython 2.0, when I replace a Point object in a list using 
> assignment, a copy of the Point object is used to replace the original 
> list element rather than the original Point itself. This does not 
> happen with user-defined objects. It also does not happen with Point 
> objects in IronPython 1.1.

Point objects are value types, so some operations cause them to be 
copied. Probably the way they are used under the hood has changed with 
IronPython 2.

Michael

>
> IronPython 2.0 (2.0.0.0) on .NET 2.0.50727.3053
> Type "help", "copyright", "credits" or "license" for more information.
> >>> import clr
> >>> clr.AddReference("System.Drawing")
> >>> from System.Drawing import *
> >>>
> >>> class C(object): pass
> ...
> >>> def test1():
> ...     p1 = C()                # Create two objects
> ...     p2 = C()
> ...     l = [p1]                # Put one in a list
> ...     print id(l[-1]), id(p2) # Print ids of object in list (p1) and p2
> ...     l[-1] = p2              # Replace p1 with p2 in list
> ...     print id(l[-1]), id(p2) # Print ids of object in list and p2
> ...
> >>> def test2():
> ...     p1 = Point(1,2)         # Create two Points
> ...     p2 = Point(3,4)
> ...     l = [p1]                # Put one in a list
> ...     print id(l[-1]), id(p2) # Print ids of Point in list (p1) and p2
> ...     l[-1] = p2              # Replace p1 with p2 in list
> ...     print id(l[-1]), id(p2) # Print ids of Point in list and p2
> ...
> >>> test1()
> 43 44
> 44 44
> >>> test2()
> 45 46
> 47 46   # The id of the Point before and after adding to the list is 
> different
> >>>
>
> Here is the same program in IronPython 1.1
>
> IronPython 1.1 (1.1) on .NET 2.0.50727.3053
> Copyright (c) Microsoft Corporation. All rights reserved.
>
> <snip>
>
> >>> test1()
> 43 44
> 44 44
> >>> test2()
> 45 46
> 46 46   # The id of the Point before and after adding to the list is 
> the same
> >>>
>
> ... or am I missing something.
>
>
> ------------------------------------------------------------------------
> Send e-mail faster without improving your typing skills. Get your 
> Hotmail? account. 
> <http://windowslive.com/online/hotmail?ocid=TXT_TAGLM_WL_hotmail_acq_speed_122008> 
>
> ------------------------------------------------------------------------
>
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>   


-- 
http://www.ironpythoninaction.com/
http://www.voidspace.org.uk/blog




From kfarmer at thuban.org  Tue Dec 23 23:46:40 2008
From: kfarmer at thuban.org (Keith J. Farmer)
Date: Tue, 23 Dec 2008 14:46:40 -0800
Subject: [IronPython] Announcing IronPython 2.0
References: <EECFB552375FDE49A23A13CE85EEC2FD1E9834A4D5@NA-EXMSG-C112.redmond.corp.microsoft.com><EEDCD954DD4A0D488BBDBDDCC260754466FB95@meta-exchange.circle-cross.home><d2155e360812231157r1a20350bt91448ee135f90574@mail.gmail.com>
	<EEDCD954DD4A0D488BBDBDDCC26075442CC94D@meta-exchange.circle-cross.home>
Message-ID: <EEDCD954DD4A0D488BBDBDDCC26075442CC950@meta-exchange.circle-cross.home>

http://www.codeplex.com/IronPython/WorkItem/View.aspx?WorkItemId=20344
 
CodePlex's TFS server seems to be down suddenly.. it's a conspiracy.

________________________________

From: users-bounces at lists.ironpython.com on behalf of Keith J. Farmer
Sent: Tue 12/23/2008 12:25 PM
To: Discussion of IronPython; Discussion of IronPython
Subject: Re: [IronPython] Announcing IronPython 2.0


I recall this was a matter of some debate in the past, so I wasn't quite sure if it was *worth* adding :)
 
Of course, MSI was a big debate in the past as well...

________________________________

From: users-bounces at lists.ironpython.com on behalf of Curt Hagenlocher
Sent: Tue 12/23/2008 11:57 AM
To: Discussion of IronPython
Subject: Re: [IronPython] Announcing IronPython 2.0


As with all feature requests, please add it to CodePlex. :)


On Mon, Dec 22, 2008 at 5:02 PM, Keith J. Farmer <kfarmer at thuban.org> wrote:


	Since there's an MSI now, could we hope for updates/new-major-versions via Windows Update?  Like online banking, I've come to rely on things updating themselves, rather than having to keep track of various support download sites.  Not to be exclusive - I would expect downloads at CodePlex to remain available -- but for the ease of installation and maintainance, it's hard to beat built-in, unified updaters.

	 

	If side-by-side versioning is required, I could see a current-version environment variable that could be added to the path.

	 

	From: users-bounces at lists.ironpython.com [mailto:users-bounces at lists.ironpython.com] On Behalf Of Dave Fugate
	Sent: Wednesday, December 10, 2008 1:19 PM
	To: Discussion of IronPython
	Cc: python-announce-list at python.org
	Subject: [IronPython] Announcing IronPython 2.0

	 

	Hello Python Community,

	 

	The IronPython and Dynamic Language Runtime teams are proud to announce the release of IronPython 2.0 final.  IronPython 2.0 is the culmination of nearly two years worth of work resulting in a CPython 2.5 compatible release on .NET 2.0 SP1.  By far, the biggest change to 2.0 is that our 1.1 codebase was refactored to run on top of the Dynamic Language Runtime <http://www.codeplex.com/dlr> .  With this we automatically get improvements in many feature areas such as better .NET interop support and hosting Python from managed code.  There have been many other major improvements as well.  The most notable are:

	?         An MSI installer for Windows platforms which includes parts of the CPython 2.5 standard library

	?         IronPython assemblies targeting Silverlight and tools such as Chiron to improve the Silverlight dynamic development experience

	?         The addition of more C-based standard modules such as cmath and _winreg

	?         Significant improvements in importing compatibility and features

	?         Distribution of IronPython under the Microsoft Public License which has been approved by OSI

	?         Performance improvements.  On that note, a new Wiki page has been created for IronPython performance reports - see http://www.codeplex.com/IronPython/Wiki/View.aspx?title=IronPython%20Performance

	?         Over 500 bugs have been closed in 2.0.  453 of these were reported on CodePlex

	?         Support for precompilation of Python source files into a single dll

	 

	This seems like an opportune time to remind everyone that we fix bugs based on the number of votes they have on CodePlex.  As we're planning on releasing IronPython 2.0.1 fairly soon, please vote for your favorite bugs at http://www.codeplex.com/IronPython/WorkItem/AdvancedList.aspx to help ensure they get fixed in time for the next release.

	 

	We'd like to extend our gratitude to everyone in the IronPython community who reported bugs thereby making this a better release:  milind, romank, chadaustin, sjmachin, davidfraser, TimothyFitz, drewid, sanxiyn, bashmohandes, pobrien, perhaps, haypo, Undebtedly, ayarrow, tscottw_cp, rope, arman0, eshaish, nivaldo, fuzzyman, CurtHagenlocher, Eloff, brucec, py_sunil, jacobg23, mziller, beaugunderson, gbraad, Oceanborn, tarlano, jbevain, glchapman, anthonybaxter, jdhardy, jjlee, haibo, doubleyewdee, jackeyoo, whit537, sdahlbac, PeteHufnagel, jtenney, nriley, junfeng, grizlupo, rridge, lewisle, JoelBondurant, johnplatt, lthompson, debackerl, googen, tscottw, VoteFw, leppie, Qvin, heyssion2, CriGoT, baxeno, sbergman, Laurion, luntain, oldman, christmas, 05031972, kevgu, wilberforce, Korbinian, lclj, sorokerr, Eriol, tatwright, ais, TraumaPony, pelikhan, asafk, felixpollan, srid, atifaziz, vernondcole, fwereade, zpy, yanne, facorreia, Daneel, zvikag, psykotic, Cavingdeep, BEaton, sborde, orbital56, fbourgeois, antont, krosavcheg, ktc1, awilkins, ben2004uk, paulfelix, axl, JeffreySax, Lawouach, and KKI.

	 

	You can download IronPython 2.0 at:  http://www.codeplex.com/IronPython/Release/ProjectReleases.aspx?ReleaseId=8365

	 

	The IronPython Team

	 

	 

	BTW

	Updates to the IronPython 1.0 samples will be released shortly.  Stay tuned to the IronPython mailing list for details.

	 


	_______________________________________________
	Users mailing list
	Users at lists.ironpython.com
	http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
	
	


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081223/1c621929/attachment.html>

From empirebuilder at gmail.com  Wed Dec 24 00:15:49 2008
From: empirebuilder at gmail.com (Dody Gunawinata)
Date: Wed, 24 Dec 2008 00:15:49 +0100
Subject: [IronPython] TCL for .Net
Message-ID: <8cd017b80812231515gf149a65m1dc5549e4971eb57@mail.gmail.com>

Check this out
http://www.codeplex.com/eagle/Release/ProjectReleases.aspx?ReleaseId=18226#ReleaseFiles
It looks very interesting.

Dody G.

-- 
nomadlife.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081224/d5341077/attachment.html>

From fuzzyman at voidspace.org.uk  Wed Dec 24 00:19:15 2008
From: fuzzyman at voidspace.org.uk (Michael Foord)
Date: Tue, 23 Dec 2008 23:19:15 +0000
Subject: [IronPython] TCL for .Net
In-Reply-To: <8cd017b80812231515gf149a65m1dc5549e4971eb57@mail.gmail.com>
References: <8cd017b80812231515gf149a65m1dc5549e4971eb57@mail.gmail.com>
Message-ID: <495171F3.4040005@voidspace.org.uk>

Dody Gunawinata wrote:
> Check this 
> out http://www.codeplex.com/eagle/Release/ProjectReleases.aspx?ReleaseId=18226#ReleaseFiles 
>
>
> It looks very interesting.

Does look interesting. It would be even more interesting if it was DLR 
based. :-)

Michael

>
> Dody G.
>
> -- 
> nomadlife.org <http://nomadlife.org>
>
>
> ------------------------------------------------------------------------
>
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>   


-- 
http://www.ironpythoninaction.com/
http://www.voidspace.org.uk/blog




From empirebuilder at gmail.com  Wed Dec 24 00:33:58 2008
From: empirebuilder at gmail.com (Dody Gunawinata)
Date: Wed, 24 Dec 2008 00:33:58 +0100
Subject: [IronPython] TCL for .Net
In-Reply-To: <495171F3.4040005@voidspace.org.uk>
References: <8cd017b80812231515gf149a65m1dc5549e4971eb57@mail.gmail.com>
	<495171F3.4040005@voidspace.org.uk>
Message-ID: <8cd017b80812231533u676b4790k725a0be18b79bd24@mail.gmail.com>

It's not. Only IronScheme is the other outsider DLR based implementation
language.

On Wed, Dec 24, 2008 at 12:19 AM, Michael Foord
<fuzzyman at voidspace.org.uk>wrote:

> Dody Gunawinata wrote:
>
>> Check this out
>> http://www.codeplex.com/eagle/Release/ProjectReleases.aspx?ReleaseId=18226#ReleaseFiles
>>
>> It looks very interesting.
>>
>
> Does look interesting. It would be even more interesting if it was DLR
> based. :-)
>
> Michael
>
>
>> Dody G.
>>
>> --
>> nomadlife.org <http://nomadlife.org>
>>
>>
>> ------------------------------------------------------------------------
>>
>> _______________________________________________
>> Users mailing list
>> Users at lists.ironpython.com
>> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>>
>>
>
>
> --
> http://www.ironpythoninaction.com/
> http://www.voidspace.org.uk/blog
>
>
>


-- 
nomadlife.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081224/39fba506/attachment.html>

From jdhardy at gmail.com  Wed Dec 24 01:56:03 2008
From: jdhardy at gmail.com (Jeff Hardy)
Date: Tue, 23 Dec 2008 17:56:03 -0700
Subject: [IronPython] Detecting IronPython Installation Directory
In-Reply-To: <d2155e360812231153y6ba3a0fw7cd152cd0f17cb5f@mail.gmail.com>
References: <ea89d0e50812221810y5086f53ay8104c7c98e03781e@mail.gmail.com>
	<d2155e360812230928v116fcc4eua7564c7bd5e1af99@mail.gmail.com>
	<A5DE25409C1E51458D103EDB73EA953A79B0B9D4CF@NA-EXMSG-C104.redmond.corp.microsoft.com>
	<d2155e360812231021j58254485ob7b3ec5861fea9e2@mail.gmail.com>
	<ea89d0e50812231140j592875d2u3eb2618bea397a8f@mail.gmail.com>
	<d2155e360812231153y6ba3a0fw7cd152cd0f17cb5f@mail.gmail.com>
Message-ID: <ea89d0e50812231656t48b46c36p658b566586d6ac7c@mail.gmail.com>

On Tue, Dec 23, 2008 at 12:53 PM, Curt Hagenlocher <curt at hagenlocher.org> wrote:
> Given that there are clean APIs for pulling the data out, what are your
> objections to using the installer data?

Oh, nothing too convincing - just an irrational aversion to using
"sekrit" information. Also, the WMI query is a bit slow.
But, as you said, the APIs are stable, so I can work with it.

- Jeff


From timr at probo.com  Wed Dec 24 19:05:20 2008
From: timr at probo.com (Tim Roberts)
Date: Wed, 24 Dec 2008 10:05:20 -0800
Subject: [IronPython] Is this a bug in IP 2.0?
In-Reply-To: <mailman.1.1230066797.29013.users-ironpython.com@lists.ironpython.com>
References: <mailman.1.1230066797.29013.users-ironpython.com@lists.ironpython.com>
Message-ID: <495279E0.4040400@probo.com>

On Tue, 23 Dec 2008 21:01:21 +0000, Michael Foord
<fuzzyman at voidspace.org.uk> wrote:
> Mark Russo wrote:
>   
>> In IronPython 2.0, when I replace a Point object in a list using 
>> assignment, a copy of the Point object is used to replace the original 
>> list element rather than the original Point itself. This does not 
>> happen with user-defined objects. It also does not happen with Point 
>> objects in IronPython 1.1.
>>     
>
> Point objects are value types, so some operations cause them to be 
> copied. Probably the way they are used under the hood has changed with 
> IronPython 2.
>   

But, this has to be considered a rather serious bug, doesn't it?  If I
create an object p2, then insert it into a list at l[0], the Python
contract guarantees that l[0] and p2 are references to the same object,
so that modifying one must modify the other.

Am I missing the point here?  Is there another level of indirection in
the implementation that means these really are references to the same
object, despite having different IDs?

-- 
Tim Roberts, timr at probo.com
Providenza & Boekelheide, Inc.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081224/b9fe18bf/attachment.html>

From curt at hagenlocher.org  Wed Dec 24 19:29:13 2008
From: curt at hagenlocher.org (Curt Hagenlocher)
Date: Wed, 24 Dec 2008 10:29:13 -0800
Subject: [IronPython] Is this a bug in IP 2.0?
In-Reply-To: <495279E0.4040400@probo.com>
References: <mailman.1.1230066797.29013.users-ironpython.com@lists.ironpython.com>
	<495279E0.4040400@probo.com>
Message-ID: <d2155e360812241029g7e4415e7la69695c7c98a4b02@mail.gmail.com>

On Wed, Dec 24, 2008 at 10:05 AM, Tim Roberts <timr at probo.com> wrote:

>
> But, this has to be considered a rather serious bug, doesn't it?  If I
> create an object p2, then insert it into a list at l[0], the Python contract
> guarantees that l[0] and p2 are references to the same object, so that
> modifying one must modify the other.
>

CLR value types simply don't map well onto Python objects, because CPython
doesn't have any value types.  All objects are basically references.  In the
CLR, we convert value types into reference types by "boxing" them, and it's
only in this state that they have an "identity".  When a value type is
unboxed and reboxed, it acquires a whole new identity, and therefore a new
"id".  If you're familiar with the CPython "C" language API, you can think
of this as the distinction between the numeric value stored inside an "int"
PyObject, and the PyObject itself.

I believe that the current "best practices" for CLR value types are to
encourage people to treat them as immutable.  If you do this, the lack of
reference equality is no longer quite as important :).

I don't know whether or not the specific behavior you've identified is a
bug, but you should definitely enter it into CodePlex so that it can be
tracked and investigated properly.

--
Curt Hagenlocher
curt at hagenlocher.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081224/c81fe617/attachment.html>

From curt at hagenlocher.org  Wed Dec 24 19:30:41 2008
From: curt at hagenlocher.org (Curt Hagenlocher)
Date: Wed, 24 Dec 2008 10:30:41 -0800
Subject: [IronPython] Is this a bug in IP 2.0?
In-Reply-To: <d2155e360812241029g7e4415e7la69695c7c98a4b02@mail.gmail.com>
References: <mailman.1.1230066797.29013.users-ironpython.com@lists.ironpython.com>
	<495279E0.4040400@probo.com>
	<d2155e360812241029g7e4415e7la69695c7c98a4b02@mail.gmail.com>
Message-ID: <d2155e360812241030h2db42a6di995bef454e785125@mail.gmail.com>

On Wed, Dec 24, 2008 at 10:29 AM, Curt Hagenlocher <curt at hagenlocher.org>wrote:

>
> I don't know whether or not the specific behavior you've identified is a
> bug, but you should definitely enter it into CodePlex so that it can be
> tracked and investigated properly.
>

Oh, sorry -- didn't see that you weren't the original poster.  Please
consider "you" to mean Mark in this sentence.

--
Curt Hagenlocher
curt at hagenlocher.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081224/ce40f8c6/attachment.html>

From chenrong2003 at gmail.com  Thu Dec 25 04:18:34 2008
From: chenrong2003 at gmail.com (=?ISO-2022-JP?B?TmVpbCgbJEJMWkxuOFEbKEIp?=)
Date: Thu, 25 Dec 2008 11:18:34 +0800
Subject: [IronPython] ANN : Davy's IronPython Editor
In-Reply-To: <4950CE1F.7060308@voidspace.org.uk>
References: <20253b0c0812200322w4734bf90u6947528ce97e92c8@mail.gmail.com>
	<4950CE1F.7060308@voidspace.org.uk>
Message-ID: <26756bf60812241918g62e13ebew4aa56301ab9710e5@mail.gmail.com>

it's a good tool but in my opinion the name 'DIE' is not good, isn't it?


From christian2.schmidt at gmx.de  Thu Dec 25 14:37:55 2008
From: christian2.schmidt at gmx.de (Christian Schmidt)
Date: Thu, 25 Dec 2008 14:37:55 +0100
Subject: [IronPython] Memory Leak in CreateScriptSource?
Message-ID: <49538CB3.70706@gmx.de>

Hi,

the following code snippet allocates constantly more and more memory. 
Moving .Compile() outside the loop works as expected.

private static void LeakTest()
{
   var runtime = Python.CreateRuntime();
   var engine = runtime.GetEngine("py");
   var sum = 0.0;

   while (true)
   {
     var code = engine.CreateScriptSourceFromString("1.0+2.0").Compile();
     sum += code.Execute<double>();
     GC.Collect();
     GC.WaitForPendingFinalizers();
     GC.Collect();
     Console.WriteLine(sum);
   }
}

Any ideas?

Merry Christmas,

Christian


From empirebuilder at gmail.com  Fri Dec 26 16:31:02 2008
From: empirebuilder at gmail.com (Dody Gunawinata)
Date: Fri, 26 Dec 2008 17:31:02 +0200
Subject: [IronPython] Memory Leak in CreateScriptSource?
In-Reply-To: <49538CB3.70706@gmx.de>
References: <49538CB3.70706@gmx.de>
Message-ID: <8cd017b80812260731r44d7b4d2h9f8ae06a79e0459d@mail.gmail.com>

I run above code in a console using IronPython 2.0 Final Version and here is
the output (on Vista) using Tasklist command. The final result doubles the
in memory size; I stopped the measurement when it hits 12,000

Image Name                     PID Session Name        Session#    Mem Usage
========================= ======== ================ =========== ============
IronPythonLeaker.exe          5248 Console                    1     33,532 K

Image Name                     PID Session Name        Session#    Mem Usage
========================= ======== ================ =========== ============
IronPythonLeaker.exe          5248 Console                    1     39,204 K

Image Name                     PID Session Name        Session#    Mem Usage
========================= ======== ================ =========== ============
IronPythonLeaker.exe          5248 Console                    1     43,692 K

Image Name                     PID Session Name        Session#    Mem Usage
========================= ======== ================ =========== ============
IronPythonLeaker.exe          5248 Console                    1     42,952 K

Image Name                     PID Session Name        Session#    Mem Usage
========================= ======== ================ =========== ============
IronPythonLeaker.exe          5248 Console                    1     45,888 K

Image Name                     PID Session Name        Session#    Mem Usage
========================= ======== ================ =========== ============
IronPythonLeaker.exe          5248 Console                    1     45,928 K

Image Name                     PID Session Name        Session#    Mem Usage
========================= ======== ================ =========== ============
IronPythonLeaker.exe          5248 Console                    1     45,956 K

Image Name                     PID Session Name        Session#    Mem Usage
========================= ======== ================ =========== ============
IronPythonLeaker.exe          5248 Console                    1     44,992 K

Image Name                     PID Session Name        Session#    Mem Usage
========================= ======== ================ =========== ============
IronPythonLeaker.exe          5248 Console                    1     49,520 K

Image Name                     PID Session Name        Session#    Mem Usage
========================= ======== ================ =========== ============
IronPythonLeaker.exe          5248 Console                    1     50,592 K

Image Name                     PID Session Name        Session#    Mem Usage
========================= ======== ================ =========== ============
IronPythonLeaker.exe          5248 Console                    1     50,636 K

Image Name                     PID Session Name        Session#    Mem Usage
========================= ======== ================ =========== ============
IronPythonLeaker.exe          5248 Console                    1     52,632 K

Image Name                     PID Session Name        Session#    Mem Usage
========================= ======== ================ =========== ============
IronPythonLeaker.exe          5248 Console                    1     52,788 K

Image Name                     PID Session Name        Session#    Mem Usage
========================= ======== ================ =========== ============
IronPythonLeaker.exe          5248 Console                    1     52,884 K

Image Name                     PID Session Name        Session#    Mem Usage
========================= ======== ================ =========== ============
IronPythonLeaker.exe          5248 Console                    1     53,032 K

Image Name                     PID Session Name        Session#    Mem Usage
========================= ======== ================ =========== ============
IronPythonLeaker.exe          5248 Console                    1     62,916 K

Image Name                     PID Session Name        Session#    Mem Usage
========================= ======== ================ =========== ============
IronPythonLeaker.exe          5248 Console                    1     62,456 K

Image Name                     PID Session Name        Session#    Mem Usage
========================= ======== ================ =========== ============
IronPythonLeaker.exe          5248 Console                    1     67,668 K

Image Name                     PID Session Name        Session#    Mem Usage
========================= ======== ================ =========== ============
IronPythonLeaker.exe          5248 Console                    1     65,880 K

Image Name                     PID Session Name        Session#    Mem Usage
========================= ======== ================ =========== ============
IronPythonLeaker.exe          5248 Console                    1     72,260 K

Image Name                     PID Session Name        Session#    Mem Usage
========================= ======== ================ =========== ============
IronPythonLeaker.exe          5248 Console                    1     73,644 K


On Thu, Dec 25, 2008 at 3:37 PM, Christian Schmidt <
christian2.schmidt at gmx.de> wrote:

> Hi,
>
> the following code snippet allocates constantly more and more memory.
> Moving .Compile() outside the loop works as expected.
>
> private static void LeakTest()
> {
>  var runtime = Python.CreateRuntime();
>  var engine = runtime.GetEngine("py");
>  var sum = 0.0;
>
>  while (true)
>  {
>    var code = engine.CreateScriptSourceFromString("1.0+2.0").Compile();
>    sum += code.Execute<double>();
>    GC.Collect();
>    GC.WaitForPendingFinalizers();
>    GC.Collect();
>    Console.WriteLine(sum);
>  }
> }
>
> Any ideas?
>
> Merry Christmas,
>
> Christian
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>



-- 
nomadlife.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081226/ab3bb844/attachment.html>

From empirebuilder at gmail.com  Fri Dec 26 16:36:37 2008
From: empirebuilder at gmail.com (Dody Gunawinata)
Date: Fri, 26 Dec 2008 17:36:37 +0200
Subject: [IronPython] Memory Leak in CreateScriptSource?
In-Reply-To: <8cd017b80812260731r44d7b4d2h9f8ae06a79e0459d@mail.gmail.com>
References: <49538CB3.70706@gmx.de>
	<8cd017b80812260731r44d7b4d2h9f8ae06a79e0459d@mail.gmail.com>
Message-ID: <8cd017b80812260736j3ed4927bw5f1209f93370d141@mail.gmail.com>

With the GC.Collect lines removed, this is how it looks like (after counting
to 40,000). I started the tasklist command after around 25K.

Image Name                     PID Session Name        Session#    Mem Usage
========================= ======== ================ =========== ============
IronPythonLeaker.exe          5296 Console                    1    146,840 K

Image Name                     PID Session Name        Session#    Mem Usage
========================= ======== ================ =========== ============
IronPythonLeaker.exe          5296 Console                    1    147,948 K

Image Name                     PID Session Name        Session#    Mem Usage
========================= ======== ================ =========== ============
IronPythonLeaker.exe          5296 Console                    1    148,344 K

Image Name                     PID Session Name        Session#    Mem Usage
========================= ======== ================ =========== ============
IronPythonLeaker.exe          5296 Console                    1    153,616 K

Image Name                     PID Session Name        Session#    Mem Usage
========================= ======== ================ =========== ============
IronPythonLeaker.exe          5296 Console                    1    154,064 K

Image Name                     PID Session Name        Session#    Mem Usage
========================= ======== ================ =========== ============
IronPythonLeaker.exe          5296 Console                    1    156,364 K

Image Name                     PID Session Name        Session#    Mem Usage
========================= ======== ================ =========== ============
IronPythonLeaker.exe          5296 Console                    1    160,612 K

Image Name                     PID Session Name        Session#    Mem Usage
========================= ======== ================ =========== ============
IronPythonLeaker.exe          5296 Console                    1    174,568 K

Image Name                     PID Session Name        Session#    Mem Usage
========================= ======== ================ =========== ============
IronPythonLeaker.exe          5296 Console                    1    180,232 K

Image Name                     PID Session Name        Session#    Mem Usage
========================= ======== ================ =========== ============
IronPythonLeaker.exe          5296 Console                    1    189,904 K

Image Name                     PID Session Name        Session#    Mem Usage
========================= ======== ================ =========== ============
IronPythonLeaker.exe          5296 Console                    1    192,856 K

Image Name                     PID Session Name        Session#    Mem Usage
========================= ======== ================ =========== ============
IronPythonLeaker.exe          5296 Console                    1    198,900 K

Image Name                     PID Session Name        Session#    Mem Usage
========================= ======== ================ =========== ============
IronPythonLeaker.exe          5296 Console                    1    201,572 K

Image Name                     PID Session Name        Session#    Mem Usage
========================= ======== ================ =========== ============
IronPythonLeaker.exe          5296 Console                    1    201,604 K


On Fri, Dec 26, 2008 at 5:31 PM, Dody Gunawinata <empirebuilder at gmail.com>wrote:

> I run above code in a console using IronPython 2.0 Final Version and here
> is the output (on Vista) using Tasklist command. The final result doubles
> the in memory size; I stopped the measurement when it hits 12,000
>
> Image Name                     PID Session Name        Session#    Mem
> Usage
> ========================= ======== ================ ===========
> ============
> IronPythonLeaker.exe          5248 Console                    1     33,532
> K
>
> Image Name                     PID Session Name        Session#    Mem
> Usage
> ========================= ======== ================ ===========
> ============
> IronPythonLeaker.exe          5248 Console                    1     39,204
> K
>
> Image Name                     PID Session Name        Session#    Mem
> Usage
> ========================= ======== ================ ===========
> ============
> IronPythonLeaker.exe          5248 Console                    1     43,692
> K
>
> Image Name                     PID Session Name        Session#    Mem
> Usage
> ========================= ======== ================ ===========
> ============
> IronPythonLeaker.exe          5248 Console                    1     42,952
> K
>
> Image Name                     PID Session Name        Session#    Mem
> Usage
> ========================= ======== ================ ===========
> ============
> IronPythonLeaker.exe          5248 Console                    1     45,888
> K
>
> Image Name                     PID Session Name        Session#    Mem
> Usage
> ========================= ======== ================ ===========
> ============
> IronPythonLeaker.exe          5248 Console                    1     45,928
> K
>
> Image Name                     PID Session Name        Session#    Mem
> Usage
> ========================= ======== ================ ===========
> ============
> IronPythonLeaker.exe          5248 Console                    1     45,956
> K
>
> Image Name                     PID Session Name        Session#    Mem
> Usage
> ========================= ======== ================ ===========
> ============
> IronPythonLeaker.exe          5248 Console                    1     44,992
> K
>
> Image Name                     PID Session Name        Session#    Mem
> Usage
> ========================= ======== ================ ===========
> ============
> IronPythonLeaker.exe          5248 Console                    1     49,520
> K
>
> Image Name                     PID Session Name        Session#    Mem
> Usage
> ========================= ======== ================ ===========
> ============
> IronPythonLeaker.exe          5248 Console                    1     50,592
> K
>
> Image Name                     PID Session Name        Session#    Mem
> Usage
> ========================= ======== ================ ===========
> ============
> IronPythonLeaker.exe          5248 Console                    1     50,636
> K
>
> Image Name                     PID Session Name        Session#    Mem
> Usage
> ========================= ======== ================ ===========
> ============
> IronPythonLeaker.exe          5248 Console                    1     52,632
> K
>
> Image Name                     PID Session Name        Session#    Mem
> Usage
> ========================= ======== ================ ===========
> ============
> IronPythonLeaker.exe          5248 Console                    1     52,788
> K
>
> Image Name                     PID Session Name        Session#    Mem
> Usage
> ========================= ======== ================ ===========
> ============
> IronPythonLeaker.exe          5248 Console                    1     52,884
> K
>
> Image Name                     PID Session Name        Session#    Mem
> Usage
> ========================= ======== ================ ===========
> ============
> IronPythonLeaker.exe          5248 Console                    1     53,032
> K
>
> Image Name                     PID Session Name        Session#    Mem
> Usage
> ========================= ======== ================ ===========
> ============
> IronPythonLeaker.exe          5248 Console                    1     62,916
> K
>
> Image Name                     PID Session Name        Session#    Mem
> Usage
> ========================= ======== ================ ===========
> ============
> IronPythonLeaker.exe          5248 Console                    1     62,456
> K
>
> Image Name                     PID Session Name        Session#    Mem
> Usage
> ========================= ======== ================ ===========
> ============
> IronPythonLeaker.exe          5248 Console                    1     67,668
> K
>
> Image Name                     PID Session Name        Session#    Mem
> Usage
> ========================= ======== ================ ===========
> ============
> IronPythonLeaker.exe          5248 Console                    1     65,880
> K
>
> Image Name                     PID Session Name        Session#    Mem
> Usage
> ========================= ======== ================ ===========
> ============
> IronPythonLeaker.exe          5248 Console                    1     72,260
> K
>
> Image Name                     PID Session Name        Session#    Mem
> Usage
> ========================= ======== ================ ===========
> ============
> IronPythonLeaker.exe          5248 Console                    1     73,644
> K
>
>
> On Thu, Dec 25, 2008 at 3:37 PM, Christian Schmidt <
> christian2.schmidt at gmx.de> wrote:
>
>> Hi,
>>
>> the following code snippet allocates constantly more and more memory.
>> Moving .Compile() outside the loop works as expected.
>>
>> private static void LeakTest()
>> {
>>  var runtime = Python.CreateRuntime();
>>  var engine = runtime.GetEngine("py");
>>  var sum = 0.0;
>>
>>  while (true)
>>  {
>>    var code = engine.CreateScriptSourceFromString("1.0+2.0").Compile();
>>    sum += code.Execute<double>();
>>    GC.Collect();
>>    GC.WaitForPendingFinalizers();
>>    GC.Collect();
>>    Console.WriteLine(sum);
>>  }
>> }
>>
>> Any ideas?
>>
>> Merry Christmas,
>>
>> Christian
>> _______________________________________________
>> Users mailing list
>> Users at lists.ironpython.com
>> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>>
>
>
>
> --
> nomadlife.org
>
>


-- 
nomadlife.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081226/e7005fee/attachment.html>

From miha.valencic at gmail.com  Fri Dec 26 20:05:00 2008
From: miha.valencic at gmail.com (Miha Valencic)
Date: Fri, 26 Dec 2008 20:05:00 +0100
Subject: [IronPython] DLR, OS X/Linux & other languages
Message-ID: <233dfa1d0812261105i37b254fcieb69283c12c7254f@mail.gmail.com>

<thinking outloud/>

DLR is a subset of CLR, with specific improvements for dynamic languages
(dynamic call sites, dispatch and whatnot). At least that's how I understand
it.

Currently, it runs in the browser on all major platforms (Win, OS X and with
Linux (with Moonlight?)). It would be nice to be able to write apps on OS X
(& linux for that matter) in C# or VB, for instance, that would target the
DLR (not CLR). You would have access only to the subset of the CLR (whatever
DLR brings to the table) but nontheless.

For whatever is missing, one could add assemblies, which would be loaded
into the application... (like with SL). No?

Silverlight is intended for running in the browser -- but could we use this
same technology to write "native" apps? We need a host of some sort for the
UI (something that currently browser provides) or maybe we already have
something like that?

Could I write an application for OSX and run it with DLR (with a small
download and as low complexity as possible)?

I know I can download Mono and install it and use that for development and
so on -- what about DLR?

Miha

ps: I could launch the same word on Win and on OS X, if Word would be
written in .NET completely.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081226/1f0edaab/attachment.html>

From me at alcidesfonseca.com  Fri Dec 26 20:15:50 2008
From: me at alcidesfonseca.com (Alcides Fonseca)
Date: Fri, 26 Dec 2008 19:15:50 +0000
Subject: [IronPython] DLR, OS X/Linux & other languages
In-Reply-To: <233dfa1d0812261105i37b254fcieb69283c12c7254f@mail.gmail.com>
References: <233dfa1d0812261105i37b254fcieb69283c12c7254f@mail.gmail.com>
Message-ID: <94F26493-2C86-4533-80D9-C7DC70364021@alcidesfonseca.com>


Em 2008/12/26, ?s 19:05, Miha Valencic escreveu:

> <thinking outloud/>
>
> DLR is a subset of CLR, with specific improvements for dynamic  
> languages (dynamic call sites, dispatch and whatnot). At least  
> that's how I understand it.

That's not exactly correct. Silverlight has this thing called CoreCLR,  
and that's what implements the subset of the CLR.

DLR is another layer on top of the CLR (although I believe it is going  
to be included in a future version). The DLR adds support for dynamic  
stuff on the CLR (both the desktop and the CoreCLR).


>
>
> Currently, it runs in the browser on all major platforms (Win, OS X  
> and with Linux (with Moonlight?)). It would be nice to be able to  
> write apps on OS X (& linux for that matter) in C# or VB, for  
> instance, that would target the DLR (not CLR). You would have access  
> only to the subset of the CLR (whatever DLR brings to the table) but  
> nontheless.

Silverlight apps target the CoreCLR, not exactly the DLR. You can  
write a Silverlight app in C# and it won't make use of the CLR. If you  
use IronPython, IronRuby or JScript, you are using the DLR to access  
the CoreCLR in Silverlight.

>
>
> For whatever is missing, one could add assemblies, which would be  
> loaded into the application... (like with SL). No?
>
> Silverlight is intended for running in the browser -- but could we  
> use this same technology to write "native" apps? We need a host of  
> some sort for the UI (something that currently browser provides) or  
> maybe we already have something like that?

At least in Linux, you can run silverlight apps in the desktop. Miguel  
calls them Moonlight Desklets, and you can learn more about them in  
his blog: http://tirania.org/blog/archive/2008/Apr-17.html

>
>
> Could I write an application for OSX and run it with DLR (with a  
> small download and as low complexity as possible)?
>
> I know I can download Mono and install it and use that for  
> development and so on -- what about DLR?

In order to run .NET apps in OSX/Linux, you need a virtual machine.  
Two are available at this time: Mono and Moonlight. The first is the  
best for your needs. It has a larger API and you can use C# to write  
applications, or Iron* with the DLR+CLR. The other is using the  
desklets.

The DLR is only used to access the CLR underneath. So it alone can't  
run anything.

>
>
> Miha
>
> ps: I could launch the same word on Win and on OS X, if Word would  
> be written in .NET completely.

True, but Mono doesn't have the same APIs as Windows's Microsoft .NET.  
WPF isn't implemented, as well as other APIs.


Alcides




From miha.valencic at gmail.com  Fri Dec 26 20:31:47 2008
From: miha.valencic at gmail.com (Miha Valencic)
Date: Fri, 26 Dec 2008 20:31:47 +0100
Subject: [IronPython] DLR, OS X/Linux & other languages
In-Reply-To: <94F26493-2C86-4533-80D9-C7DC70364021@alcidesfonseca.com>
References: <233dfa1d0812261105i37b254fcieb69283c12c7254f@mail.gmail.com>
	<94F26493-2C86-4533-80D9-C7DC70364021@alcidesfonseca.com>
Message-ID: <233dfa1d0812261131l7e209960w7261d86bf2744a10@mail.gmail.com>

Thanks for correcting me. So, how could I leverage the use of CoreCLR for OS
X/Linux... development? With c# or Iron*?

Actually, Desklets sound like the right solution. But with the same API
across all platforms. Something that Java "delivered" with Swing -- sort of.


A way to have a powerful (but at the same time lightweight (== not bloated))
framework/infrastructure, that enables you to write apps in the language of
your choice -- be it M, C#, VB, python, ruby or javascript (BTW, given that
a lot of web developers use javascript at their work, using managed jscript
for development has some interesting potential -- like VB).

Wouldn't it be cool to deliver developer tooling to support that? I could
leverage whatever language I am proficient in, I could leverage XAML for the
UI (hopefully the same on for desktop as for the web) and the software I
would write would run on most platforms and the ones that don't have the
"runtime", could get it with a quick download (like Silverlight).

It seems obvious to me that something like that is going to happen -- I'm
just wondering why it hasn't yet. Where is the missing link? If Miguel has
the tooling to run "Silverlight apps" on the desktop on Linux, I think they
could also run on Windows and/or MacOSX.

What about accessing functionality (BCL) which is not included in the
CoreCLR? Could I just reference (and add to deployment)
System.ServiceModel.dll assembly and have access to full WCF? (there are of
course other implications to that, but let's brainstorm for a minute)

Hey, Ironmen, what do you think?

Miha

2008/12/26 Alcides Fonseca <me at alcidesfonseca.com>

>
> Em 2008/12/26, ?s 19:05, Miha Valencic escreveu:
>
>  <thinking outloud/>
>>
>> DLR is a subset of CLR, with specific improvements for dynamic languages
>> (dynamic call sites, dispatch and whatnot). At least that's how I understand
>> it.
>>
>
> That's not exactly correct. Silverlight has this thing called CoreCLR, and
> that's what implements the subset of the CLR.
>
> DLR is another layer on top of the CLR (although I believe it is going to
> be included in a future version). The DLR adds support for dynamic stuff on
> the CLR (both the desktop and the CoreCLR).
>
>
>
>>
>> Currently, it runs in the browser on all major platforms (Win, OS X and
>> with Linux (with Moonlight?)). It would be nice to be able to write apps on
>> OS X (& linux for that matter) in C# or VB, for instance, that would target
>> the DLR (not CLR). You would have access only to the subset of the CLR
>> (whatever DLR brings to the table) but nontheless.
>>
>
> Silverlight apps target the CoreCLR, not exactly the DLR. You can write a
> Silverlight app in C# and it won't make use of the CLR. If you use
> IronPython, IronRuby or JScript, you are using the DLR to access the CoreCLR
> in Silverlight.
>
>
>>
>> For whatever is missing, one could add assemblies, which would be loaded
>> into the application... (like with SL). No?
>>
>> Silverlight is intended for running in the browser -- but could we use
>> this same technology to write "native" apps? We need a host of some sort for
>> the UI (something that currently browser provides) or maybe we already have
>> something like that?
>>
>
> At least in Linux, you can run silverlight apps in the desktop. Miguel
> calls them Moonlight Desklets, and you can learn more about them in his
> blog: http://tirania.org/blog/archive/2008/Apr-17.html
>
>
>>
>> Could I write an application for OSX and run it with DLR (with a small
>> download and as low complexity as possible)?
>>
>> I know I can download Mono and install it and use that for development and
>> so on -- what about DLR?
>>
>
> In order to run .NET apps in OSX/Linux, you need a virtual machine. Two are
> available at this time: Mono and Moonlight. The first is the best for your
> needs. It has a larger API and you can use C# to write applications, or
> Iron* with the DLR+CLR. The other is using the desklets.
>
> The DLR is only used to access the CLR underneath. So it alone can't run
> anything.
>
>
>>
>> Miha
>>
>> ps: I could launch the same word on Win and on OS X, if Word would be
>> written in .NET completely.
>>
>
> True, but Mono doesn't have the same APIs as Windows's Microsoft .NET. WPF
> isn't implemented, as well as other APIs.
>
>
> Alcides
>
>
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081226/598c6bfc/attachment.html>

From kfarmer at thuban.org  Fri Dec 26 21:14:20 2008
From: kfarmer at thuban.org (Keith J. Farmer)
Date: Fri, 26 Dec 2008 12:14:20 -0800
Subject: [IronPython] DLR, OS X/Linux & other languages
References: <233dfa1d0812261105i37b254fcieb69283c12c7254f@mail.gmail.com><94F26493-2C86-4533-80D9-C7DC70364021@alcidesfonseca.com>
	<233dfa1d0812261131l7e209960w7261d86bf2744a10@mail.gmail.com>
Message-ID: <EEDCD954DD4A0D488BBDBDDCC26075442CC951@meta-exchange.circle-cross.home>

My memory is sketchy on this point, but for Silverlight, you have to either provide the assembly itself, or you have to have it blessed.  The problem with having the assembly is that you end up having to bring in others.
 
What's wrong with targetting a full WPF host (assuming that Mono's WPF support is sufficient at the time)?
 
Either stick to the reduced set or not.  Don't try to turn the one into the other :)
 
(and looking outside, it's snowing again...)

________________________________

From: users-bounces at lists.ironpython.com on behalf of Miha Valencic
Sent: Fri 12/26/2008 11:31 AM
To: Discussion of IronPython
Subject: Re: [IronPython] DLR, OS X/Linux & other languages


Thanks for correcting me. So, how could I leverage the use of CoreCLR for OS X/Linux... development? With c# or Iron*?

Actually, Desklets sound like the right solution. But with the same API across all platforms. Something that Java "delivered" with Swing -- sort of. 

A way to have a powerful (but at the same time lightweight (== not bloated)) framework/infrastructure, that enables you to write apps in the language of your choice -- be it M, C#, VB, python, ruby or javascript (BTW, given that a lot of web developers use javascript at their work, using managed jscript for development has some interesting potential -- like VB).

Wouldn't it be cool to deliver developer tooling to support that? I could leverage whatever language I am proficient in, I could leverage XAML for the UI (hopefully the same on for desktop as for the web) and the software I would write would run on most platforms and the ones that don't have the "runtime", could get it with a quick download (like Silverlight).

It seems obvious to me that something like that is going to happen -- I'm just wondering why it hasn't yet. Where is the missing link? If Miguel has the tooling to run "Silverlight apps" on the desktop on Linux, I think they could also run on Windows and/or MacOSX.

What about accessing functionality (BCL) which is not included in the CoreCLR? Could I just reference (and add to deployment) System.ServiceModel.dll assembly and have access to full WCF? (there are of course other implications to that, but let's brainstorm for a minute)

Hey, Ironmen, what do you think?

Miha


2008/12/26 Alcides Fonseca <me at alcidesfonseca.com>



	Em 2008/12/26, ?s 19:05, Miha Valencic escreveu: 



		<thinking outloud/>
		
		DLR is a subset of CLR, with specific improvements for dynamic languages (dynamic call sites, dispatch and whatnot). At least that's how I understand it.
		


	That's not exactly correct. Silverlight has this thing called CoreCLR, and that's what implements the subset of the CLR.
	
	DLR is another layer on top of the CLR (although I believe it is going to be included in a future version). The DLR adds support for dynamic stuff on the CLR (both the desktop and the CoreCLR). 






		Currently, it runs in the browser on all major platforms (Win, OS X and with Linux (with Moonlight?)). It would be nice to be able to write apps on OS X (& linux for that matter) in C# or VB, for instance, that would target the DLR (not CLR). You would have access only to the subset of the CLR (whatever DLR brings to the table) but nontheless.
		


	Silverlight apps target the CoreCLR, not exactly the DLR. You can write a Silverlight app in C# and it won't make use of the CLR. If you use IronPython, IronRuby or JScript, you are using the DLR to access the CoreCLR in Silverlight. 





		For whatever is missing, one could add assemblies, which would be loaded into the application... (like with SL). No?
		
		Silverlight is intended for running in the browser -- but could we use this same technology to write "native" apps? We need a host of some sort for the UI (something that currently browser provides) or maybe we already have something like that?
		


	At least in Linux, you can run silverlight apps in the desktop. Miguel calls them Moonlight Desklets, and you can learn more about them in his blog: http://tirania.org/blog/archive/2008/Apr-17.html 





		Could I write an application for OSX and run it with DLR (with a small download and as low complexity as possible)?
		
		I know I can download Mono and install it and use that for development and so on -- what about DLR?
		


	In order to run .NET apps in OSX/Linux, you need a virtual machine. Two are available at this time: Mono and Moonlight. The first is the best for your needs. It has a larger API and you can use C# to write applications, or Iron* with the DLR+CLR. The other is using the desklets.
	
	The DLR is only used to access the CLR underneath. So it alone can't run anything. 





		Miha
		
		ps: I could launch the same word on Win and on OS X, if Word would be written in .NET completely.
		


	True, but Mono doesn't have the same APIs as Windows's Microsoft .NET. WPF isn't implemented, as well as other APIs.
	
	
	Alcides
	
	
	_______________________________________________
	Users mailing list
	Users at lists.ironpython.com
	http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
	


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081226/131355c2/attachment.html>

From miha.valencic at gmail.com  Fri Dec 26 21:48:47 2008
From: miha.valencic at gmail.com (Miha Valencic)
Date: Fri, 26 Dec 2008 21:48:47 +0100
Subject: [IronPython] DLR, OS X/Linux & other languages
In-Reply-To: <EEDCD954DD4A0D488BBDBDDCC26075442CC951@meta-exchange.circle-cross.home>
References: <233dfa1d0812261105i37b254fcieb69283c12c7254f@mail.gmail.com>
	<94F26493-2C86-4533-80D9-C7DC70364021@alcidesfonseca.com>
	<233dfa1d0812261131l7e209960w7261d86bf2744a10@mail.gmail.com>
	<EEDCD954DD4A0D488BBDBDDCC26075442CC951@meta-exchange.circle-cross.home>
Message-ID: <233dfa1d0812261248l1b5c10a9hcf59e42b3e7a7e@mail.gmail.com>

[I hope this is not too off-topic...]

What is wrong? Nothing I guess -- if you have full WPF at the other end.
Probably WPF (or WCF in my example) is not the right example -- due to
dependencies further down the road.

Reduced set or not, you will end up with extensions/additions to base class
library. You have extensions/additions with SL as well -- be it custom UI
controls or some other backend libraries that make your life easier. (it is
the same with desktop development). So you end up bringing those libraries
along with your apps.

Now the point of my ... writing/thinking is not how to bring in additional
DLLs but how far are we from having the Silverilght programming model (with
DLR + CoreCLR + XAML) with the whole extensibility model as a [lightweight]
desktop programming model as well?

So that, regardless of my platform of choice (Linux, OS X, Windows
XP/Vista/7?), I can develop apps in the language of choice (be it a dynamic
beast or a statically typed beast or something in between :)), and have it
run on multiple platforms -- and on the web.

Basically, we would need an environment to run Silverlight applications on
the desktop as a "first class applications" with access to the full
.NET/Mono/xy API where applicable. Something like desklets apparently
deliver[1].

It seems pretty awesome to me that I would be able to choose the language
and have a great tooling support (UI controls, BCL subset..., VisualStudio)
and that I could develop for the web and for the desktop using the same
technology and paradigms.

Miha

[1]:
http://www.davidarno.org/2008/04/21/mono-team-to-microsoft-build-moonlight-desklet-support-into-silverlight/


2008/12/26 Keith J. Farmer <kfarmer at thuban.org>

>   What's wrong with targetting a full WPF host (assuming that Mono's WPF
> support is sufficient at the time)?
>
> Either stick to the reduced set or not.  Don't try to turn the one into the
> other :)
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081226/c162f16b/attachment.html>

From daftspaniel at gmail.com  Fri Dec 26 22:04:25 2008
From: daftspaniel at gmail.com (Davy Mitchell)
Date: Fri, 26 Dec 2008 21:04:25 +0000
Subject: [IronPython] ANN : Davy's IronPython Editor
In-Reply-To: <4950CE1F.7060308@voidspace.org.uk>
References: <20253b0c0812200322w4734bf90u6947528ce97e92c8@mail.gmail.com>
	<4950CE1F.7060308@voidspace.org.uk>
Message-ID: <20253b0c0812261304y51f9e387u404dcfbaf4a12728@mail.gmail.com>

On Tue, Dec 23, 2008 at 11:40 AM, Michael Foord
<fuzzyman at voidspace.org.uk>wrote:

> Awesome! You should look at PySmell for autocomplete. It won't run under
> IronPython but is intended to be run with CPython called as a subprocess
> anyway. I'm sure Orestis will be happy to give you some pointers if you go
> down this route.
>

I am trying to avoid CPython as a dependency but may be worth it as an
option for autocomplete. Thanks for the suggestion.

>it's a good tool but in my opinion the name 'DIE' is not good, isn't it?

LOL - maybe I should stay out of marketing :-) !!! Seriously though,
avoiding the TLA or a new name may be a good idea!

Take care,
Davy Mitchell
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081226/aa54bc8e/attachment.html>

From kfarmer at thuban.org  Fri Dec 26 22:37:48 2008
From: kfarmer at thuban.org (Keith J. Farmer)
Date: Fri, 26 Dec 2008 13:37:48 -0800
Subject: [IronPython] DLR, OS X/Linux & other languages
References: <233dfa1d0812261105i37b254fcieb69283c12c7254f@mail.gmail.com><94F26493-2C86-4533-80D9-C7DC70364021@alcidesfonseca.com><233dfa1d0812261131l7e209960w7261d86bf2744a10@mail.gmail.com><EEDCD954DD4A0D488BBDBDDCC26075442CC951@meta-exchange.circle-cross.home>
	<233dfa1d0812261248l1b5c10a9hcf59e42b3e7a7e@mail.gmail.com>
Message-ID: <EEDCD954DD4A0D488BBDBDDCC26075442CC953@meta-exchange.circle-cross.home>

Right, and I would agree that desklets are a nifty way to share sandboxed code.
 
My point was more a caution against trying to undo Silverlight/Moonlight's reduction.  It's one thing to add additional controls, quite another to start messing with the networking support. The security implications could be tragic ;)
 
It may interest you to know that folks have already created Silverlight gadgets for Vista's sidebar including, IIRC, gadgets that are remotely hosted.

________________________________

From: users-bounces at lists.ironpython.com on behalf of Miha Valencic
Sent: Fri 12/26/2008 12:48 PM
To: Discussion of IronPython
Subject: Re: [IronPython] DLR, OS X/Linux & other languages


[I hope this is not too off-topic...]

What is wrong? Nothing I guess -- if you have full WPF at the other end. Probably WPF (or WCF in my example) is not the right example -- due to dependencies further down the road.

Reduced set or not, you will end up with extensions/additions to base class library. You have extensions/additions with SL as well -- be it custom UI controls or some other backend libraries that make your life easier. (it is the same with desktop development). So you end up bringing those libraries along with your apps.

Now the point of my ... writing/thinking is not how to bring in additional DLLs but how far are we from having the Silverilght programming model (with DLR + CoreCLR + XAML) with the whole extensibility model as a [lightweight] desktop programming model as well?

So that, regardless of my platform of choice (Linux, OS X, Windows XP/Vista/7?), I can develop apps in the language of choice (be it a dynamic beast or a statically typed beast or something in between :)), and have it run on multiple platforms -- and on the web.

Basically, we would need an environment to run Silverlight applications on the desktop as a "first class applications" with access to the full .NET/Mono/xy API where applicable. Something like desklets apparently deliver[1].

It seems pretty awesome to me that I would be able to choose the language and have a great tooling support (UI controls, BCL subset..., VisualStudio) and that I could develop for the web and for the desktop using the same technology and paradigms.

Miha

[1]: http://www.davidarno.org/2008/04/21/mono-team-to-microsoft-build-moonlight-desklet-support-into-silverlight/



2008/12/26 Keith J. Farmer <kfarmer at thuban.org>


	  
	What's wrong with targetting a full WPF host (assuming that Mono's WPF support is sufficient at the time)?
	 
	Either stick to the reduced set or not.  Don't try to turn the one into the other :)


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081226/bc1a8384/attachment.html>

From miha.valencic at gmail.com  Fri Dec 26 23:07:34 2008
From: miha.valencic at gmail.com (Miha Valencic)
Date: Fri, 26 Dec 2008 23:07:34 +0100
Subject: [IronPython] DLR, OS X/Linux & other languages
In-Reply-To: <EEDCD954DD4A0D488BBDBDDCC26075442CC953@meta-exchange.circle-cross.home>
References: <233dfa1d0812261105i37b254fcieb69283c12c7254f@mail.gmail.com>
	<94F26493-2C86-4533-80D9-C7DC70364021@alcidesfonseca.com>
	<233dfa1d0812261131l7e209960w7261d86bf2744a10@mail.gmail.com>
	<EEDCD954DD4A0D488BBDBDDCC26075442CC951@meta-exchange.circle-cross.home>
	<233dfa1d0812261248l1b5c10a9hcf59e42b3e7a7e@mail.gmail.com>
	<EEDCD954DD4A0D488BBDBDDCC26075442CC953@meta-exchange.circle-cross.home>
Message-ID: <233dfa1d0812261407q2847752cncdc1af66799e04b1@mail.gmail.com>

Keith,

you are right that what people have done with Vista gadgets interests me,
but usually (I didn't check how they did it though!) it involves
workarounds. Like the ones how they use Cardspace and WCF from gadgets... or
how they access Outlook, etc.

With regards to Desklets, the point is that you have two options: sandboxed
and "full access", which makes it very compelling, IMHO.

With regards to to "reduction" of the API: it is probably a philosophical
debate, but the point with reduction IMHO is not security or sandboxing, but
how to make the "framework or vritual engine" small and make it
"transferrable" to client computers with a fast download.

Sandboxing is absolutely necessary in the browser, but not on the desktop.
The point you make with SL vista gadgets though, makes me think about
something like clickonce for desklets - where you can cache the app for
offline usage, but when you're online, it launches (or at least checks for a
new version) from the remote server(s). (that would need security sandboxing
though).

So, a question to Ironmen and Silverlight guys (Jimmy et al?): is it
possible to somehow run Silverlight app as a desktop app on Windows?
Something like Desklets?

An ugly workaround I can think of at the moment would be a simple winforms
application which hosts a browser control which in turn loads a silverlight
application from a local system. (or maybe even starts a simple local-only
lightweight http server (like chiron or devserver) if it is not possible to
launch SL app from filesystem). That way, we could (could we really?) run
DLRConsle as a desktop app...

Miha

2008/12/26 Keith J. Farmer <kfarmer at thuban.org>

>  Right, and I would agree that desklets are a nifty way to share sandboxed
> code.
>
> My point was more a caution against trying to undo Silverlight/Moonlight's
> reduction.  It's one thing to add additional controls, quite another to
> start messing with the networking support. The security implications could
> be tragic ;)
>
> It may interest you to know that folks have already created Silverlight
> gadgets for Vista's sidebar including, IIRC, gadgets that are remotely
> hosted.
>
> ------------------------------
> *From:* users-bounces at lists.ironpython.com on behalf of Miha Valencic
> *Sent:* Fri 12/26/2008 12:48 PM
> *To:* Discussion of IronPython
> *Subject:* Re: [IronPython] DLR, OS X/Linux & other languages
>
> [I hope this is not too off-topic...]
>
> What is wrong? Nothing I guess -- if you have full WPF at the other end.
> Probably WPF (or WCF in my example) is not the right example -- due to
> dependencies further down the road.
>
> Reduced set or not, you will end up with extensions/additions to base class
> library. You have extensions/additions with SL as well -- be it custom UI
> controls or some other backend libraries that make your life easier. (it is
> the same with desktop development). So you end up bringing those libraries
> along with your apps.
>
> Now the point of my ... writing/thinking is not how to bring in additional
> DLLs but how far are we from having the Silverilght programming model (with
> DLR + CoreCLR + XAML) with the whole extensibility model as a [lightweight]
> desktop programming model as well?
>
> So that, regardless of my platform of choice (Linux, OS X, Windows
> XP/Vista/7?), I can develop apps in the language of choice (be it a dynamic
> beast or a statically typed beast or something in between :)), and have it
> run on multiple platforms -- and on the web.
>
> Basically, we would need an environment to run Silverlight applications on
> the desktop as a "first class applications" with access to the full
> .NET/Mono/xy API where applicable. Something like desklets apparently
> deliver[1].
>
> It seems pretty awesome to me that I would be able to choose the language
> and have a great tooling support (UI controls, BCL subset..., VisualStudio)
> and that I could develop for the web and for the desktop using the same
> technology and paradigms.
>
> Miha
>
> [1]:
> http://www.davidarno.org/2008/04/21/mono-team-to-microsoft-build-moonlight-desklet-support-into-silverlight/
>
>
> 2008/12/26 Keith J. Farmer <kfarmer at thuban.org>
>
>>    What's wrong with targetting a full WPF host (assuming that Mono's WPF
>> support is sufficient at the time)?
>>
>> Either stick to the reduced set or not.  Don't try to turn the one into
>> the other :)
>>
>
>
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081226/cb962783/attachment.html>

From kfarmer at thuban.org  Fri Dec 26 23:32:23 2008
From: kfarmer at thuban.org (Keith J. Farmer)
Date: Fri, 26 Dec 2008 14:32:23 -0800
Subject: [IronPython] DLR, OS X/Linux & other languages
References: <233dfa1d0812261105i37b254fcieb69283c12c7254f@mail.gmail.com><94F26493-2C86-4533-80D9-C7DC70364021@alcidesfonseca.com><233dfa1d0812261131l7e209960w7261d86bf2744a10@mail.gmail.com><EEDCD954DD4A0D488BBDBDDCC26075442CC951@meta-exchange.circle-cross.home><233dfa1d0812261248l1b5c10a9hcf59e42b3e7a7e@mail.gmail.com><EEDCD954DD4A0D488BBDBDDCC26075442CC953@meta-exchange.circle-cross.home>
	<233dfa1d0812261407q2847752cncdc1af66799e04b1@mail.gmail.com>
Message-ID: <EEDCD954DD4A0D488BBDBDDCC26075442CC956@meta-exchange.circle-cross.home>

If you think sandboxing is not necessary on the desktop, consider a Trojan attack using desklets.
 
A buddy of mine in MSN suggests you check out the "Web Sandbox" materials for PDC 2008 for backgrounder, and comments that while the desktop used to be considered an isolation boundary -- it used to be considered fully-trusted.  However, we now know this to be foolish, so if you want to secure downloaded code you either block things completely, or you go into more interesting security.
 
I also don't personally consider your "ugly workaround" all that ugly.  If it makes it more palatable, you could consider it a security scoping mechanism, which I think would be a great thing to have in WPF, and just have it use Silverlight as an implementation detail.
 
-- Keith (ex-Silverlight DataGrid guy)

________________________________

From: users-bounces at lists.ironpython.com on behalf of Miha Valencic
Sent: Fri 12/26/2008 2:07 PM
To: Discussion of IronPython
Subject: Re: [IronPython] DLR, OS X/Linux & other languages


Keith,

you are right that what people have done with Vista gadgets interests me, but usually (I didn't check how they did it though!) it involves workarounds. Like the ones how they use Cardspace and WCF from gadgets... or how they access Outlook, etc.

With regards to Desklets, the point is that you have two options: sandboxed and "full access", which makes it very compelling, IMHO.

With regards to to "reduction" of the API: it is probably a philosophical debate, but the point with reduction IMHO is not security or sandboxing, but how to make the "framework or vritual engine" small and make it "transferrable" to client computers with a fast download.

Sandboxing is absolutely necessary in the browser, but not on the desktop. The point you make with SL vista gadgets though, makes me think about something like clickonce for desklets - where you can cache the app for offline usage, but when you're online, it launches (or at least checks for a new version) from the remote server(s). (that would need security sandboxing though).

So, a question to Ironmen and Silverlight guys (Jimmy et al?): is it possible to somehow run Silverlight app as a desktop app on Windows? Something like Desklets? 

An ugly workaround I can think of at the moment would be a simple winforms application which hosts a browser control which in turn loads a silverlight application from a local system. (or maybe even starts a simple local-only lightweight http server (like chiron or devserver) if it is not possible to launch SL app from filesystem). That way, we could (could we really?) run DLRConsle as a desktop app... 

Miha


2008/12/26 Keith J. Farmer <kfarmer at thuban.org>


	Right, and I would agree that desklets are a nifty way to share sandboxed code.
	 
	My point was more a caution against trying to undo Silverlight/Moonlight's reduction.  It's one thing to add additional controls, quite another to start messing with the networking support. The security implications could be tragic ;)
	 
	It may interest you to know that folks have already created Silverlight gadgets for Vista's sidebar including, IIRC, gadgets that are remotely hosted.

________________________________

	
	From: users-bounces at lists.ironpython.com on behalf of Miha Valencic
	
	Sent: Fri 12/26/2008 12:48 PM 

	To: Discussion of IronPython
	Subject: Re: [IronPython] DLR, OS X/Linux & other languages
	

	[I hope this is not too off-topic...]
	
	What is wrong? Nothing I guess -- if you have full WPF at the other end. Probably WPF (or WCF in my example) is not the right example -- due to dependencies further down the road.
	
	Reduced set or not, you will end up with extensions/additions to base class library. You have extensions/additions with SL as well -- be it custom UI controls or some other backend libraries that make your life easier. (it is the same with desktop development). So you end up bringing those libraries along with your apps.
	
	Now the point of my ... writing/thinking is not how to bring in additional DLLs but how far are we from having the Silverilght programming model (with DLR + CoreCLR + XAML) with the whole extensibility model as a [lightweight] desktop programming model as well?
	
	So that, regardless of my platform of choice (Linux, OS X, Windows XP/Vista/7?), I can develop apps in the language of choice (be it a dynamic beast or a statically typed beast or something in between :)), and have it run on multiple platforms -- and on the web.
	
	Basically, we would need an environment to run Silverlight applications on the desktop as a "first class applications" with access to the full .NET/Mono/xy API where applicable. Something like desklets apparently deliver[1].
	
	It seems pretty awesome to me that I would be able to choose the language and have a great tooling support (UI controls, BCL subset..., VisualStudio) and that I could develop for the web and for the desktop using the same technology and paradigms.
	
	Miha
	
	[1]: http://www.davidarno.org/2008/04/21/mono-team-to-microsoft-build-moonlight-desklet-support-into-silverlight/
	
	
	
	2008/12/26 Keith J. Farmer <kfarmer at thuban.org>
	

		  
		What's wrong with targetting a full WPF host (assuming that Mono's WPF support is sufficient at the time)?
		 
		Either stick to the reduced set or not.  Don't try to turn the one into the other :)



	_______________________________________________
	Users mailing list
	Users at lists.ironpython.com
	http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
	
	


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081226/64624d1d/attachment.html>

From miha.valencic at gmail.com  Fri Dec 26 23:54:19 2008
From: miha.valencic at gmail.com (Miha Valencic)
Date: Fri, 26 Dec 2008 23:54:19 +0100
Subject: [IronPython] DLR, OS X/Linux & other languages
In-Reply-To: <EEDCD954DD4A0D488BBDBDDCC26075442CC956@meta-exchange.circle-cross.home>
References: <233dfa1d0812261105i37b254fcieb69283c12c7254f@mail.gmail.com>
	<94F26493-2C86-4533-80D9-C7DC70364021@alcidesfonseca.com>
	<233dfa1d0812261131l7e209960w7261d86bf2744a10@mail.gmail.com>
	<EEDCD954DD4A0D488BBDBDDCC26075442CC951@meta-exchange.circle-cross.home>
	<233dfa1d0812261248l1b5c10a9hcf59e42b3e7a7e@mail.gmail.com>
	<EEDCD954DD4A0D488BBDBDDCC26075442CC953@meta-exchange.circle-cross.home>
	<233dfa1d0812261407q2847752cncdc1af66799e04b1@mail.gmail.com>
	<EEDCD954DD4A0D488BBDBDDCC26075442CC956@meta-exchange.circle-cross.home>
Message-ID: <233dfa1d0812261454j4ec13d49pa7c086b02f36a77f@mail.gmail.com>

Keith,

I "knew" I had that coming. :) It is exactly as you point out + more... but
given the context, I think that we don't need sandboxing in the same way we
don't have sandboxing for <insert downloadable application name>. It would
be nice and UAC can be viewed as a step in this direction (although just for
"admin-required" operations). Management would be harder... but come to
think of it, modern mobile phones have some sort of sandboxing with
different permission sets built into JVMs. (on some devices, you can specify
which operations you allow and you can specify for which operations you want
to be asked every time if you allow them).

Will check the PDC materials, thanks.

I think we're completely off-topic now.... although it's a very interesting
topic so far. :) [redirect to private email?]

To recap: desklet in my opinion is not just sth you download from the web.
It is something which can also be signed and distributed on a CD, you
"install it" like a regular program and launch it. It has access to system
resources (file system, CD drive, audio, camera,...) and whatnot.
It could very well be run in two modes. Sandboxed (same sandboxing as a
browser) and "Free", where it has the same access as every other app on the
system.

With next-gen OSs, we might have sandboxing for every app (like virtualized
environment for the running app -- something like that is possible today
with ThinApp and other solutions), but until then, I wouldn't treat them
differently than any other user-installable applicaiton. The point here is
"user-installable", of course.

Miha

2008/12/26 Keith J. Farmer <kfarmer at thuban.org>

>  If you think sandboxing is not necessary on the desktop, consider a
> Trojan attack using desklets.
>
> A buddy of mine in MSN suggests you check out the "Web Sandbox" materials
> for PDC 2008 for backgrounder, and comments that while the desktop used to
> be considered an isolation boundary -- it used to be considered
> fully-trusted.  However, we now know this to be foolish, so if you want to
> secure downloaded code you either block things completely, or you go into
> more interesting security.
>
> I also don't personally consider your "ugly workaround" all that ugly.  If
> it makes it more palatable, you could consider it a security scoping
> mechanism, which I think would be a great thing to have in WPF, and just
> have it use Silverlight as an implementation detail.
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081226/76147af5/attachment.html>

From fuzzyman at voidspace.org.uk  Sat Dec 27 00:07:58 2008
From: fuzzyman at voidspace.org.uk (Michael Foord)
Date: Fri, 26 Dec 2008 23:07:58 +0000
Subject: [IronPython] ANN : Davy's IronPython Editor
In-Reply-To: <20253b0c0812261304y51f9e387u404dcfbaf4a12728@mail.gmail.com>
References: <20253b0c0812200322w4734bf90u6947528ce97e92c8@mail.gmail.com>	<4950CE1F.7060308@voidspace.org.uk>
	<20253b0c0812261304y51f9e387u404dcfbaf4a12728@mail.gmail.com>
Message-ID: <495563CE.90508@voidspace.org.uk>

Davy Mitchell wrote:
> On Tue, Dec 23, 2008 at 11:40 AM, Michael Foord 
> <fuzzyman at voidspace.org.uk <mailto:fuzzyman at voidspace.org.uk>> wrote:
>
>     Awesome! You should look at PySmell for autocomplete. It won't run
>     under IronPython but is intended to be run with CPython called as
>     a subprocess anyway. I'm sure Orestis will be happy to give you
>     some pointers if you go down this route.
>
>
> I am trying to avoid CPython as a dependency but may be worth it as an 
> option for autocomplete. Thanks for the suggestion.

It doesn't need to be an external dependency - but could be included in 
the project files (using py2exe is one way).

Michael
>
> >it's a good tool but in my opinion the name 'DIE' is not good, isn't it?
>
> LOL - maybe I should stay out of marketing :-) !!! Seriously though, 
> avoiding the TLA or a new name may be a good idea!
>
> Take care,
> Davy Mitchell
> ------------------------------------------------------------------------
>
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>   


-- 
http://www.ironpythoninaction.com/
http://www.voidspace.org.uk/blog




From andy.sy at neotitans.com  Sat Dec 27 19:47:09 2008
From: andy.sy at neotitans.com (Andy Sy)
Date: Sun, 28 Dec 2008 02:47:09 +0800
Subject: [IronPython] DLR, OS X/Linux & other languages
In-Reply-To: <94F26493-2C86-4533-80D9-C7DC70364021@alcidesfonseca.com>
References: <233dfa1d0812261105i37b254fcieb69283c12c7254f@mail.gmail.com>
	<94F26493-2C86-4533-80D9-C7DC70364021@alcidesfonseca.com>
Message-ID: <4956782D.3030506@neotitans.com>

Alcides Fonseca wrote:

> Em 2008/12/26, ?s 19:05, Miha Valencic escreveu:
> 
>> <thinking outloud/>
>>
>> DLR is a subset of CLR, with specific improvements for dynamic 
>> languages (dynamic call sites, dispatch and whatnot). At least that's 
>> how I understand it.
> 
> That's not exactly correct. Silverlight has this thing called CoreCLR, 
> and that's what implements the subset of the CLR.
> 
> DLR is another layer on top of the CLR (although I believe it is going 
> to be included in a future version). The DLR adds support for dynamic 
> stuff on the CLR (both the desktop and the CoreCLR).

Just to add in my 2 cents...

The way I understand DLR's purpose is... a language like IronPython
could be implemented on top of the CLR directly (and this is actually
the case with IronPython v1.x ).

However, dynamic languages like Python and Ruby share some commonalities
and thus it made sense to create yet another shared layer on top of the
CLR that provides facilities addressing those commonalities.

Thus implementors of dynamic languages - Ruby, Python and other future
ones - can just reuse the stuff provided by the DLR rather than
reinventing certain wheels.

So I believe IronRuby and IronPython 2.x are already being built on
top of the DLR?

================================
http://www.webmechs.com/webpress
The Webmechs Press


From miha.valencic at gmail.com  Sat Dec 27 20:43:09 2008
From: miha.valencic at gmail.com (Miha Valencic)
Date: Sat, 27 Dec 2008 20:43:09 +0100
Subject: [IronPython] DLR, OS X/Linux & other languages
In-Reply-To: <4956782D.3030506@neotitans.com>
References: <233dfa1d0812261105i37b254fcieb69283c12c7254f@mail.gmail.com>
	<94F26493-2C86-4533-80D9-C7DC70364021@alcidesfonseca.com>
	<4956782D.3030506@neotitans.com>
Message-ID: <233dfa1d0812271143k60bf176asdb27ae84813ed594@mail.gmail.com>

Andy, well put. I think it offers even more than that. A common hosting API
(to host dynamic languages inside CLR applications and to offer facilities
for dynamic language implementations to access .NET objects (like
IronPython's clr.AddReference and import clr ...). And of course tools for
the language writer (AST and all of the other goodies).

This can be seen from the DLR hosting spec, which is publicly available at
http://compilerlab.members.winisp.net/.

And yes, from my understanding, IPY 2 is build on top of the DLR. (and
IronRuby as well).

Regards,
 Miha.
2008/12/27 Andy Sy <andy.sy at neotitans.com>

> Alcides Fonseca wrote:
>
>  Em 2008/12/26, ?s 19:05, Miha Valencic escreveu:
>>
>>  <thinking outloud/>
>>>
>>> DLR is a subset of CLR, with specific improvements for dynamic languages
>>> (dynamic call sites, dispatch and whatnot). At least that's how I understand
>>> it.
>>>
>>
>> That's not exactly correct. Silverlight has this thing called CoreCLR, and
>> that's what implements the subset of the CLR.
>>
>> DLR is another layer on top of the CLR (although I believe it is going to
>> be included in a future version). The DLR adds support for dynamic stuff on
>> the CLR (both the desktop and the CoreCLR).
>>
>
> Just to add in my 2 cents...
>
> The way I understand DLR's purpose is... a language like IronPython
> could be implemented on top of the CLR directly (and this is actually
> the case with IronPython v1.x ).
>
> However, dynamic languages like Python and Ruby share some commonalities
> and thus it made sense to create yet another shared layer on top of the
> CLR that provides facilities addressing those commonalities.
>
> Thus implementors of dynamic languages - Ruby, Python and other future
> ones - can just reuse the stuff provided by the DLR rather than
> reinventing certain wheels.
>
> So I believe IronRuby and IronPython 2.x are already being built on
> top of the DLR?
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081227/7a4ff9a9/attachment.html>

From jdhardy at gmail.com  Sun Dec 28 01:36:18 2008
From: jdhardy at gmail.com (Jeff Hardy)
Date: Sat, 27 Dec 2008 17:36:18 -0700
Subject: [IronPython] NWSGI 1.0 Released
Message-ID: <ea89d0e50812271636r1b42e678w31b23c713ba17f3b@mail.gmail.com>

I'm very pleased to announce the release of NWSGI 1.0. NWSGI is an
implementation of the WSGI specification (PEP 333) for IronPython. It
is implemented as an ASP.NET HttpHandler and is supported on IIS 6 and
IIS 7 (and should work on any other ASP.NET-capable web server). NWSGI
allows the integration of Python applications with existing ASP.NET
facilities such as caching, sessions, and user management.

NWSGI 1.0 can be downloaded from
http://www.codeplex.com/NWSGI/Release/ProjectReleases.aspx?ReleaseId=18995.
More details are available at
http://jdhardy.blogspot.com/2008/12/nwsgi-10.html

- Jeff


From empirebuilder at gmail.com  Sun Dec 28 09:16:15 2008
From: empirebuilder at gmail.com (Dody Gunawinata)
Date: Sun, 28 Dec 2008 09:16:15 +0100
Subject: [IronPython] NWSGI 1.0 Released
In-Reply-To: <ea89d0e50812271636r1b42e678w31b23c713ba17f3b@mail.gmail.com>
References: <ea89d0e50812271636r1b42e678w31b23c713ba17f3b@mail.gmail.com>
Message-ID: <8cd017b80812280016u3cea2b23i72f37e9fa50b0fa0@mail.gmail.com>

Bravo!

On Sun, Dec 28, 2008 at 1:36 AM, Jeff Hardy <jdhardy at gmail.com> wrote:

> I'm very pleased to announce the release of NWSGI 1.0. NWSGI is an
> implementation of the WSGI specification (PEP 333) for IronPython. It
> is implemented as an ASP.NET HttpHandler and is supported on IIS 6 and
> IIS 7 (and should work on any other ASP.NET-capable web server). NWSGI
> allows the integration of Python applications with existing ASP.NET
> facilities such as caching, sessions, and user management.
>
> NWSGI 1.0 can be downloaded from
> http://www.codeplex.com/NWSGI/Release/ProjectReleases.aspx?ReleaseId=18995
> .
> More details are available at
> http://jdhardy.blogspot.com/2008/12/nwsgi-10.html
>
> - Jeff
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>



-- 
nomadlife.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081228/5e18d393/attachment.html>

From D.Moor at sim-soft.co.uk  Sun Dec 28 12:31:47 2008
From: D.Moor at sim-soft.co.uk (Dave Moor)
Date: Sun, 28 Dec 2008 11:31:47 +0000
Subject: [IronPython] ANN : Davy's IronPython Editor
In-Reply-To: <20253b0c0812261304y51f9e387u404dcfbaf4a12728@mail.gmail.com>
References: <20253b0c0812200322w4734bf90u6947528ce97e92c8@mail.gmail.com>
	<4950CE1F.7060308@voidspace.org.uk>
	<20253b0c0812261304y51f9e387u404dcfbaf4a12728@mail.gmail.com>
Message-ID: <WorldClient-F200812281131.AA31470002@sim-soft.co.uk>

Perhaps its my English sense of humour but an IDE called DIE catches my 
attention in a number of ways.  It's name alone has placed up far further up 
the 'must give it a go sometime' list than any other IronPython IDE.

Just my 2 pence worth.

Dave

-----Original Message-----

From: "Davy Mitchell" <daftspaniel at gmail.com>

To: "Discussion of IronPython" <users at lists.ironpython.com>

Date: Fri, 26 Dec 2008 21:04:25 +0000

Subject: Re: [IronPython] ANN : Davy's IronPython Editor



>it's a good tool but in my opinion the name 'DIE' is not good, isn't it?

LOL - maybe I should stay out of marketing :-) !!! Seriously though, 
avoiding the TLA or a new name may be a good idea!


--
Registered Office
Mill House, Hawksworth Road, Horsforth, Leeds, LS18 4JP  England.
Reg No. 1311728 Eng.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081228/460bbe42/attachment.html>

From jdhardy at gmail.com  Mon Dec 29 01:41:33 2008
From: jdhardy at gmail.com (Jeff Hardy)
Date: Sun, 28 Dec 2008 17:41:33 -0700
Subject: [IronPython] Bug in Django's lazy functions
Message-ID: <ea89d0e50812281641m1bfe0560g5849049df386722d@mail.gmail.com>

Hi all,
I've been on the trail of this bug for months, and I think I've
finally caught a piece of it. I've uploaded a somewhat-minimal repro
to http://cid-414fa1a9bd174b4b.skydrive.live.com/self.aspx/Public/lazy_cmp.zip.
functional.py is taken from Django and is used extensively by it. The
lazy() function, in particular, has some hairy code in it to enable
lazy objects that seems to be where this problem originates.

The output of lazy_cmp.py is differsbetween IP 2.0 and Python 2.5:
>python lazy_cmp.py
1

>ipy lazy_cmp.py
Traceback (most recent call last):
  File "lazy_cmp.py", line 13, in lazy_cmp.py
  File "C:\Users\Jeff\Desktop\lazy_cmp\functional.py", line 191, in __wrapper__
TypeError: expected str, got Object_1$6

Hopefully the fix is simple, as this completely breaks Django's admin views.

- Jeff
-------------- next part --------------
A non-text attachment was scrubbed...
Name: lazy_cmp.zip
Type: application/zip
Size: 7249 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081228/8699802d/attachment.zip>

From curt at hagenlocher.org  Mon Dec 29 17:43:49 2008
From: curt at hagenlocher.org (Curt Hagenlocher)
Date: Mon, 29 Dec 2008 08:43:49 -0800
Subject: [IronPython] Bug in Django's lazy functions
In-Reply-To: <ea89d0e50812281641m1bfe0560g5849049df386722d@mail.gmail.com>
References: <ea89d0e50812281641m1bfe0560g5849049df386722d@mail.gmail.com>
Message-ID: <d2155e360812290843j8314024k5c02647fd2da8a8a@mail.gmail.com>

This specific error seems to come down to the fact that IronPython
implements str.__cmp__ while CPython apparently does not.

PS C:\src> ipyd
IronPython 2.0 Beta (2.0.0.5000) on .NET 2.0.50727.3069
Type "help", "copyright", "credits" or "license" for more information.
>>> str.__cmp__
<method '__cmp__' of 'str' objects>
>>> ^Z

PS C:\src> C:\Python25\python.exe
Python 2.5.2 (r252:60911, Feb 21 2008, 13:11:45) [MSC v.1310 32 bit (Intel)]
on win32
Type "help", "copyright", "credits" or "license" for more information.
>>> str.__cmp__
<method-wrapper '__cmp__' of type object at 0x1E1E30C0>
>>>

What happens is that IronPython ends up calling str.__cmp__, which correctly
reports that its argument isn't a string.  CPython apparently calls
object.__cmp__ (or something masquerading as it) which is much less picky.
 It seems to me that this code is overly sensitive to something that should
be an implementation detail, but we'd obviously like to provide a way to
work around it.  Please add this to CodePlex.

There may also be a problem as a result of str == unicode; in fact, there's
a commented-out assertion in line 175 of functional.py that specifically
breaks because this is true.

On Sun, Dec 28, 2008 at 4:41 PM, Jeff Hardy <jdhardy at gmail.com> wrote:

> Hi all,
> I've been on the trail of this bug for months, and I think I've
> finally caught a piece of it. I've uploaded a somewhat-minimal repro
> to
> http://cid-414fa1a9bd174b4b.skydrive.live.com/self.aspx/Public/lazy_cmp.zip.
> functional.py<http://cid-414fa1a9bd174b4b.skydrive.live.com/self.aspx/Public/lazy_cmp.zip.functional.py>is taken from Django and is used extensively by it. The
> lazy() function, in particular, has some hairy code in it to enable
> lazy objects that seems to be where this problem originates.
>
> The output of lazy_cmp.py is differsbetween IP 2.0 and Python 2.5:
> >python lazy_cmp.py
> 1
>
> >ipy lazy_cmp.py
> Traceback (most recent call last):
>  File "lazy_cmp.py", line 13, in lazy_cmp.py
>  File "C:\Users\Jeff\Desktop\lazy_cmp\functional.py", line 191, in
> __wrapper__
> TypeError: expected str, got Object_1$6
>
> Hopefully the fix is simple, as this completely breaks Django's admin
> views.
>
> - Jeff
>
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081229/fae6dda2/attachment.html>

From curt at hagenlocher.org  Mon Dec 29 17:52:55 2008
From: curt at hagenlocher.org (Curt Hagenlocher)
Date: Mon, 29 Dec 2008 08:52:55 -0800
Subject: [IronPython] Bug in Django's lazy functions
In-Reply-To: <d2155e360812290843j8314024k5c02647fd2da8a8a@mail.gmail.com>
References: <ea89d0e50812281641m1bfe0560g5849049df386722d@mail.gmail.com>
	<d2155e360812290843j8314024k5c02647fd2da8a8a@mail.gmail.com>
Message-ID: <d2155e360812290852t6b66d1a3nb5f100b734d40bef@mail.gmail.com>

By the way, you can see the same behavior from CPython if you do this:
from functional import lazy

def foo():
    return "foo"

f = lazy(foo, str)
print str.strip(f())

Builtin functions and magical proxies don't appear to work well together...

On Mon, Dec 29, 2008 at 8:43 AM, Curt Hagenlocher <curt at hagenlocher.org>wrote:

> This specific error seems to come down to the fact that IronPython
> implements str.__cmp__ while CPython apparently does not.
>
> PS C:\src> ipyd
> IronPython 2.0 Beta (2.0.0.5000) on .NET 2.0.50727.3069
> Type "help", "copyright", "credits" or "license" for more information.
> >>> str.__cmp__
> <method '__cmp__' of 'str' objects>
> >>> ^Z
>
> PS C:\src> C:\Python25\python.exe
> Python 2.5.2 (r252:60911, Feb 21 2008, 13:11:45) [MSC v.1310 32 bit
> (Intel)] on win32
> Type "help", "copyright", "credits" or "license" for more information.
> >>> str.__cmp__
> <method-wrapper '__cmp__' of type object at 0x1E1E30C0>
> >>>
>
> What happens is that IronPython ends up calling str.__cmp__, which
> correctly reports that its argument isn't a string.  CPython apparently
> calls object.__cmp__ (or something masquerading as it) which is much less
> picky.  It seems to me that this code is overly sensitive to something that
> should be an implementation detail, but we'd obviously like to provide a way
> to work around it.  Please add this to CodePlex.
>
> There may also be a problem as a result of str == unicode; in fact, there's
> a commented-out assertion in line 175 of functional.py that specifically
> breaks because this is true.
>
> On Sun, Dec 28, 2008 at 4:41 PM, Jeff Hardy <jdhardy at gmail.com> wrote:
>
>> Hi all,
>> I've been on the trail of this bug for months, and I think I've
>> finally caught a piece of it. I've uploaded a somewhat-minimal repro
>> to
>> http://cid-414fa1a9bd174b4b.skydrive.live.com/self.aspx/Public/lazy_cmp.zip.
>> functional.py<http://cid-414fa1a9bd174b4b.skydrive.live.com/self.aspx/Public/lazy_cmp.zip.functional.py>is taken from Django and is used extensively by it. The
>> lazy() function, in particular, has some hairy code in it to enable
>> lazy objects that seems to be where this problem originates.
>>
>> The output of lazy_cmp.py is differsbetween IP 2.0 and Python 2.5:
>> >python lazy_cmp.py
>> 1
>>
>> >ipy lazy_cmp.py
>> Traceback (most recent call last):
>>  File "lazy_cmp.py", line 13, in lazy_cmp.py
>>  File "C:\Users\Jeff\Desktop\lazy_cmp\functional.py", line 191, in
>> __wrapper__
>> TypeError: expected str, got Object_1$6
>>
>> Hopefully the fix is simple, as this completely breaks Django's admin
>> views.
>>
>> - Jeff
>>
>> _______________________________________________
>> Users mailing list
>> Users at lists.ironpython.com
>> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>>
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081229/e1b5bb85/attachment.html>

From jdhardy at gmail.com  Mon Dec 29 20:35:24 2008
From: jdhardy at gmail.com (Jeff Hardy)
Date: Mon, 29 Dec 2008 12:35:24 -0700
Subject: [IronPython] Bug in Django's lazy functions
In-Reply-To: <d2155e360812290843j8314024k5c02647fd2da8a8a@mail.gmail.com>
References: <ea89d0e50812281641m1bfe0560g5849049df386722d@mail.gmail.com>
	<d2155e360812290843j8314024k5c02647fd2da8a8a@mail.gmail.com>
Message-ID: <ea89d0e50812291135p318ac758x5ba83a97713c5937@mail.gmail.com>

On Mon, Dec 29, 2008 at 9:43 AM, Curt Hagenlocher <curt at hagenlocher.org> wrote:
> This specific error seems to come down to the fact that IronPython
> implements str.__cmp__ while CPython apparently does not.
> ...
> What happens is that IronPython ends up calling str.__cmp__, which correctly
> reports that its argument isn't a string.  CPython apparently calls
> object.__cmp__ (or something masquerading as it) which is much less picky.
>  It seems to me that this code is overly sensitive to something that should
> be an implementation detail, but we'd obviously like to provide a way to
> work around it.  Please add this to CodePlex.

Will do. Nice detective work!

> There may also be a problem as a result of str == unicode; in fact, there's
> a commented-out assertion in line 175 of functional.py that specifically
> breaks because this is true.

functional.py isn't *exactly* the same as Django's - I commented that
line out. It does actually have a purpose when str != unicode. There's
code that hits CP #20366 as well.

- Jeff


From MHarelick at ise.com  Mon Dec 29 22:18:16 2008
From: MHarelick at ise.com (Harelick, Matthew)
Date: Mon, 29 Dec 2008 16:18:16 -0500
Subject: [IronPython] Missing python libraries / modules from Iron Python 2.0
Message-ID: <2311471F8507984BBFCCB85ABD03B06823D5750EE4@OFFC-MAIL01.office.iseoptions.com>

Hi:

There is a list of Standard Library components not included in the MSI.

Does this mean that we can't import these libraries from a CPython library  or that they were just not implemented?

In other words is some of this functionality lost to IronPython users?

Thanks

Matthew Harelick
Test Automation Engineer

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081229/cb5febe0/attachment.html>

From mr_buchan at hotmail.com  Mon Dec 29 23:37:23 2008
From: mr_buchan at hotmail.com (Andrew Buchan)
Date: Mon, 29 Dec 2008 22:37:23 +0000
Subject: [IronPython] System.MissingMemberException / pyc.py
Message-ID: <BAY119-W23D2B65ED5EF5101C98D009CE60@phx.gbl>


I'm encountering problems when compiling files using the pyc sample for 2.0:
 
My (close to first) IronPython program runs fine when called from DOS console, but when compiled into an exe using pyc, I get the following error when running the exe (debug in Visual Studio 2008):
 System.MissingMemberException:  "'namespace#' object has no attribute 'File'" 
 
The piece of code this ties back to is:
 
def isFileValid(filename):    att = System.IO.File.GetAttributes(filename)    mask = att.value__    if ((mask & System.IO.FileAttributes.Hidden.value__) or (mask & System.IO.FileAttributes.System.value__)):       return False    else:       return True
 
The above is not in a class, its just a function in a module which is imported into the main py file.
 
I'm no expert on the matter, and as far as I can see it seems to be telling me that System.IO does not have an Attribute 'File', which according to MSDN and by inference from the fact the program works when called from DOS console, it does.
 
What bugs me about this (pun half intended) is that a presumably correctly compiled exe behaves differently to how the original py file ran via ipy.exe does with reference to using libraries (before I got to this bug I also had to change/rectify how I was importing standard libraries as the compiled exe requires you to specifically add the path to standard python lib, but ipy.exe interpreter doesn't...).
 
Have I stumbled upon an exception, or a small category of 'gotchas'; or is it common for a compiled IronPython application to run differently from the py files run using ipy.exe due to a number of possible reasons?
If the latter is true, then I see little point in running the py files during development/testing of small .exe applications, you'd be better compiling afresh every run.
 
The irony of all this is that the application in question is a simple GUI tool to navigate folders, select python files and either Run (calling IPY.exe), or Compile (using pyc and copying in the required dlls etc...), all to save me the bother of doing these tasks with the DOS console. 
 
Any clues as to where to go with the "'namespace#' object has no attribute 'File'" issue?
 

 
 
_________________________________________________________________
 Live Search presents Big Snap II - win John Lewis vouchers 
http://clk.atdmt.com/UKM/go/117442309/direct/01/
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081229/c9be17e6/attachment.html>

From curt at hagenlocher.org  Mon Dec 29 23:57:35 2008
From: curt at hagenlocher.org (Curt Hagenlocher)
Date: Mon, 29 Dec 2008 14:57:35 -0800
Subject: [IronPython] Missing python libraries / modules from Iron
	Python 2.0
In-Reply-To: <2311471F8507984BBFCCB85ABD03B06823D5750EE4@OFFC-MAIL01.office.iseoptions.com>
References: <2311471F8507984BBFCCB85ABD03B06823D5750EE4@OFFC-MAIL01.office.iseoptions.com>
Message-ID: <d2155e360812291457p77bd7d9es51adc5c09d5a9464@mail.gmail.com>

The files from the standard library that were left out of the MSI are the
ones that don't work correctly with IronPython.  We thought it would be
better to omit them than to include something which doesn't work.

On Mon, Dec 29, 2008 at 1:18 PM, Harelick, Matthew <MHarelick at ise.com>wrote:

>  Hi:
>
>
>
> There is a list of Standard Library components not included in the MSI.
>
>
>
> Does this mean that we can't import these libraries from a CPython library
> or that they were just not implemented?
>
>
>
> In other words is some of this functionality lost to IronPython users?
>
>
>
> Thanks
>
>
>
> Matthew Harelick
>
> Test Automation Engineer
>
>
>
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081229/b2bc2cba/attachment.html>

From stephen at theleengs.com  Tue Dec 30 04:16:09 2008
From: stephen at theleengs.com (Stephen Ng)
Date: Mon, 29 Dec 2008 19:16:09 -0800
Subject: [IronPython] Hosting IronPython in Powershell
Message-ID: <86122ad70812291916gca274ebwa83f01a61bc542e5@mail.gmail.com>

Hello IronPython users,
I have a largish Powershell script I'd like to migrate to IronPython.
 (Powershell is a fine language, but there are a lot more Python developers
where I work).

The Powershell script manipulates VMware ESX machines, using their VI
Toolkit (http://blogs.vmware.com/vipowershell/).

So, I'm hoping that I can have my Powershell script host IronPython, and
have a minimum of Powershell to create objects which I then pass over to
IronPython and have IronPython do most of the work.

Does this approach sound reasonable?  I've done a fair amount of Powershell
and Python programming, but I'm not that familiar with IronPython (or .Net
even).  I'm having trouble even getting started.  I saw this blog entry (
http://stevegilham.blogspot.com/2008/02/embedding-powershell-in-ironpython.html)
but "$pythonEngine = [ironpython.hosting.pythonengine]::CurrentEngine" seems
out of date.  I changed it to CreateEngine(), but now I'm getting an error
loading Microsoft.Scripting.  Doing a loadFrom of Microsoft.Scripting*.dll
from the IronPython dir doesn't seem to help.

PS C:\Documents and Settings\sng>
[reflection.assembly]::loadFrom("C:\Program Files\IronPython
2.0\IronPython.dll")
GAC    Version        Location
---    -------        --------
True   v2.0.50727
C:\WINDOWS\assembly\GAC_MSIL\IronPython\2.0.0.0__31bf3...

PS C:\Documents and Settings\sng>
[IronPython.Hosting.Python]::CreateEngine()
Exception retrieving member "CreateEngine": "Could not load file or assembly
'Microsoft.Scripting, Version=0.9.0.0,
Culture=neutral,PublicKeyToken=31bf3856ad364e35' or one of its dependencies.
The system cannot find the file specified."
At line:1 char:42
+ [IronPython.Hosting.Python]::CreateEngine( <<<< )

Any help would be greatly appreciated....

Thanks,

Steve
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081229/44c4defc/attachment.html>

From empirebuilder at gmail.com  Tue Dec 30 12:33:26 2008
From: empirebuilder at gmail.com (Dody Gunawinata)
Date: Tue, 30 Dec 2008 13:33:26 +0200
Subject: [IronPython] Memory Leak in CreateScriptSource?
In-Reply-To: <8cd017b80812260736j3ed4927bw5f1209f93370d141@mail.gmail.com>
References: <49538CB3.70706@gmx.de>
	<8cd017b80812260731r44d7b4d2h9f8ae06a79e0459d@mail.gmail.com>
	<8cd017b80812260736j3ed4927bw5f1209f93370d141@mail.gmail.com>
Message-ID: <8cd017b80812300333k461f5924l4d37c68e73028dd2@mail.gmail.com>

I reported this bug on codeplex
http://www.codeplex.com/IronPython/WorkItem/View.aspx?WorkItemId=20399


On Fri, Dec 26, 2008 at 5:36 PM, Dody Gunawinata <empirebuilder at gmail.com>wrote:

> With the GC.Collect lines removed, this is how it looks like (after
> counting to 40,000). I started the tasklist command after around 25K.
>
> Image Name                     PID Session Name        Session#    Mem
> Usage
> ========================= ======== ================ ===========
> ============
> IronPythonLeaker.exe          5296 Console                    1    146,840
> K
>
> Image Name                     PID Session Name        Session#    Mem
> Usage
> ========================= ======== ================ ===========
> ============
> IronPythonLeaker.exe          5296 Console                    1    147,948
> K
>
> Image Name                     PID Session Name        Session#    Mem
> Usage
> ========================= ======== ================ ===========
> ============
> IronPythonLeaker.exe          5296 Console                    1    148,344
> K
>
> Image Name                     PID Session Name        Session#    Mem
> Usage
> ========================= ======== ================ ===========
> ============
> IronPythonLeaker.exe          5296 Console                    1    153,616
> K
>
> Image Name                     PID Session Name        Session#    Mem
> Usage
> ========================= ======== ================ ===========
> ============
> IronPythonLeaker.exe          5296 Console                    1    154,064
> K
>
> Image Name                     PID Session Name        Session#    Mem
> Usage
> ========================= ======== ================ ===========
> ============
> IronPythonLeaker.exe          5296 Console                    1    156,364
> K
>
> Image Name                     PID Session Name        Session#    Mem
> Usage
> ========================= ======== ================ ===========
> ============
> IronPythonLeaker.exe          5296 Console                    1    160,612
> K
>
> Image Name                     PID Session Name        Session#    Mem
> Usage
> ========================= ======== ================ ===========
> ============
> IronPythonLeaker.exe          5296 Console                    1    174,568
> K
>
> Image Name                     PID Session Name        Session#    Mem
> Usage
> ========================= ======== ================ ===========
> ============
> IronPythonLeaker.exe          5296 Console                    1    180,232
> K
>
> Image Name                     PID Session Name        Session#    Mem
> Usage
> ========================= ======== ================ ===========
> ============
> IronPythonLeaker.exe          5296 Console                    1    189,904
> K
>
> Image Name                     PID Session Name        Session#    Mem
> Usage
> ========================= ======== ================ ===========
> ============
> IronPythonLeaker.exe          5296 Console                    1    192,856
> K
>
> Image Name                     PID Session Name        Session#    Mem
> Usage
> ========================= ======== ================ ===========
> ============
> IronPythonLeaker.exe          5296 Console                    1    198,900
> K
>
> Image Name                     PID Session Name        Session#    Mem
> Usage
> ========================= ======== ================ ===========
> ============
> IronPythonLeaker.exe          5296 Console                    1    201,572
> K
>
> Image Name                     PID Session Name        Session#    Mem
> Usage
> ========================= ======== ================ ===========
> ============
> IronPythonLeaker.exe          5296 Console                    1    201,604
> K
>
>
> On Fri, Dec 26, 2008 at 5:31 PM, Dody Gunawinata <empirebuilder at gmail.com>wrote:
>
>> I run above code in a console using IronPython 2.0 Final Version and here
>> is the output (on Vista) using Tasklist command. The final result doubles
>> the in memory size; I stopped the measurement when it hits 12,000
>>
>> Image Name                     PID Session Name        Session#    Mem
>> Usage
>> ========================= ======== ================ ===========
>> ============
>> IronPythonLeaker.exe          5248 Console                    1     33,532
>> K
>>
>> Image Name                     PID Session Name        Session#    Mem
>> Usage
>> ========================= ======== ================ ===========
>> ============
>> IronPythonLeaker.exe          5248 Console                    1     39,204
>> K
>>
>> Image Name                     PID Session Name        Session#    Mem
>> Usage
>> ========================= ======== ================ ===========
>> ============
>> IronPythonLeaker.exe          5248 Console                    1     43,692
>> K
>>
>> Image Name                     PID Session Name        Session#    Mem
>> Usage
>> ========================= ======== ================ ===========
>> ============
>> IronPythonLeaker.exe          5248 Console                    1     42,952
>> K
>>
>> Image Name                     PID Session Name        Session#    Mem
>> Usage
>> ========================= ======== ================ ===========
>> ============
>> IronPythonLeaker.exe          5248 Console                    1     45,888
>> K
>>
>> Image Name                     PID Session Name        Session#    Mem
>> Usage
>> ========================= ======== ================ ===========
>> ============
>> IronPythonLeaker.exe          5248 Console                    1     45,928
>> K
>>
>> Image Name                     PID Session Name        Session#    Mem
>> Usage
>> ========================= ======== ================ ===========
>> ============
>> IronPythonLeaker.exe          5248 Console                    1     45,956
>> K
>>
>> Image Name                     PID Session Name        Session#    Mem
>> Usage
>> ========================= ======== ================ ===========
>> ============
>> IronPythonLeaker.exe          5248 Console                    1     44,992
>> K
>>
>> Image Name                     PID Session Name        Session#    Mem
>> Usage
>> ========================= ======== ================ ===========
>> ============
>> IronPythonLeaker.exe          5248 Console                    1     49,520
>> K
>>
>> Image Name                     PID Session Name        Session#    Mem
>> Usage
>> ========================= ======== ================ ===========
>> ============
>> IronPythonLeaker.exe          5248 Console                    1     50,592
>> K
>>
>> Image Name                     PID Session Name        Session#    Mem
>> Usage
>> ========================= ======== ================ ===========
>> ============
>> IronPythonLeaker.exe          5248 Console                    1     50,636
>> K
>>
>> Image Name                     PID Session Name        Session#    Mem
>> Usage
>> ========================= ======== ================ ===========
>> ============
>> IronPythonLeaker.exe          5248 Console                    1     52,632
>> K
>>
>> Image Name                     PID Session Name        Session#    Mem
>> Usage
>> ========================= ======== ================ ===========
>> ============
>> IronPythonLeaker.exe          5248 Console                    1     52,788
>> K
>>
>> Image Name                     PID Session Name        Session#    Mem
>> Usage
>> ========================= ======== ================ ===========
>> ============
>> IronPythonLeaker.exe          5248 Console                    1     52,884
>> K
>>
>> Image Name                     PID Session Name        Session#    Mem
>> Usage
>> ========================= ======== ================ ===========
>> ============
>> IronPythonLeaker.exe          5248 Console                    1     53,032
>> K
>>
>> Image Name                     PID Session Name        Session#    Mem
>> Usage
>> ========================= ======== ================ ===========
>> ============
>> IronPythonLeaker.exe          5248 Console                    1     62,916
>> K
>>
>> Image Name                     PID Session Name        Session#    Mem
>> Usage
>> ========================= ======== ================ ===========
>> ============
>> IronPythonLeaker.exe          5248 Console                    1     62,456
>> K
>>
>> Image Name                     PID Session Name        Session#    Mem
>> Usage
>> ========================= ======== ================ ===========
>> ============
>> IronPythonLeaker.exe          5248 Console                    1     67,668
>> K
>>
>> Image Name                     PID Session Name        Session#    Mem
>> Usage
>> ========================= ======== ================ ===========
>> ============
>> IronPythonLeaker.exe          5248 Console                    1     65,880
>> K
>>
>> Image Name                     PID Session Name        Session#    Mem
>> Usage
>> ========================= ======== ================ ===========
>> ============
>> IronPythonLeaker.exe          5248 Console                    1     72,260
>> K
>>
>> Image Name                     PID Session Name        Session#    Mem
>> Usage
>> ========================= ======== ================ ===========
>> ============
>> IronPythonLeaker.exe          5248 Console                    1     73,644
>> K
>>
>>
>> On Thu, Dec 25, 2008 at 3:37 PM, Christian Schmidt <
>> christian2.schmidt at gmx.de> wrote:
>>
>>> Hi,
>>>
>>> the following code snippet allocates constantly more and more memory.
>>> Moving .Compile() outside the loop works as expected.
>>>
>>> private static void LeakTest()
>>> {
>>>  var runtime = Python.CreateRuntime();
>>>  var engine = runtime.GetEngine("py");
>>>  var sum = 0.0;
>>>
>>>  while (true)
>>>  {
>>>    var code = engine.CreateScriptSourceFromString("1.0+2.0").Compile();
>>>    sum += code.Execute<double>();
>>>    GC.Collect();
>>>    GC.WaitForPendingFinalizers();
>>>    GC.Collect();
>>>    Console.WriteLine(sum);
>>>  }
>>> }
>>>
>>> Any ideas?
>>>
>>> Merry Christmas,
>>>
>>> Christian
>>> _______________________________________________
>>> Users mailing list
>>> Users at lists.ironpython.com
>>> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>>>
>>
>>
>>
>> --
>> nomadlife.org
>>
>>
>
>
> --
> nomadlife.org
>
>


-- 
nomadlife.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081230/5df6e60f/attachment.html>

From dan.eloff at gmail.com  Tue Dec 30 16:54:10 2008
From: dan.eloff at gmail.com (Dan Eloff)
Date: Tue, 30 Dec 2008 10:54:10 -0500
Subject: [IronPython] How do you call a generic method that requires more
	than one generic argument?
Message-ID: <4817b6fc0812300754h1cb87621jd5068fd1a0fc4e89@mail.gmail.com>

>>> from System import Action
>>> f = lambda x, y: x+y
>>> Action[object, object](f)
ValueError: The type or method has 1 generic parameter(s), but 2
generic argument(s) were provided. A generic argument must be provided
for each generic parameter.
>>> Action[(object, object)](f)
ValueError: The type or method has 1 generic parameter(s), but 2
generic argument(s) were provided. A generic argument must be provided
for each generic parameter.
>>> Action[[object,object]](f)
TypeError: expected Array[Type], got list
>>> from System import Array, Type
>>> Action[Array[Type]([object, object])](f)
ValueError: The type or method has 1 generic parameter(s), but 2
generic argument(s) were provided. A generic argument must be provided
for each generic parameter.

Is this a bug?

Thanks,
-Dan


From curt at hagenlocher.org  Tue Dec 30 17:07:56 2008
From: curt at hagenlocher.org (Curt Hagenlocher)
Date: Tue, 30 Dec 2008 08:07:56 -0800
Subject: [IronPython] How do you call a generic method that requires
	more than one generic argument?
In-Reply-To: <4817b6fc0812300754h1cb87621jd5068fd1a0fc4e89@mail.gmail.com>
References: <4817b6fc0812300754h1cb87621jd5068fd1a0fc4e89@mail.gmail.com>
Message-ID: <d2155e360812300807u3b401067gccf69d6281be1856@mail.gmail.com>

Did you AddReference System.Core.dll?

On Tue, Dec 30, 2008 at 7:54 AM, Dan Eloff <dan.eloff at gmail.com> wrote:

> >>> from System import Action
> >>> f = lambda x, y: x+y
> >>> Action[object, object](f)
> ValueError: The type or method has 1 generic parameter(s), but 2
> generic argument(s) were provided. A generic argument must be provided
> for each generic parameter.
> >>> Action[(object, object)](f)
> ValueError: The type or method has 1 generic parameter(s), but 2
> generic argument(s) were provided. A generic argument must be provided
> for each generic parameter.
> >>> Action[[object,object]](f)
> TypeError: expected Array[Type], got list
> >>> from System import Array, Type
> >>> Action[Array[Type]([object, object])](f)
> ValueError: The type or method has 1 generic parameter(s), but 2
> generic argument(s) were provided. A generic argument must be provided
> for each generic parameter.
>
> Is this a bug?
>
> Thanks,
> -Dan
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081230/51d16c1b/attachment.html>

From dan.eloff at gmail.com  Tue Dec 30 17:25:32 2008
From: dan.eloff at gmail.com (Dan Eloff)
Date: Tue, 30 Dec 2008 11:25:32 -0500
Subject: [IronPython] How do you call a generic method that requires
	more than one generic argument?
In-Reply-To: <d2155e360812300807u3b401067gccf69d6281be1856@mail.gmail.com>
References: <4817b6fc0812300754h1cb87621jd5068fd1a0fc4e89@mail.gmail.com>
	<d2155e360812300807u3b401067gccf69d6281be1856@mail.gmail.com>
Message-ID: <4817b6fc0812300825h409bac7fw50d88fc0c42da33a@mail.gmail.com>

On Tue, Dec 30, 2008 at 11:07 AM, Curt Hagenlocher <curt at hagenlocher.org> wrote:
> Did you AddReference System.Core.dll?
>

Thanks, that was it. Second time that's got me.

-Dan


From curt at hagenlocher.org  Tue Dec 30 18:01:21 2008
From: curt at hagenlocher.org (Curt Hagenlocher)
Date: Tue, 30 Dec 2008 09:01:21 -0800
Subject: [IronPython] Hosting IronPython in Powershell
In-Reply-To: <86122ad70812291916gca274ebwa83f01a61bc542e5@mail.gmail.com>
References: <86122ad70812291916gca274ebwa83f01a61bc542e5@mail.gmail.com>
Message-ID: <d2155e360812300901y103d4002q9370432fd152e9f8@mail.gmail.com>

This looks like a classic and continually frustrating Fusion problem.
 (Fusion is the .NET loader.)  When you use Assembly.LoadFrom, it doesn't
automatically add the assembly's directory into any kind of search path.  So
the dependencies of IronPython.dll can't automatically be found to be
loaded.
Your best bet for solving this is to put IronPython and all its dependencies
into the GAC.  It might also work to put the IronPython directory into the
path.

On Mon, Dec 29, 2008 at 7:16 PM, Stephen Ng <stephen at theleengs.com> wrote:

> Hello IronPython users,
> I have a largish Powershell script I'd like to migrate to IronPython.
>  (Powershell is a fine language, but there are a lot more Python developers
> where I work).
>
>  The Powershell script manipulates VMware ESX machines, using their VI
> Toolkit (http://blogs.vmware.com/vipowershell/).
>
> So, I'm hoping that I can have my Powershell script host IronPython, and
> have a minimum of Powershell to create objects which I then pass over to
> IronPython and have IronPython do most of the work.
>
> Does this approach sound reasonable?  I've done a fair amount of Powershell
> and Python programming, but I'm not that familiar with IronPython (or .Net
> even).  I'm having trouble even getting started.  I saw this blog entry (
> http://stevegilham.blogspot.com/2008/02/embedding-powershell-in-ironpython.html)
> but "$pythonEngine = [ironpython.hosting.pythonengine]::CurrentEngine" seems
> out of date.  I changed it to CreateEngine(), but now I'm getting an error
> loading Microsoft.Scripting.  Doing a loadFrom of Microsoft.Scripting*.dll
> from the IronPython dir doesn't seem to help.
>
> PS C:\Documents and Settings\sng>
> [reflection.assembly]::loadFrom("C:\Program Files\IronPython
> 2.0\IronPython.dll")
> GAC    Version        Location
> ---    -------        --------
> True   v2.0.50727
> C:\WINDOWS\assembly\GAC_MSIL\IronPython\2.0.0.0__31bf3...
>
> PS C:\Documents and Settings\sng>
> [IronPython.Hosting.Python]::CreateEngine()
> Exception retrieving member "CreateEngine": "Could not load file or
> assembly 'Microsoft.Scripting, Version=0.9.0.0,
> Culture=neutral,PublicKeyToken=31bf3856ad364e35' or one of its dependencies.
> The system cannot find the file specified."
> At line:1 char:42
> + [IronPython.Hosting.Python]::CreateEngine( <<<< )
>
> Any help would be greatly appreciated....
>
> Thanks,
>
> Steve
>
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081230/17395d1e/attachment.html>

From curt at hagenlocher.org  Tue Dec 30 18:02:57 2008
From: curt at hagenlocher.org (Curt Hagenlocher)
Date: Tue, 30 Dec 2008 09:02:57 -0800
Subject: [IronPython] Bug in Django's lazy functions
In-Reply-To: <ea89d0e50812291135p318ac758x5ba83a97713c5937@mail.gmail.com>
References: <ea89d0e50812281641m1bfe0560g5849049df386722d@mail.gmail.com>
	<d2155e360812290843j8314024k5c02647fd2da8a8a@mail.gmail.com>
	<ea89d0e50812291135p318ac758x5ba83a97713c5937@mail.gmail.com>
Message-ID: <d2155e360812300902n275d45efh89e28ffeb0344716@mail.gmail.com>

On Mon, Dec 29, 2008 at 11:35 AM, Jeff Hardy <jdhardy at gmail.com> wrote:

> On Mon, Dec 29, 2008 at 9:43 AM, Curt Hagenlocher <curt at hagenlocher.org>
> wrote:
>
> > There may also be a problem as a result of str == unicode; in fact,
> there's
> > a commented-out assertion in line 175 of functional.py that specifically
> > breaks because this is true.
>
> functional.py isn't *exactly* the same as Django's - I commented that
> line out. It does actually have a purpose when str != unicode. There's
> code that hits CP #20366 as well.


If nothing else, the code should at least be rearranged to put the tests for
unicode and str in the same order in every place they appear.

--
Curt Hagenlocher
curt at hagenlocher.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081230/82cff4c3/attachment.html>

From curt at hagenlocher.org  Tue Dec 30 18:13:25 2008
From: curt at hagenlocher.org (Curt Hagenlocher)
Date: Tue, 30 Dec 2008 09:13:25 -0800
Subject: [IronPython] Problems with hashlib and threading
In-Reply-To: <4a3ced360812201356i4914639gefa19fcb8d326d4f@mail.gmail.com>
References: <4a3ced360812201356i4914639gefa19fcb8d326d4f@mail.gmail.com>
Message-ID: <d2155e360812300913g337d08d4o44e913b83711aa91@mail.gmail.com>

I'm going to guess that this is because we use a static instance
of MD5CryptoServiceProvider to do the hashing, something which may not be
thread-safe.  It looks like all of the IronPython crypto implementations do
the same thing, so the SHA functions may have a similar issue.
Please file this as a bug on Codeplex.

On Sat, Dec 20, 2008 at 1:56 PM, Brian Merrell <brian at merrells.org> wrote:

> I get the following error periodically:
>
> Exception in thread Thread-2:
> Traceback (most recent call last):
>   File "C:\Program Files\IronPython 2.0\Lib\threading.py", line 486, in
> _Thread__bootstrap_inner
>     self.run()
>   File "initial_ingest.py", line 33, in run
>     md5 = hashlib.md5(contents).hexdigest()
> SystemError: Hash not valid for use in specified state.
>
> I am not sharing a md5() object between threads and have tried to avoid any
> shared memory issues:
>
>
> class Convert(threading.Thread):
>     def __init__(self, queue):
>         threading.Thread.__init__(self)
>         self.queue = queue
>     def run(self):
>         import hashlib
>         while True:
>             abs_filename = self.queue.get()
>             f = open(abs_filename, "rb")
>             contents = f.read()
>             f.close()
>             md5 = hashlib.md5(contents).hexdigest()
>             print md5, abs_filename
>             self.queue.task_done()
>
> Any ideas why this is failing?  Thanks,
>
> -brian
>
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081230/d2fedf90/attachment.html>

From beppler at gmail.com  Tue Dec 30 18:13:55 2008
From: beppler at gmail.com (Carlos Alberto Costa Beppler)
Date: Tue, 30 Dec 2008 15:13:55 -0200
Subject: [IronPython] Problems in module binascii
Message-ID: <b0057c270812300913gf8f7ad6l1d1f76bcad979150@mail.gmail.com>

Hello.

The function crc32 contained on module binascii requires 2 arguments,
but the python docs says that the second argument is optional.

Then a was looking for the implementation and found it in binascii.cs
file on IronPython.Modules. But the implementation of this function
and others (a2b_qp, b2a_qp, a2b_hqx, rledecode_hqx, b2a_hqx and
crc_hqx) are implemented to throw NotImplementdException.

Is this a bug or this module is not supported on IronPython?


From curt at hagenlocher.org  Tue Dec 30 18:19:15 2008
From: curt at hagenlocher.org (Curt Hagenlocher)
Date: Tue, 30 Dec 2008 09:19:15 -0800
Subject: [IronPython] Problems in module binascii
In-Reply-To: <b0057c270812300913gf8f7ad6l1d1f76bcad979150@mail.gmail.com>
References: <b0057c270812300913gf8f7ad6l1d1f76bcad979150@mail.gmail.com>
Message-ID: <d2155e360812300919x26137357xa88ccd898454d665@mail.gmail.com>

It throws a NotImplementedException because it hasn't been implemented yet.
:)
You can vote on this feature at
http://www.codeplex.com/IronPython/WorkItem/View.aspx?WorkItemId=17465

On Tue, Dec 30, 2008 at 9:13 AM, Carlos Alberto Costa Beppler <
beppler at gmail.com> wrote:

> Hello.
>
> The function crc32 contained on module binascii requires 2 arguments,
> but the python docs says that the second argument is optional.
>
> Then a was looking for the implementation and found it in binascii.cs
> file on IronPython.Modules. But the implementation of this function
> and others (a2b_qp, b2a_qp, a2b_hqx, rledecode_hqx, b2a_hqx and
> crc_hqx) are implemented to throw NotImplementdException.
>
> Is this a bug or this module is not supported on IronPython?
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081230/d170a524/attachment.html>

From beppler at gmail.com  Tue Dec 30 18:42:52 2008
From: beppler at gmail.com (Carlos Alberto Costa Beppler)
Date: Tue, 30 Dec 2008 15:42:52 -0200
Subject: [IronPython] Problems in module binascii
In-Reply-To: <d2155e360812300919x26137357xa88ccd898454d665@mail.gmail.com>
References: <b0057c270812300913gf8f7ad6l1d1f76bcad979150@mail.gmail.com>
	<d2155e360812300919x26137357xa88ccd898454d665@mail.gmail.com>
Message-ID: <b0057c270812300942g60bea92fl383983f241995a01@mail.gmail.com>

Thanks.

On Tue, Dec 30, 2008 at 15:19, Curt Hagenlocher <curt at hagenlocher.org> wrote:
> It throws a NotImplementedException because it hasn't been implemented yet.
> :)
> You can vote on this feature
> at http://www.codeplex.com/IronPython/WorkItem/View.aspx?WorkItemId=17465
>
> On Tue, Dec 30, 2008 at 9:13 AM, Carlos Alberto Costa Beppler
> <beppler at gmail.com> wrote:
>>
>> Hello.
>>
>> The function crc32 contained on module binascii requires 2 arguments,
>> but the python docs says that the second argument is optional.
>>
>> Then a was looking for the implementation and found it in binascii.cs
>> file on IronPython.Modules. But the implementation of this function
>> and others (a2b_qp, b2a_qp, a2b_hqx, rledecode_hqx, b2a_hqx and
>> crc_hqx) are implemented to throw NotImplementdException.
>>
>> Is this a bug or this module is not supported on IronPython?
>> _______________________________________________
>> Users mailing list
>> Users at lists.ironpython.com
>> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>
>
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>
>


From stephen at theleengs.com  Tue Dec 30 21:39:27 2008
From: stephen at theleengs.com (Stephen Ng)
Date: Tue, 30 Dec 2008 12:39:27 -0800
Subject: [IronPython] Hosting IronPython in Powershell
In-Reply-To: <d2155e360812300901y103d4002q9370432fd152e9f8@mail.gmail.com>
References: <86122ad70812291916gca274ebwa83f01a61bc542e5@mail.gmail.com>
	<d2155e360812300901y103d4002q9370432fd152e9f8@mail.gmail.com>
Message-ID: <86122ad70812301239w1cc675a9r8159907b4066b3ea@mail.gmail.com>

Thanks, dropping all the dlls from the IronPython2 directory into the GAC
gets me a little further (adding to the path didn't work).  Sorry to need to
be led by the hand...now I get:
PS C:\Program Files\IronPython 2.0>
[reflection.assembly]::loadFrom("C:\Program Files\IronPython
2.0\IronPython.dll")

GAC    Version        Location
---    -------        --------
True   v2.0.50727
C:\WINDOWS\assembly\GAC_MSIL\IronPython\2.0.0.0__31bf3...

PS C:\Program Files\IronPython 2.0> $py =
[IronPython.Hosting.Python]::CreateEngine()
PS C:\Program Files\IronPython 2.0> $py.Execute("3")
Exception calling "Execute" with "1" argument(s): "Late bound operations
cannot
 be performed on types or methods for which ContainsGenericParameters is
true."
At line:1 char:12
+ $py.Execute( <<<< "3")
PS C:\Program Files\IronPython 2.0> $py.Execute("print 'hello world!'")
Exception calling "Execute" with "1" argument(s): "Late bound operations
cannot be performed on types or methods for which ContainsGenericParameters
is true."
At line:1 char:12
+ $py.Execute( <<<< "print 'hello world!'")
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081230/654de131/attachment.html>

From curt at hagenlocher.org  Tue Dec 30 21:43:55 2008
From: curt at hagenlocher.org (Curt Hagenlocher)
Date: Tue, 30 Dec 2008 12:43:55 -0800
Subject: [IronPython] Hosting IronPython in Powershell
In-Reply-To: <86122ad70812301239w1cc675a9r8159907b4066b3ea@mail.gmail.com>
References: <86122ad70812291916gca274ebwa83f01a61bc542e5@mail.gmail.com>
	<d2155e360812300901y103d4002q9370432fd152e9f8@mail.gmail.com>
	<86122ad70812301239w1cc675a9r8159907b4066b3ea@mail.gmail.com>
Message-ID: <d2155e360812301243i1ecdead4of1a7acf21c6db228@mail.gmail.com>

This is a PowerShell error.  The DLR hosting interfaces overload Execute
(and other methods) with generic and non-generic versions, and it appears
that PowerShell won't allow calling of the generic version.  For this to
work, you'll need to find a way to force PowerShell to use the non-generic
overload.  I'm afraid my PS skills aren't up to that task.

On Tue, Dec 30, 2008 at 12:39 PM, Stephen Ng <stephen at theleengs.com> wrote:

> Thanks, dropping all the dlls from the IronPython2 directory into the GAC
> gets me a little further (adding to the path didn't work).  Sorry to need to
> be led by the hand...now I get:
> PS C:\Program Files\IronPython 2.0>
> [reflection.assembly]::loadFrom("C:\Program Files\IronPython
> 2.0\IronPython.dll")
>
> GAC    Version        Location
> ---    -------        --------
> True   v2.0.50727
> C:\WINDOWS\assembly\GAC_MSIL\IronPython\2.0.0.0__31bf3...
>
> PS C:\Program Files\IronPython 2.0> $py =
> [IronPython.Hosting.Python]::CreateEngine()
> PS C:\Program Files\IronPython 2.0> $py.Execute("3")
> Exception calling "Execute" with "1" argument(s): "Late bound operations
> cannot
>  be performed on types or methods for which ContainsGenericParameters is
> true."
> At line:1 char:12
> + $py.Execute( <<<< "3")
> PS C:\Program Files\IronPython 2.0> $py.Execute("print 'hello world!'")
> Exception calling "Execute" with "1" argument(s): "Late bound operations
> cannot be performed on types or methods for which ContainsGenericParameters
> is true."
> At line:1 char:12
> + $py.Execute( <<<< "print 'hello world!'")
>
>
>
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081230/4c3dd6b7/attachment.html>

From fuzzyman at voidspace.org.uk  Tue Dec 30 21:48:59 2008
From: fuzzyman at voidspace.org.uk (Michael Foord)
Date: Tue, 30 Dec 2008 20:48:59 +0000
Subject: [IronPython] Hosting IronPython in Powershell
In-Reply-To: <d2155e360812301243i1ecdead4of1a7acf21c6db228@mail.gmail.com>
References: <86122ad70812291916gca274ebwa83f01a61bc542e5@mail.gmail.com>	<d2155e360812300901y103d4002q9370432fd152e9f8@mail.gmail.com>	<86122ad70812301239w1cc675a9r8159907b4066b3ea@mail.gmail.com>
	<d2155e360812301243i1ecdead4of1a7acf21c6db228@mail.gmail.com>
Message-ID: <495A893B.5060704@voidspace.org.uk>

Curt Hagenlocher wrote:
> This is a PowerShell error.  The DLR hosting interfaces overload 
> Execute (and other methods) with generic and non-generic versions, and 
> it appears that PowerShell won't allow calling of the generic version. 
>  For this to work, you'll need to find a way to force PowerShell to 
> use the non-generic overload.  I'm afraid my PS skills aren't up to 
> that task.
>

This topic is covered in IronPython in Action. :-)

I get round the problem using Lee Holmes script that uses reflection for 
calling generic methods:

http://www.leeholmes.com/blog/InvokingGenericMethodsOnNonGenericClassesInPowerShell.aspx

All the best,

Michael Foord

> On Tue, Dec 30, 2008 at 12:39 PM, Stephen Ng <stephen at theleengs.com 
> <mailto:stephen at theleengs.com>> wrote:
>
>     Thanks, dropping all the dlls from the IronPython2 directory into
>     the GAC gets me a little further (adding to the path didn't work).
>      Sorry to need to be led by the hand...now I get:
>
>     PS C:\Program Files\IronPython 2.0>
>     [reflection.assembly]::loadFrom("C:\Program Files\IronPython
>     2.0\IronPython.dll")
>
>     GAC    Version        Location
>     ---    -------        --------
>     True   v2.0.50727    
>     C:\WINDOWS\assembly\GAC_MSIL\IronPython\2.0.0.0__31bf3...
>
>     PS C:\Program Files\IronPython 2.0> $py =
>     [IronPython.Hosting.Python]::CreateEngine()
>     PS C:\Program Files\IronPython 2.0> $py.Execute("3")
>     Exception calling "Execute" with "1" argument(s): "Late bound
>     operations cannot
>      be performed on types or methods for which
>     ContainsGenericParameters is true."
>     At line:1 char:12
>     + $py.Execute( <<<< "3")
>     PS C:\Program Files\IronPython 2.0> $py.Execute("print 'hello
>     world!'")
>     Exception calling "Execute" with "1" argument(s): "Late bound
>     operations cannot be performed on types or methods for which
>     ContainsGenericParameters is true."
>     At line:1 char:12
>     + $py.Execute( <<<< "print 'hello world!'")
>
>
>
>     _______________________________________________
>     Users mailing list
>     Users at lists.ironpython.com <mailto:Users at lists.ironpython.com>
>     http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>
>
> ------------------------------------------------------------------------
>
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>   


-- 
http://www.ironpythoninaction.com/
http://www.voidspace.org.uk/blog




From fuzzyman at voidspace.org.uk  Tue Dec 30 23:04:53 2008
From: fuzzyman at voidspace.org.uk (Michael Foord)
Date: Tue, 30 Dec 2008 22:04:53 +0000
Subject: [IronPython] Embedding IronPython
In-Reply-To: <20081225210901.F261A1C70480@smtp.webfaction.com>
References: <20081225210901.F261A1C70480@smtp.webfaction.com>
Message-ID: <495A9B05.8060703@voidspace.org.uk>

Hello Josh,

Sorry for the delayed reply - life is hectic at the moment and I'm still 
copy editing IronPython in Action.

I've copied your email to the IronPython list as I don't know whether 
the CreateScope and other methods of the ScriptEngine are thread safe. 
My instinct would be to say no, but on the other hand the DLR has to 
handle running multi-threaded code so maybe...

Michael Foord

greenbergjosh at hotmail.com wrote:
> Some one has sent you a message via your contact form. They filled in the
> following values :
>
> comment = I read your article on IronPython and I was hoping you might have an answer to the following question.
>
> I would like to embed IronPython in a web application.  I would like to initialize the engine and runtime in the Application_Start() method and store an instance of the ScriptEngine in the Application object.  Then, any time I need to run some python I intend to call the ScriptEngine::CreateScope() method and use it to get a new scope in which to execute my code.  Is this approach thread safe?  I will have only one instance of the ScriptEngine shared by many threads.  Each thread will call CreateScope whenever it chooses.
>
>  
>
> It will look something like:
>
>  
>
> protected void Application_Start(object sender, EventArgs e)
>
> {
>
> ScriptRuntime truntime = ScriptRuntime.Create();
>
> ScriptEngine teng = truntime.GetEngine("py");
>
> this.Application.Add(???PythonEngine???, teng);
>
> }
>
>  
>
> Then, in some web page or dll called by a web page I would do
>
>  
>
> ..
>
> ScriptEngine se = (ScriptEngine)Application[???PythonEngine???];
>
> ScriptScope ss = se.CreateScope();
>
> ScriptSource src = se.CreateScriptSourceFromString(???y=24???, SourceCodeKind.Statements);
>
> CompiledCode cc = src.Compile();
>
> cc.Execute(scope);
>
> ...
>
>  
>
> Is this going to work or will I run into threading problems?  In other words, is the ScriptEngine thread-safe?  Will it serialize all of my cc.Execute() calls?  Any other insight into its behavior in this regard would be appreciated.
>
>  
>
> Many Thanks,
>
> Josh
>
> name = Josh
> email = greenbergjosh at hotmail.com
> subject = Embedding IronPython
>   


-- 
http://www.ironpythoninaction.com/
http://www.voidspace.org.uk/blog




From stephen at theleengs.com  Wed Dec 31 01:14:44 2008
From: stephen at theleengs.com (Stephen Ng)
Date: Tue, 30 Dec 2008 16:14:44 -0800
Subject: [IronPython] Hosting IronPython in Powershell
In-Reply-To: <495A893B.5060704@voidspace.org.uk>
References: <86122ad70812291916gca274ebwa83f01a61bc542e5@mail.gmail.com>
	<d2155e360812300901y103d4002q9370432fd152e9f8@mail.gmail.com>
	<86122ad70812301239w1cc675a9r8159907b4066b3ea@mail.gmail.com>
	<d2155e360812301243i1ecdead4of1a7acf21c6db228@mail.gmail.com>
	<495A893B.5060704@voidspace.org.uk>
Message-ID: <86122ad70812301614y3264700dx73dbe363ef4968da@mail.gmail.com>

Thanks for the help....  I'll get the pdf version of the book this weekend
and read up to try to get more of a clue.  Am I using Invoke-GenericMethod
correctly?

PS C:\Program Files\IronPython 2.0> & 'C:\Program Files\IronPython
2.0\Invoke-GenericMethod.ps1' $py Execute String '3'
Exception calling "Invoke" with "2" argument(s): "Exception has been thrown
by the target of an invocation."
At C:\Program Files\IronPython 2.0\Invoke-GenericMethod.ps1:96 char:30
+             $newMethod.Invoke( <<<< $instance, $methodParameters)
PS C:\Program Files\IronPython 2.0> & 'C:\Program Files\IronPython
2.0\Invoke-GenericMethod.ps1' $py Execute String 'print 3'
Exception calling "Invoke" with "2" argument(s): "Exception has been thrown
by the target of an invocation."
At C:\Program Files\IronPython 2.0\Invoke-GenericMethod.ps1:96 char:30
+             $newMethod.Invoke( <<<< $instance, $methodParameters)
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081230/c32cff49/attachment.html>

From jdhardy at gmail.com  Wed Dec 31 06:00:35 2008
From: jdhardy at gmail.com (Jeff Hardy)
Date: Tue, 30 Dec 2008 22:00:35 -0700
Subject: [IronPython] Frames in IronPython
Message-ID: <ea89d0e50812302100o7b2f4ba3l2383e9396438ff52@mail.gmail.com>

I don't know what the feature list is going to be for 2.1, but I'd
like to see frame objects near the top. Almost every piece of Python
code I've tried uses them in some way. Often it's for debugging
purposes (and can be worked around), but not always. Trac has this
lovely piece of code:

@staticmethod
    def implements(*interfaces):
        """Can be used in the class definiton of `Component` subclasses to
        declare the extension points that are extended.
        """
        import sys

        frame = sys._getframe(1)
        locals_ = frame.f_locals

        # Some sanity checks
        assert locals_ is not frame.f_globals and '__module__' in locals_, \
               'implements() can only be used in a class definition'

        locals_.setdefault('_implements', []).extend(interfaces)

This is the core of Trac's plugin system, and is used in nearly every
class. I can't see anyway to work around this kind of code without
touching every class in Trac. SCons has similar code as well, to
handle its SConscript files.

The following issues deal with this:
http://www.codeplex.com/IronPython/WorkItem/View.aspx?WorkItemId=12876
http://www.codeplex.com/IronPython/WorkItem/View.aspx?WorkItemId=15399
http://www.codeplex.com/IronPython/WorkItem/View.aspx?WorkItemId=15400

I realize that this is a major feature, but I hope that it can get in
sooner rather than later.

- Jeff


From miha.valencic at gmail.com  Wed Dec 31 09:19:38 2008
From: miha.valencic at gmail.com (Miha Valencic)
Date: Wed, 31 Dec 2008 09:19:38 +0100
Subject: [IronPython] Embedding IronPython
In-Reply-To: <495A9B05.8060703@voidspace.org.uk>
References: <20081225210901.F261A1C70480@smtp.webfaction.com>
	<495A9B05.8060703@voidspace.org.uk>
Message-ID: <233dfa1d0812310019r5c4b8905g3d78395519dfbcb@mail.gmail.com>

We're doing something similar with IPY 1.1 and it works so far. Basically,
we use IP for "content templating" and each request executes a script with
certain variables.

I don;t know if it is really thread safe, but we've had no issues.

hth,
  Miha.

2008/12/30 Michael Foord <fuzzyman at voidspace.org.uk>

> Hello Josh,
>
> Sorry for the delayed reply - life is hectic at the moment and I'm still
> copy editing IronPython in Action.
>
> I've copied your email to the IronPython list as I don't know whether the
> CreateScope and other methods of the ScriptEngine are thread safe. My
> instinct would be to say no, but on the other hand the DLR has to handle
> running multi-threaded code so maybe...
>
> Michael Foord
>
> greenbergjosh at hotmail.com wrote:
>
>> Some one has sent you a message via your contact form. They filled in the
>> following values :
>>
>> comment = I read your article on IronPython and I was hoping you might
>> have an answer to the following question.
>>
>> I would like to embed IronPython in a web application.  I would like to
>> initialize the engine and runtime in the Application_Start() method and
>> store an instance of the ScriptEngine in the Application object.  Then, any
>> time I need to run some python I intend to call the
>> ScriptEngine::CreateScope() method and use it to get a new scope in which to
>> execute my code.  Is this approach thread safe?  I will have only one
>> instance of the ScriptEngine shared by many threads.  Each thread will call
>> CreateScope whenever it chooses.
>>
>>
>> It will look something like:
>>
>>
>> protected void Application_Start(object sender, EventArgs e)
>>
>> {
>>
>> ScriptRuntime truntime = ScriptRuntime.Create();
>>
>> ScriptEngine teng = truntime.GetEngine("py");
>>
>> this.Application.Add(???PythonEngine?? , teng);
>>
>> }
>>
>>
>> Then, in some web page or dll called by a web page I would do
>>
>>
>> ..
>>
>> ScriptEngine se = (ScriptEngine)Application[???PythonEngine?? ];
>>
>> ScriptScope ss = se.CreateScope();
>>
>> ScriptSource src = se.CreateScriptSourceFromString(???y=24?? ,
>> SourceCodeKind.Statements);
>>
>> CompiledCode cc = src.Compile();
>>
>> cc.Execute(scope);
>>
>> ...
>>
>>
>> Is this going to work or will I run into threading problems?  In other
>> words, is the ScriptEngine thread-safe?  Will it serialize all of my
>> cc.Execute() calls?  Any other insight into its behavior in this regard
>> would be appreciated.
>>
>>
>> Many Thanks,
>>
>> Josh
>>
>> name = Josh
>> email = greenbergjosh at hotmail.com
>> subject = Embedding IronPython
>>
>>
>
>
> --
> http://www.ironpythoninaction.com/
> http://www.voidspace.org.uk/blog
>
>
>
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081231/63b6d907/attachment.html>

From christian2.schmidt at gmx.de  Wed Dec 31 11:50:16 2008
From: christian2.schmidt at gmx.de (Christian Schmidt)
Date: Wed, 31 Dec 2008 11:50:16 +0100
Subject: [IronPython] Embedding IronPython
In-Reply-To: <495A9B05.8060703@voidspace.org.uk>
References: <20081225210901.F261A1C70480@smtp.webfaction.com>
	<495A9B05.8060703@voidspace.org.uk>
Message-ID: <495B4E68.8050606@gmx.de>

Hi Josh,

>> CompiledCode cc = src.Compile();

This will leak memory. So you will run out of memory after a while (in 
our case after 100'000 compiles on x86).
[http://www.codeplex.com/IronPython/WorkItem/View.aspx?WorkItemId=20399]

>> Is this going to work or will I run into threading problems?  In other 
>> words, is the ScriptEngine thread-safe?  Will it serialize all of my 
>> cc.Execute() calls?  Any other insight into its behavior in this 
>> regard would be appreciated.

I think globals and imports will be shared. So modifying them might give 
unexpected results.

Cheers,
Christian


From empirebuilder at gmail.com  Wed Dec 31 13:35:24 2008
From: empirebuilder at gmail.com (Dody Gunawinata)
Date: Wed, 31 Dec 2008 13:35:24 +0100
Subject: [IronPython] Embedding IronPython
In-Reply-To: <495B4E68.8050606@gmx.de>
References: <20081225210901.F261A1C70480@smtp.webfaction.com>
	<495A9B05.8060703@voidspace.org.uk> <495B4E68.8050606@gmx.de>
Message-ID: <8cd017b80812310435t719d1a90p9023a57df886ebe0@mail.gmail.com>

Please we need more vote on the memory leak issue because it has huge impact
on the usage of Iron Python for the web environment as well.

On Wed, Dec 31, 2008 at 11:50 AM, Christian Schmidt <
christian2.schmidt at gmx.de> wrote:

> Hi Josh,
>
>  CompiledCode cc = src.Compile();
>>>
>>
> This will leak memory. So you will run out of memory after a while (in our
> case after 100'000 compiles on x86).
> [http://www.codeplex.com/IronPython/WorkItem/View.aspx?WorkItemId=20399]
>
>  Is this going to work or will I run into threading problems?  In other
>>> words, is the ScriptEngine thread-safe?  Will it serialize all of my
>>> cc.Execute() calls?  Any other insight into its behavior in this regard
>>> would be appreciated.
>>>
>>
> I think globals and imports will be shared. So modifying them might give
> unexpected results.
>
> Cheers,
> Christian
>
> _______________________________________________
> Users mailing list
> Users at lists.ironpython.com
> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com
>



-- 
nomadlife.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20081231/892d0420/attachment.html>

From sanxiyn at gmail.com  Wed Dec 31 17:07:44 2008
From: sanxiyn at gmail.com (Seo Sanghyeon)
Date: Thu, 1 Jan 2009 01:07:44 +0900
Subject: [IronPython] Frames in IronPython
In-Reply-To: <ea89d0e50812302100o7b2f4ba3l2383e9396438ff52@mail.gmail.com>
References: <ea89d0e50812302100o7b2f4ba3l2383e9396438ff52@mail.gmail.com>
Message-ID: <5b0248170812310807s3afab7d1sf8178222060bf52e@mail.gmail.com>

2008/12/31 Jeff Hardy <jdhardy at gmail.com>:
> I don't know what the feature list is going to be for 2.1, but I'd
> like to see frame objects near the top. Almost every piece of Python
> code I've tried uses them in some way. Often it's for debugging
> purposes (and can be worked around), but not always. Trac has this
> lovely piece of code:

This is pretty sad state of affair, especially considering that such
(like Trac's) use of getting frame local is usually unnecessary with
a bit of metaclass.

-- 
Seo Sanghyeon


From curt at hagenlocher.org  Wed Dec 31 17:22:40 2008
From: curt at hagenlocher.org (Curt Hagenlocher)
Date: Wed, 31 Dec 2008 08:22:40 -0800
Subject: [IronPython] Frames in IronPython
In-Reply-To: <ea89d0e50812302100o7b2f4ba3l2383e9396438ff52@mail.gmail.com>
References: <ea89d0e50812302100o7b2f4ba3l2383e9396438ff52@mail.gmail.com>
Message-ID: <d2155e360812310822r6315f462p39e916c59220518@mail.gmail.com>

On Tue, Dec 30, 2008 at 9:00 PM, Jeff Hardy <jdhardy at gmail.com> wrote:
>
> I don't know what the feature list is going to be for 2.1, but I'd
> like to see frame objects near the top.

As I understand it, the biggest problem with getframe has always been
its performance implications.  Would you want all IronPython code to
be slowed down so that this feature is supported?  What if you needed
to opt-in on the command-line with a -X:getframe?

Also, I don't see a lot of votes here. *wink*

--
Curt Hagenlocher
curt at hagenlocher.org


From dan.eloff at gmail.com  Wed Dec 31 22:51:37 2008
From: dan.eloff at gmail.com (Dan Eloff)
Date: Wed, 31 Dec 2008 16:51:37 -0500
Subject: [IronPython] Frames in IronPython
In-Reply-To: <d2155e360812310822r6315f462p39e916c59220518@mail.gmail.com>
References: <ea89d0e50812302100o7b2f4ba3l2383e9396438ff52@mail.gmail.com>
	<d2155e360812310822r6315f462p39e916c59220518@mail.gmail.com>
Message-ID: <4817b6fc0812311351q2e6be652y6114a1c7220e6b5c@mail.gmail.com>

On Wed, Dec 31, 2008 at 11:22 AM, Curt Hagenlocher <curt at hagenlocher.org> wrote:
> On Tue, Dec 30, 2008 at 9:00 PM, Jeff Hardy <jdhardy at gmail.com> wrote:
>>
>> I don't know what the feature list is going to be for 2.1, but I'd
>> like to see frame objects near the top.
>
> As I understand it, the biggest problem with getframe has always been
> its performance implications.  Would you want all IronPython code to
> be slowed down so that this feature is supported?  What if you needed
> to opt-in on the command-line with a -X:getframe?
>
> Also, I don't see a lot of votes here. *wink*
>

I would want it to be opt-in. If I really need it (e.g. Trac) or if I
want it only for debugging, I wouldn't mind the performance hit. If I
don't need it, I wouldn't want it enabled.

Just my thoughts,
-Dan