From keating at acm.org  Mon Jan  4 01:11:09 2010
From: keating at acm.org (Paul Keating)
Date: Mon, 04 Jan 2010 01:11:09 +0100
Subject: [python-win32] Question concerning wmi c.CIM_DataFile
References: <2E8EE617FF5865459F6EF9217BC466E2063E3C@opr-pr-exc-01.rgfc.main>
Message-ID: <014a01ca8cd2$6d32ff80$4801a8c0@vuurpijlhuisje.local>

Don't do this:

#File to query       
string = "C:\DOCUME~1\profileName\LOCALS~1\Temp\\Temporary Directory 1 for deluxetreee (2).zip\Christmas.exe"
 
#Substitute "\" to "\\"
path = re.sub('\\\\', '\\\\\\\\', string)

I can see nothing wrong with this particular bit of code, but sooner or later doing this sort of thing will trip you up. Instead use a raw string (r"mys\t\ri\ng") that will leave your whacks intact. All you need is this:

path = r"C:\DOCUME~1\profileName\LOCALS~1\Temp\\Temporary Directory 1 for deluxetreee (2).zip\Christmas.exe"

I notice you have a doubled \ in path after Temp which may be the problem, depending on the target filesystem.

--
Paul Keating
The Hague 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-win32/attachments/20100104/6e4735ce/attachment.htm>

From Sunny.Carter at metaswitch.com  Mon Jan  4 17:51:42 2010
From: Sunny.Carter at metaswitch.com (Sunny Carter)
Date: Mon, 4 Jan 2010 16:51:42 +0000
Subject: [python-win32] Boolean type changed in Python 3.0?
Message-ID: <44EE1E31095AE349AC6C3C0B69FFBF02932E0B3D6F@ENFIMBOX1.ad.datcon.co.uk>

Hi all,

I'm trying my posting again with a different subject so that it is more generic.

I am having problems calling across the win32 API using a Boolean argument in Python 3.0 (False in my python script) which is not recognised as a Boolean (The error I get back from setattr is 'Boolean value expected').

Has the way that a Boolean is represented changed in Python 3.0?  This worked fine in Python 2.6.  I cannot get it to work and have tried values of False,True,0 and 1, all to no avail.

It sounds like a bug in the win32com API for 3.0 to me.

Further details below.

Many thanks to anyone that can help,

Sunny Carter
==============================================================================================================

I am writing a Python script to call across to Adobe Photoshop, to automate some image manipulation.  (For historical reasons I'm writing using Python rather than javascript or VB)

My problem is that my script works fine using Python 2.6, but fails after upgrading to Python 3.0 or 3.1.1.  I've converted the script to Python 3.0 format, and it all works except the call across the Photoshop API.

The problem seems to be with the win32 API call using a Boolean value (False in my python script) which is not recognised as a Boolean by Photoshop.

Details:
Script based in part on http://techarttiki.blogspot.com/2008/08/photoshop-scripting-with-python.html
Script uses pywin32 from http://sourceforge.net/projects/pywin32/files/
    - I have installed pywin32-214.win32-py2.6.exe for Python 2.6 and pywin32-214.win32-py3.0.exe for use with Python 3.0

Subset of the script:

    import win32com.client
    psApp = win32com.client.Dispatch('Photoshop.Application')
    doc = psApp.Open(os.path.abspath(psd_image.psd_filename))
    #  Here I do the image manipulation which I haven't included here as it is irrelevant (and working fine -
    # calling Photoshop and making it do stuff)
    psd_save_options = win32com.client.Dispatch("Photoshop.PNGSaveOptions")
    psd_save_options.Interlaced = False
    print("Save as %s" % (os.path.join(output_directory, psd_image.final_filename)))
    doc.SaveAs(os.path.join(output_directory, psd_image.final_filename),
               psd_save_options,
               True,
               2)  # lower case extension

Exception hit:

Unexpected error detail: <class 'pywintypes.com_error'>
Traceback (most recent call last):
  File "C:\brander\brander.py", line 6464, in <module> main()
  File "C:\brander\brander.py", line 6407, in main recf)
  File "C:\brander\brander.py", line 4303, in createBranding brand_widgets)
  File "C:\brander\brander.py", line 1961, in reconstructImageDirectory    brand_widgets)
  File "C:\brander\brander.py", line 1961, in reconstructImageDirectory     brand_widgets)
  File "C:\brander\brander.py", line 1961, in reconstructImageDirectory     brand_widgets)
  File "C:\brander\brander.py", line 1999, in reconstructImageDirectory     new_color_palette)
  File "C:\brander\brander.py", line 2519, in reconstructPSDImageFile   psd_save_options.Interlaced = False
  File "C:\Program Files\Python31\lib\site-packages\win32com\client\dynamic.py", line 550, in __setattr__     self._oleobj_.Invoke(entry.dispid, 0, invoke_type, 0, value)
pywintypes.com_error: (-2147352567, 'Exception occurred.', (0, 'Adobe Photoshop' , 'Boolean value expected', None, 0, -2147220276), None)

Investigation tracing from pywintypes.py
The dll being loaded is
C:\WINDOWS\system32\pywintypes30.dll

Things tried so far
I've already tried setting psd_save_options.Interlaced to True, "False", 0, 1 and None.  In this, None gets further on in the script but then fails on the SaveAs call.  If I delete the psd_save_options.Interlaced line, then the script fails in exactly the same way on the SaveAs call (this time the parameter True is the culprit).

As I say - the same script works fine using Python 2.6 and pywin32-214.win32-py2.6.exe.  I am running on Windows XP.

Help?
If anyone can help with a solution that would be great.  If there is no way of fixing, is it possible to run my python script using Python 2.6 but call out to a separate script running Python 3.0 for the part where I require 3.0?  (not included here)

Thanks so much,

Sunny Carter
<http://www.metaswitch.com/>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-win32/attachments/20100104/8425b1cd/attachment.htm>

From wade.dyck at gmail.com  Mon Jan  4 18:06:20 2010
From: wade.dyck at gmail.com (Wade Dyck)
Date: Mon, 4 Jan 2010 10:06:20 -0700
Subject: [python-win32] Exception: object has no attribute 'Win32_Process'
Message-ID: <a0dfa4f11001040906m41009b57o214c58272768dedd@mail.gmail.com>

I create a wmi connection to a server in a separate domain from where the
script is running like this:

    remote = wmi.connect_server(server=server, user=r"%s\%s" % (domain,
user),
        password=password, impersonation_level="impersonate")
    self.c = wmi.WMI(wmi=remote)

I then successfully run a few Win32_Process.create() commands (which fail if
the initial connection above uses impersonation_level="delegate"). Next I
use adsi to trust the server for delegation:

    computer.Put('userAccountControl', value)
    computer.Setinfo()

At this point I want to use Win32_Process.create() to xcopy some files from
a third server, so I switch to delegation since impersonate won't allow
access to network resources:

    remote = wmi.connect_server(server=self.server, user=r"%s\%s"
            % (self.domain, self.user), password=self.password,
            impersonation_level="delegate")
    self.c = wmi.WMI(wmi=remote)

But then when I try to create the process using:

    pid, res = self.c.Win32_Process.Create('cmd.exe /c xcopy %s %s \
                    /s /e /i /y /c /q' % (src, dest))

I get the following exception:

AttributeError: '<win32com.gen_py.Microsoft WMI Scripting V1.2
Library.ISWbemSer
vices instance at 0x12790040>' object has no attribute 'Win32_Process'

However, this works fine if I don't set the userAccountControl attribute
above using adsi. It also seems specific to actually setting the attribute,
as the code is always using adsi to get DNs and the current value of
userAccountControl.

Am I doing something wrong, or does anyone have suggestions on what to try
next?
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-win32/attachments/20100104/887f99ae/attachment.htm>

From timr at probo.com  Mon Jan  4 19:57:45 2010
From: timr at probo.com (Tim Roberts)
Date: Mon, 04 Jan 2010 10:57:45 -0800
Subject: [python-win32] Question concerning wmi c.CIM_DataFile
In-Reply-To: <2E8EE617FF5865459F6EF9217BC466E2063E3C@opr-pr-exc-01.rgfc.main>
References: <2E8EE617FF5865459F6EF9217BC466E2063E3C@opr-pr-exc-01.rgfc.main>
Message-ID: <4B423A29.9050305@probo.com>

Kelvin Lomboy Mendez wrote:
>
> I?m currently writing a script to access approximately 400 computers
> for removing a particular app from the system. I?m having problems
> getting results using the wmi c.CIM_DataFile (name=file). When I use
> file as a variable which equals to the path to the app I want to
> remove I don?t get nothing back, however, if I hard code the path, I
> get results. See snippet below for an example of what I?m talking
> about. I appreciate in advance for any help I can get. Thank you.
>
>  
>
> import wmi
>
> import re
>
>  
>
> def queryFile(file, host):
>
>     print '[Debug nukeXmas()]:' + file
>
>     c = wmi.WMI(host)
>
>     for f in c.CIM_DataFile (name=file):
>
>         print 'Install Date:', f.InstallDate
>
>        
>
> def queryFile2(file, host):
>
>     print '[Debug nukeXmas()]:' + file
>
>     c = wmi.WMI(host)
>
>     for f in c.CIM_DataFile
> (name="C:\\DOCUME~1\\profileName\\LOCALS~1\\Temp\\Temporary Directory
> 1 for deluxetreee (2).zip\\Christmas.exe"):
>
>         print 'Install Date:', f.InstallDate
>
>        
>
>  #File to query      
>
> string = "C:\DOCUME~1\profileName\LOCALS~1\Temp\\Temporary Directory 1
> for deluxetreee (2).zip\Christmas.exe"
>
>  
>
> #Substitute "\" to "\\"
>
> path = re.sub('\\\\', '\\\\\\\\', string)
>

Don't do this.  You WANT the string to contain single backslashes.  This
is one of the most confusing things about working with strings in Python
on Windows.  Here's a test.  How many characters are in this string?

    xyz = "a\\b\tc"

The answer is 5.  There's a letter "a", a backslash, a letter "b", a
tab, and a letter "c".

When you pass a file name into an API, the name should only contain
single backslashes (or forward slashes -- both work equally well in the
Win32 API).  In order to GET single backslashes in a Python string
literal, you need to type TWO (or use the raw r"string" concept), but
the string itself only contains one.

> #Close query path with double quotes
>
> file = '"'+path+'"'
>

Don't do this either.  The file name does not contain quotes.  If you
are typing the file name on a command-line, you have to provide quotes
so they get handled by the command line parser properly, but when you're
calling an API, you should NEVER use quotes.

> #I'm having problems here, query brings back nothing
>
> queryFile(file, '172.27.1.5')
>

Right, because the file name you are passing here is literally this:
    "C:\\DOCUME~1\\profileName\\LOCALS~1\\Temp\\Temporary Directory 1
for deluxetreee (2).zip\\Christmas.exe"

>  
>
> #Here, I hard code the path to the exe and it works
>
> queryFile2(file, '172.27.1.5')
>

Right, because the file name you are passing here is literally this:
    C:\DOCUME~1\profileName\LOCALS~1\Temp\Temporary Directory 1 for
deluxetreee (2).zip\Christmas.exe

And THAT is what the file name really is.

By the way, your whole concept is loony.  This will not work at all on
Vista or Win 7, because the "Documents and Settings" tree is now called
"Users".  You are assuming that every system is going to have this file
as "...tree (2).zip", but depending on how many times it has been
downloaded, it might be called something different (like "...tree.zip"
or "...tree(3).zip").  Further, the directory names you are using are
configurable.  They don't have to be called "Documents and Setting" or
"Local Settings", and they don't have to be located on "C:".  You should
be using the shell folder APIs to find the names of the "well-known
folders".

Plus, the whole "Local Settings\Temp" folder is volatile.  You should
just be able to wipe out that whole directory without causing any damage.

-- 
Tim Roberts, timr at probo.com
Providenza & Boekelheide, Inc.


From mdriscoll at co.marshall.ia.us  Mon Jan  4 22:18:43 2010
From: mdriscoll at co.marshall.ia.us (Mike Driscoll)
Date: Mon, 04 Jan 2010 15:18:43 -0600
Subject: [python-win32] Question concerning wmi c.CIM_DataFile
In-Reply-To: <4B423A29.9050305@probo.com>
References: <2E8EE617FF5865459F6EF9217BC466E2063E3C@opr-pr-exc-01.rgfc.main>
	<4B423A29.9050305@probo.com>
Message-ID: <4B425B33.2060003@co.marshall.ia.us>

On 1:59 PM, Tim Roberts wrote:
> Kelvin Lomboy Mendez wrote:
>    
>> I'm currently writing a script to access approximately 400 computers
>> for removing a particular app from the system. I'm having problems
>> getting results using the wmi c.CIM_DataFile (name=file). When I use
>> file as a variable which equals to the path to the app I want to
>> remove I don't get nothing back, however, if I hard code the path, I
>> get results. See snippet below for an example of what I'm talking
>> about. I appreciate in advance for any help I can get. Thank you.
>>
>>
>>
>> import wmi
>>
>> import re
>>
>>
>>
>> def queryFile(file, host):
>>
>>      print '[Debug nukeXmas()]:' + file
>>
>>      c = wmi.WMI(host)
>>
>>      for f in c.CIM_DataFile (name=file):
>>
>>          print 'Install Date:', f.InstallDate
>>
>>
>>
>> def queryFile2(file, host):
>>
>>      print '[Debug nukeXmas()]:' + file
>>
>>      c = wmi.WMI(host)
>>
>>      for f in c.CIM_DataFile
>> (name="C:\\DOCUME~1\\profileName\\LOCALS~1\\Temp\\Temporary Directory
>> 1 for deluxetreee (2).zip\\Christmas.exe"):
>>
>>          print 'Install Date:', f.InstallDate
>>
>>
>>
>>   #File to query
>>
>> string = "C:\DOCUME~1\profileName\LOCALS~1\Temp\\Temporary Directory 1
>> for deluxetreee (2).zip\Christmas.exe"
>>
>>
>>
>> #Substitute "\" to "\\"
>>
>> path = re.sub('\\\\', '\\\\\\\\', string)
>>
>>      
> Don't do this.  You WANT the string to contain single backslashes.  This
> is one of the most confusing things about working with strings in Python
> on Windows.  Here's a test.  How many characters are in this string?
>
>      xyz = "a\\b\tc"
>
> The answer is 5.  There's a letter "a", a backslash, a letter "b", a
> tab, and a letter "c".
>
> When you pass a file name into an API, the name should only contain
> single backslashes (or forward slashes -- both work equally well in the
> Win32 API).  In order to GET single backslashes in a Python string
> literal, you need to type TWO (or use the raw r"string" concept), but
> the string itself only contains one.
>
>    
>> #Close query path with double quotes
>>
>> file = '"'+path+'"'
>>
>>      
> Don't do this either.  The file name does not contain quotes.  If you
> are typing the file name on a command-line, you have to provide quotes
> so they get handled by the command line parser properly, but when you're
> calling an API, you should NEVER use quotes.
>
>    
>> #I'm having problems here, query brings back nothing
>>
>> queryFile(file, '172.27.1.5')
>>
>>      
> Right, because the file name you are passing here is literally this:
>      "C:\\DOCUME~1\\profileName\\LOCALS~1\\Temp\\Temporary Directory 1
> for deluxetreee (2).zip\\Christmas.exe"
>
>    
>>
>>
>> #Here, I hard code the path to the exe and it works
>>
>> queryFile2(file, '172.27.1.5')
>>
>>      
> Right, because the file name you are passing here is literally this:
>      C:\DOCUME~1\profileName\LOCALS~1\Temp\Temporary Directory 1 for
> deluxetreee (2).zip\Christmas.exe
>
> And THAT is what the file name really is.
>
> By the way, your whole concept is loony.  This will not work at all on
> Vista or Win 7, because the "Documents and Settings" tree is now called
> "Users".  You are assuming that every system is going to have this file
> as "...tree (2).zip", but depending on how many times it has been
> downloaded, it might be called something different (like "...tree.zip"
> or "...tree(3).zip").  Further, the directory names you are using are
> configurable.  They don't have to be called "Documents and Setting" or
> "Local Settings", and they don't have to be located on "C:".  You should
> be using the shell folder APIs to find the names of the "well-known
> folders".
>
> Plus, the whole "Local Settings\Temp" folder is volatile.  You should
> just be able to wipe out that whole directory without causing any damage.
>
>    

The OP should check out Tim Golden's winshell module as it gives access 
to the common folders quite easily:

http://timgolden.me.uk/python/winshell.html



-- 
*Mike Driscoll*

Blog: http://blog.pythonlibrary.org

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-win32/attachments/20100104/c0833297/attachment.htm>

From python at bdurham.com  Tue Jan  5 00:43:24 2010
From: python at bdurham.com (python at bdurham.com)
Date: Mon, 04 Jan 2010 18:43:24 -0500
Subject: [python-win32] Create a QueryInfo type Windows Shell Extension to
 provide InfoTip data in Windows Explorer
Message-ID: <1262648604.4336.1352935453@webmail.messagingengine.com>

Wondering if its possible to create Windows Shell Extensions
using pywin32 and if so, has anyone has tried to create a
QueryInfo type Windows Shell Extension?

I would like to create a Python equivalent of the simple InfoTip
described in this well written tutorial.
http://www.codeproject.com/KB/shell/shellextguide3.aspx

Here's a quick excerpt from this article:

Quote:

The Active Desktop shell introduced a new feature: tooltips that
show a description of certain objects if you hover the mouse over
them. For example, hovering over My Documents shows this tooltip:

[My Computer tooltip - 6K]

Other objects like My Computer and Control Panel have similar
tooltips. These bits of text are called infotips, since they are
tooltips that provide information about the file, folder, or
object that the mouse is over. We can provide our own infotip
text for other objects in the shell, using an infotip extension.
An example of an infotip extension that you have probably already
seen is in WinZip, which shows the contents of compressed files:

[WinZip tooltip - 5K]

This article's sample extension will be a quick text file viewer
- it will display the first line of the file, along with the
total file size. This information will appear in an infotip when
the user hovers the mouse over a .TXT file.
Extracting the info I want to display in an InfoTIp is easy :)
... its the plumbing to support the display of information that
has me stuck.

Thank you,
Malcolm
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-win32/attachments/20100104/bee0d1b6/attachment.htm>

From timr at probo.com  Tue Jan  5 02:15:03 2010
From: timr at probo.com (Tim Roberts)
Date: Mon, 04 Jan 2010 17:15:03 -0800
Subject: [python-win32] Create a QueryInfo type Windows Shell Extension
 to provide InfoTip data in Windows Explorer
In-Reply-To: <1262648604.4336.1352935453@webmail.messagingengine.com>
References: <1262648604.4336.1352935453@webmail.messagingengine.com>
Message-ID: <4B429297.4000400@probo.com>

python at bdurham.com wrote:
> Wondering if its possible to create Windows Shell Extensions using
> pywin32 and if so, has anyone has tried to create a QueryInfo type
> Windows Shell Extension?

Sure, it's possible.  A shell extension is just a COM server.  There are
lots of example showing how to build a COM server in Python.  You'll
need to support IPersistFile and IQueryInfo.

Here's a web page from Tim Golden showing a sample shell extension in
Python.  You'll need to modify this to support IPersistFile and
IQueryInfo based on the information in that article, but this will show
you the mechanics of it:
   
http://timgolden.me.uk/python/win32_how_do_i/add-my-own-icon-overlays.html

-- 
Tim Roberts, timr at probo.com
Providenza & Boekelheide, Inc.


From keesvanschaik at gmail.com  Tue Jan  5 03:28:47 2010
From: keesvanschaik at gmail.com (Kees van Schaik)
Date: Tue, 05 Jan 2010 02:28:47 +0000
Subject: [python-win32] win32com calling IE: OLE error 0x800704a6
Message-ID: <4B42A3DF.20003@gmail.com>

(This is a cross post from the Python mailing list where unfortunately 
nobody responded. I would be very grateful if someone could give me some 
insights!)

Dear all,

I have written a Python script that runs on Windows XP with Python 2.5
and uses Internet Explorer to 'pretty print' some HTML output. This
has been working perfectly fine for a few years until recently IE was
upgraded to version 8. Now occasionally Python throws an error that
originates from the win32com.client.Dispatch() method and has
description 'OLE error 0x800704a6'. (Occasionally means that sometimes
it happens, sometimes it doesn't. I haven't been able to figure out
what the trigger exactly is.) The most recent version of win32com is
installed. Google on the error code didn't give me a lot of hits, and
no useful ones.

So, I was hoping somebody might be able to give me some pointers what
might be going on.

Here is the relevant (very straightforward) Python code:

import win32com.client

class IE:
    def __init__(self,url=None,vis=0,standalone=False):
        #Pas registerwaardes aan voor het printen:
        if not standalone:
            self.regOld=self.getRegValues()
            self.setRegValues
({'margin_left':'0.00000','margin_right':'0.00000','footer':'','header':''}) 

        #Start instantie van IE op:
        self.ie=win32com.client.Dispatch
('InternetExplorer.Application')
        self.ie.Visible = vis
        if url!=None:
            self.ie.Navigate(url)

and the class furthermore implements some methods to open & print some
HTML-file.

Unfortunately I only have a screenshot of the error:
http://tweakers.net/ext/f/cAdKjyjs7w4yZRu6s4RqNwjW/full.png 
<http://www.google.com/url?sa=D&q=http://tweakers.net/ext/f/cAdKjyjs7w4yZRu6s4RqNwjW/full.png&usg=AFQjCNEHr-QJwwMhc9AyTj-w46fL20XX9Q> 


Thanks in advance for your thoughts, Kees


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-win32/attachments/20100105/36e2fa34/attachment-0001.htm>

From skippy.hammond at gmail.com  Tue Jan  5 08:58:21 2010
From: skippy.hammond at gmail.com (Mark Hammond)
Date: Tue, 05 Jan 2010 18:58:21 +1100
Subject: [python-win32] Boolean type changed in Python 3.0?
In-Reply-To: <44EE1E31095AE349AC6C3C0B69FFBF02932E0B3D6F@ENFIMBOX1.ad.datcon.co.uk>
References: <44EE1E31095AE349AC6C3C0B69FFBF02932E0B3D6F@ENFIMBOX1.ad.datcon.co.uk>
Message-ID: <4B42F11D.3010109@gmail.com>

On 5/01/2010 3:51 AM, Sunny Carter wrote:
> Hi all,
> I'm trying my posting again with a different subject so that it is more
> generic.
> I am having problems calling across the win32 API using a Boolean
> argument in Python 3.0 (False in my python script) which is not
> recognised as a Boolean (The error I get back from setattr is 'Boolean
> value expected').
> Has the way that a Boolean is represented changed in Python 3.0? This
> worked fine in Python 2.6. I cannot get it to work and have tried values
> of False,True,0 and 1, all to no avail.
> It sounds like a bug in the win32com API for 3.0 to me.
> Further details below.
> Many thanks to anyone that can help,

I expect something subtle is going on with boolean conversions in 3.x, 
but the test suite does have coverage of booleans, so I'm really not 
sure what the problem could be.  I don't have photoshop so I'm unable to 
reproduce - I intend revisiting the tests to see if I can spot an edge 
case which isn't tested, but I'm yet to get to that after the holidays...

Cheers,

Mark

From timr at probo.com  Tue Jan  5 19:09:19 2010
From: timr at probo.com (Tim Roberts)
Date: Tue, 05 Jan 2010 10:09:19 -0800
Subject: [python-win32] win32com calling IE: OLE error 0x800704a6
In-Reply-To: <4B42A3DF.20003@gmail.com>
References: <4B42A3DF.20003@gmail.com>
Message-ID: <4B43804F.1010501@probo.com>

Kees van Schaik wrote:
>
> I have written a Python script that runs on Windows XP with Python 2.5
> and uses Internet Explorer to 'pretty print' some HTML output. This
> has been working perfectly fine for a few years until recently IE was
> upgraded to version 8. Now occasionally Python throws an error that
> originates from the win32com.client.Dispatch() method and has
> description 'OLE error 0x800704a6'. (Occasionally means that sometimes
> it happens, sometimes it doesn't. I haven't been able to figure out
> what the trigger exactly is.) The most recent version of win32com is
> installed. Google on the error code didn't give me a lot of hits, and
> no useful ones.
>

The 80070xxx errors are often COM-wrapped versions of the standard
Windows errors.  0x4A6 is 1190, which is "ERROR_SHUTDOWN_IS_SCHEDULED". 
Is it possible this happens after a Windows Update has occurred but
before the reboot?

-- 
Tim Roberts, timr at probo.com
Providenza & Boekelheide, Inc.


From keesvanschaik at gmail.com  Tue Jan  5 20:28:37 2010
From: keesvanschaik at gmail.com (Kees van Schaik)
Date: Tue, 5 Jan 2010 19:28:37 +0000
Subject: [python-win32] win32com calling IE: OLE error 0x800704a6
Message-ID: <5f2152f71001051128r4661f997m8e84aa61aad610d1@mail.gmail.com>

>The 80070xxx errors are often COM-wrapped versions of the standard
>Windows errors.  0x4A6 is 1190, which is "ERROR_SHUTDOWN_IS_SCHEDULED".
>Is it possible this happens after a Windows Update has occurred but
>before the reboot?

Thanks. It is not impossible, but unlikely given that it has occured
so frequently over the last period and the pc gets rebooted each
morning. I mean, I don't assume updates that require restarting occur
so frequently, do they?
Then again, the pc is used by somebody who might be missing such things.

Although this is not the right thread, I'm looking at alternatives in
case I can't get it resolved. The most straightforward way seems to be
to print directly using a win32ui.CreateDC() object. There is only one
problem left with that approach, namely

dc.TextOut(0, 0, "Text")

works perfect to place plain text on the page, but I'd also need to
print a limited number of non-ASCII characters (accents in names,
eurosign) and feeding the above method a unicode-string seems to throw
an error. Is it possible to get
non-ASCII characters printed this way?
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-win32/attachments/20100105/3a5ceb0a/attachment.htm>

From dreadpiratejeff at gmail.com  Tue Jan  5 21:42:44 2010
From: dreadpiratejeff at gmail.com (J)
Date: Tue, 5 Jan 2010 15:42:44 -0500
Subject: [python-win32] Still looking for a method to get CPUID info ...
Message-ID: <36dec4ff1001051242q2ae88c0cx332ced7065b12121@mail.gmail.com>

I've been at this a while :)

Thought I'd ask if anyone knew either of a way in python, or just a
stand alone tool, to use in Windows to get CPU information.
Specifically, I'm looking for Feature sets or Flags.

What I'm doing is building a tool for work that gets certain bits of
system information to automate test configuration and deployment.

So far, I'm using a combination of dmidecode.exe and the Python WMI
module to get info from WMI and the BIOS.  The problem I'm still stuck
with is actually getting the feature sets of a given processor.

dmidecode returns a nice list of flags, however, it only seems to
support and report the original register (I think it's ECX) and thus,
only shows 31 flags.  The new feature flags are being dumped, from
what I've beed reading, into a different register (I think this one is
EDX) and dmidecode (and dmi) can not access that.  DMI is dead, in any
case, and CIM is the new standard....

The problem is, after pouring through the CIM/WMI stuff over and over,
I've found no part of CIM or WMI that reports CPU feature sets.

In linux, this is pretty easy... I can just parse /proc/cpuinfo to get
every flag, and the new ones show up (assuming I'm running a new
enough kernel to handle the new flags).  In my case, I'll always be
running the newest GA linux distros.  BUT, I'll also be using this
code on Windows systems running Server 2008 and beyond.  So, the
problem now is not so much as how do I get the flags, because I can at
least get the basic ones, but how to I get ALL the flags.

What I'm specifically interested in are flags like VMX and SVN for
virtualization extensions, and maybe a couple others that have come
with the new generation AMD and Intel processors.

Any ideas, outside of what I've already tried (WMI/CIM and dmidecode?)
 I'm license limited, so any tools I try to include and use have to be
GPL.

Cheers
Jeff, who has a headache right now because of all this.

-- 

Samuel Goldwyn  - "I'm willing to admit that I may not always be
right, but I am never wrong." -
http://www.brainyquote.com/quotes/authors/s/samuel_goldwyn.html

From timr at probo.com  Tue Jan  5 22:53:59 2010
From: timr at probo.com (Tim Roberts)
Date: Tue, 05 Jan 2010 13:53:59 -0800
Subject: [python-win32] Still looking for a method to get CPUID info ...
In-Reply-To: <36dec4ff1001051242q2ae88c0cx332ced7065b12121@mail.gmail.com>
References: <36dec4ff1001051242q2ae88c0cx332ced7065b12121@mail.gmail.com>
Message-ID: <4B43B4F7.8070803@probo.com>

J wrote:
> I've been at this a while :)
>
> Thought I'd ask if anyone knew either of a way in python, or just a
> stand alone tool, to use in Windows to get CPU information.
> Specifically, I'm looking for Feature sets or Flags.
>
> What I'm doing is building a tool for work that gets certain bits of
> system information to automate test configuration and deployment.
>   

In my opinion, you could have solved this all much quicker and easier
just by writing a trivial DLL.

c:\tmp>type cpuid.asm
    .586
    .model   flat, stdcall
    .code

GetCpuidEcx   proc   public
    xor eax, eax
    cpuid
    mov eax, ecx
    ret
GetCpuidEcx   endp

GetCpuidEdx   proc   public
    xor eax, eax
    cpuid
    mov eax, edx
    ret
GetCpuidEdx   endp

start   proc    public, x:DWORD, y:DWORD, z:DWORD
        ret
start   endp               

    end start

C:\tmp>ml /Fl /c cpuid.asm
Microsoft (R) Macro Assembler Version 9.00.21022.08
Copyright (C) Microsoft Corporation.  All rights reserved.

 Assembling: cpuid.asm

C:\tmp>link /dll /out:cpuid.dll /export:GetCpuidEcx /export:GetCpuidEdx
/entry:start cpuid.obj
Microsoft (R) Incremental Linker Version 9.00.21022.08
Copyright (C) Microsoft Corporation.  All rights reserved.

   Creating library cpuid.lib and object cpuid.exp

C:\tmp>python
Python 2.6.2 (r262:71605, Apr 14 2009, 22:40:02) [MSC v.1500 32 bit
(Intel)] onwin32
Type "help", "copyright", "credits" or "license" for more information.
 >>> import ctypes
 >>> c = ctypes.windll.cpuid
 >>> hex(c.GetCpuidEcx())
'0x444d4163'
 >>> hex(c.GetCpuidEdx())
'0x69746e65'
 >>>

C:\tmp>

Modifying this to accept the "cpuid" eax value as a parameter is left as
an exercise for the reader.

-- 
Tim Roberts, timr at probo.com
Providenza & Boekelheide, Inc.


From dreadpiratejeff at gmail.com  Tue Jan  5 23:03:05 2010
From: dreadpiratejeff at gmail.com (J)
Date: Tue, 5 Jan 2010 17:03:05 -0500
Subject: [python-win32] Still looking for a method to get CPUID info ...
In-Reply-To: <4B43B4F7.8070803@probo.com>
References: <36dec4ff1001051242q2ae88c0cx332ced7065b12121@mail.gmail.com> 
	<4B43B4F7.8070803@probo.com>
Message-ID: <36dec4ff1001051403s28b1590fib76238599d365a61@mail.gmail.com>

On Tue, Jan 5, 2010 at 16:53, Tim Roberts <timr at probo.com> wrote:
> In my opinion, you could have solved this all much quicker and easier
> just by writing a trivial DLL.

Indeed... and I may well have tried that, were I actually capable of
reading and writing assembly and/or low-level C...  Unfortunately, I'm
not capable of such things and trivial for you is Magic Blue Smoke to
me...  I do understand the basics about registers and kinda how things
are moved in and out, but when it comes to actual real assembly, or
anything more low-level than userspace, I don't know nothin' useful.
Just enough to be dangerous.

However, that's not meant to be snarky, I'm just honest about my
abilities/knowledge.

BUT, thanks for this... this seems simple enough that I can dissect it
and play with it and actually learn some basic assembly (which I've
never done before).

> C:\tmp>python
> Python 2.6.2 (r262:71605, Apr 14 2009, 22:40:02) [MSC v.1500 32 bit
> (Intel)] onwin32
> Type "help", "copyright", "credits" or "license" for more information.
> ?>>> import ctypes
> ?>>> c = ctypes.windll.cpuid
> ?>>> hex(c.GetCpuidEcx())
> '0x444d4163'
> ?>>> hex(c.GetCpuidEdx())
> '0x69746e65'

Also have to figure out how to decode the hex value into a more user
friendly format ;-)  BUT I think there was a white paper or something
that I have bookmarked somewhere that does have hex values for the
various flags, at least according to the Intel specs, assuming that
vendors (in this case IBM) hasn't mucked with them.

Seriously though, thanks, even if it turns out I can't really use this
for what I'm working on, it certainly does give me something new to
play with and learn about.

Cheers
Jeff
-- 

Joan Crawford  - "I, Joan Crawford, I believe in the dollar.
Everything I earn, I spend." -
http://www.brainyquote.com/quotes/authors/j/joan_crawford.html

From timr at probo.com  Tue Jan  5 23:15:31 2010
From: timr at probo.com (Tim Roberts)
Date: Tue, 05 Jan 2010 14:15:31 -0800
Subject: [python-win32] Still looking for a method to get CPUID info ...
In-Reply-To: <36dec4ff1001051403s28b1590fib76238599d365a61@mail.gmail.com>
References: <36dec4ff1001051242q2ae88c0cx332ced7065b12121@mail.gmail.com>
	<4B43B4F7.8070803@probo.com>
	<36dec4ff1001051403s28b1590fib76238599d365a61@mail.gmail.com>
Message-ID: <4B43BA03.8070600@probo.com>

J wrote:
> On Tue, Jan 5, 2010 at 16:53, Tim Roberts <timr at probo.com> wrote:
>   
>
>> C:\tmp>python
>> Python 2.6.2 (r262:71605, Apr 14 2009, 22:40:02) [MSC v.1500 32 bit
>> (Intel)] onwin32
>> Type "help", "copyright", "credits" or "license" for more information.
>>  >>> import ctypes
>>  >>> c = ctypes.windll.cpuid
>>  >>> hex(c.GetCpuidEcx())
>> '0x444d4163'
>>  >>> hex(c.GetCpuidEdx())
>> '0x69746e65'
>>     
>
> Also have to figure out how to decode the hex value into a more user
> friendly format ;-)  BUT I think there was a white paper or something
> that I have bookmarked somewhere that does have hex values for the
> various flags, at least according to the Intel specs, assuming that
> vendors (in this case IBM) hasn't mucked with them.
>   

Well, those hex values are actually part of the string "AuthenticAMD"
that gets returned in ebx, edx, and ecx in response to the standard
"cpuid" (since I'm running an AMD 64X2).  That's because I used the
"eax=0" variant of cpuid.  For the version you want, you'd set eax to 1
before invoking "cpuid", which would return the feature capability
bits.  Replace the code with this:

GetCpuidEcx   proc   public, style:DWORD
    mov eax, [style]
    cpuid
    mov eax, ecx
    ret
GetCpuidEcx   endp

GetCpuidEdx   proc   public, style:DWORD
    mov eax, [style]
    cpuid
    mov eax, edx
    ret
GetCpuidEdx   endp

Now I can pass the cpuid code as a parameter:

C:\tmp>python
Python 2.6.2 (r262:71605, Apr 14 2009, 22:40:02) [MSC v.1500 32 bit
(Intel)] on
win32
Type "help", "copyright", "credits" or "license" for more information.
 >>> import ctypes
 >>> c = ctypes.windll.cpuid
 >>> c.GetCpuidEcx(1)
1
>>> hex(c.GetCpuidEdx(1))
'0x178bfbff'
 >>>

-- 
Tim Roberts, timr at probo.com
Providenza & Boekelheide, Inc.


From timr at probo.com  Tue Jan  5 23:40:50 2010
From: timr at probo.com (Tim Roberts)
Date: Tue, 05 Jan 2010 14:40:50 -0800
Subject: [python-win32] Still looking for a method to get CPUID info ...
In-Reply-To: <4B43BA03.8070600@probo.com>
References: <36dec4ff1001051242q2ae88c0cx332ced7065b12121@mail.gmail.com>	<4B43B4F7.8070803@probo.com>	<36dec4ff1001051403s28b1590fib76238599d365a61@mail.gmail.com>
	<4B43BA03.8070600@probo.com>
Message-ID: <4B43BFF2.7040105@probo.com>

Tim Roberts wrote:
> J wrote:
>   
>> On Tue, Jan 5, 2010 at 16:53, Tim Roberts <timr at probo.com> wrote:
>>   
>>
>>     
>>> C:\tmp>python
>>> Python 2.6.2 (r262:71605, Apr 14 2009, 22:40:02) [MSC v.1500 32 bit
>>> (Intel)] onwin32
>>> Type "help", "copyright", "credits" or "license" for more information.
>>>  >>> import ctypes
>>>  >>> c = ctypes.windll.cpuid
>>>  >>> hex(c.GetCpuidEcx())
>>> '0x444d4163'
>>>  >>> hex(c.GetCpuidEdx())
>>> '0x69746e65'
>>>     
>>>       
>> Also have to figure out how to decode the hex value into a more user
>> friendly format ;-)  

OK, so I was being overly clever, showing off a bit, demonstrating that
it's possible to build a minimal DLL in assembler.  It turns out you can
do the same thing in Visual C++:

c:\tmp>type cpuid.c
#include <intrin.h>

int
GetCpuidEcx( int var )
{
    int parts[4];
    __cpuid( parts, var );
    return parts[2];
}

int
GetCpuidEdx( int var )
{
    int parts[4];
    __cpuid( parts, var );
    return parts[3];
}
C:\tmp>cl /LD cpuid.c
Microsoft (R) 32-bit C/C++ Optimizing Compiler Version 15.00.21022.08
for 80x86
Copyright (C) Microsoft Corporation.  All rights reserved.

cpuid.c
Microsoft (R) Incremental Linker Version 9.00.21022.08
Copyright (C) Microsoft Corporation.  All rights reserved.

/out:cpuid.dll
/dll
/implib:cpuid.lib
cpuid.obj
   Creating library cpuid.lib and object cpuid.exp

C:\tmp>python
Python 2.6.2 (r262:71605, Apr 14 2009, 22:40:02) [MSC v.1500 32 bit
(Intel)] on win32
Type "help", "copyright", "credits" or "license" for more information.
  >>> import ctypes
  >>> c = ctypes.cdll.cpuid
  >>> c.GetCpuidEcx(1)
1
  >>> hex(c.GetCpuidEdx(1))
'0x178bfbff'
  >>>

-- 
Tim Roberts, timr at probo.com
Providenza & Boekelheide, Inc.


From timr at probo.com  Wed Jan  6 00:46:34 2010
From: timr at probo.com (Tim Roberts)
Date: Tue, 05 Jan 2010 15:46:34 -0800
Subject: [python-win32] Still looking for a method to get CPUID info ...
In-Reply-To: <4B43BFF2.7040105@probo.com>
References: <36dec4ff1001051242q2ae88c0cx332ced7065b12121@mail.gmail.com>	<4B43B4F7.8070803@probo.com>	<36dec4ff1001051403s28b1590fib76238599d365a61@mail.gmail.com>	<4B43BA03.8070600@probo.com>
	<4B43BFF2.7040105@probo.com>
Message-ID: <4B43CF5A.20504@probo.com>

Tim Roberts wrote:
> Tim Roberts wrote:
>   
>> J wrote:
>>   
>>     
>>> On Tue, Jan 5, 2010 at 16:53, Tim Roberts <timr at probo.com> wrote:
>>>   
>>>
>>>     
>>>       
>>>> C:\tmp>python
>>>> Python 2.6.2 (r262:71605, Apr 14 2009, 22:40:02) [MSC v.1500 32 bit
>>>> (Intel)] onwin32
>>>> Type "help", "copyright", "credits" or "license" for more information.
>>>>  >>> import ctypes
>>>>  >>> c = ctypes.windll.cpuid
>>>>  >>> hex(c.GetCpuidEcx())
>>>> '0x444d4163'
>>>>  >>> hex(c.GetCpuidEdx())
>>>> '0x69746e65'
>>>>     
>>>>       
>>>>         
>>> Also have to figure out how to decode the hex value into a more user
>>> friendly format ;-)  
>>>       
>
> OK, so I was being overly clever, showing off a bit, demonstrating that
> it's possible to build a minimal DLL in assembler.  It turns out you can
> do the same thing in Visual C++:
>
> c:\tmp>type cpuid.c
> #include <intrin.h>
>
> int
> GetCpuidEcx( int var )
> {
>     int parts[4];
>     __cpuid( parts, var );
>     return parts[2];
> }
>
> int
> GetCpuidEdx( int var )
> {
>     int parts[4];
>     __cpuid( parts, var );
>     return parts[3];
> }
>   

Well, almost.  Replace both of the "int" lines with this:
    int __declspec(dllexport)

-- 
Tim Roberts, timr at probo.com
Providenza & Boekelheide, Inc.


From rwupole at msn.com  Wed Jan  6 09:58:30 2010
From: rwupole at msn.com (Roger Upole)
Date: Wed, 6 Jan 2010 03:58:30 -0500
Subject: [python-win32]  Boolean type changed in Python 3.0?
Message-ID: <BLU0-SMTP924708C4FF2089A57B0800A3720@phx.gbl>

>From looking at the source of 
PyCom_VariantFromPyObject in oleargs.cpp, 
the check for a boolean is done after the check for
an int.  PyLong_Check is used to determine
if the object should be converted to one of the variant
integer types.  Apparently in Python 3.x, this now returns
True for a boolean whereas it doesn't in 2.x.  We
should be able to just move the boolean check above
the int check without breaking anything in 2.x.


       Roger



From gklein at xs4all.nl  Wed Jan  6 12:33:32 2010
From: gklein at xs4all.nl (Gertjan Klein)
Date: Wed, 06 Jan 2010 12:33:32 +0100
Subject: [python-win32] Still looking for a method to get CPUID info ...
References: <36dec4ff1001051242q2ae88c0cx332ced7065b12121@mail.gmail.com>
	<4B43B4F7.8070803@probo.com>
Message-ID: <i9s8k55vh1kq0kcqdfr55joe3s0046u0kp@4ax.com>

Tim Roberts wrote:

>In my opinion, you could have solved this all much quicker and easier
>just by writing a trivial DLL.

Out of pure curiousity (I used to write a lot of assembler, but those
days are long gone) I tried your code, and it works as advertised -- on
Python 2.6. It fails on Python 2.5 with "WindowsError: exception code
0x20800", but oddly I've also seen error code 0x1020800. I couldn't find
much about either of these errors on the Internet. Any idea what could
be wrong? I know Python 2.5 and 2.6 use different C runtimes, but the
created DLL is completely independent.

Gertjan.



From timr at probo.com  Wed Jan  6 19:32:06 2010
From: timr at probo.com (Tim Roberts)
Date: Wed, 06 Jan 2010 10:32:06 -0800
Subject: [python-win32] Still looking for a method to get CPUID info ...
In-Reply-To: <i9s8k55vh1kq0kcqdfr55joe3s0046u0kp@4ax.com>
References: <36dec4ff1001051242q2ae88c0cx332ced7065b12121@mail.gmail.com>	<4B43B4F7.8070803@probo.com>
	<i9s8k55vh1kq0kcqdfr55joe3s0046u0kp@4ax.com>
Message-ID: <4B44D726.6010101@probo.com>

Gertjan Klein wrote:
> Out of pure curiousity (I used to write a lot of assembler, but those
> days are long gone) I tried your code, and it works as advertised -- on
> Python 2.6. It fails on Python 2.5 with "WindowsError: exception code
> 0x20800", but oddly I've also seen error code 0x1020800. I couldn't find
> much about either of these errors on the Internet. Any idea what could
> be wrong? I know Python 2.5 and 2.6 use different C runtimes, but the
> created DLL is completely independent.
>   

Very interesting.  I tried this on Python 2.5.2 on my Windows 7 64
system (using 32-bit Python).  The C version works fine, but the
assembler version gets a "DLL initialization error".  I fixed that by
adding "mov eax, 1" to the "start" routine.  See if that helps your issue.

-- 
Tim Roberts, timr at probo.com
Providenza & Boekelheide, Inc.


From l at urent.org  Wed Jan  6 20:15:39 2010
From: l at urent.org (lduchesne)
Date: Wed, 6 Jan 2010 11:15:39 -0800 (PST)
Subject: [python-win32]  mapi.ATTACH_EMBEDDED_MSG
Message-ID: <27026941.post@talk.nabble.com>


Hi,

I'm tying to embed a mail message in another mail using the following code:

[...]
imsg = draftFolder.CreateMessage(None, 0)
message = outboxFolder.CreateMessage(None, 0)
attach = message.CreateAttach(None, 0)

attach[1].SetProps([(mapitags.PR_ATTACH_METHOD, mapi.ATTACH_EMBEDDED_MSG),
                    (mapitags.PR_ATTACH_DATA_OBJ, imsg),
                    (mapitags.PR_DISPLAY_NAME, filename),
                    (mapitags.PR_ATTACH_LONG_FILENAME, filename)])
[...]

The problem is, I always get a: "TypeError: Unsupported MAPI property type
0xD" when trying to set the PR_ATTACH_DATA_OBJ property. I tried various
ways of creating a new IMessage to embed (the one above and using CDO) but
it seems the property cannot be set.

Using CDO all the way is not an option.. so has anybody succedded in
embeding a message using the extended MAPI interface?

Note that the embedded message will be created from a message/rfc822 source.

Thanks,
Laurent
-- 
View this message in context: http://old.nabble.com/mapi.ATTACH_EMBEDDED_MSG-tp27026941p27026941.html
Sent from the Python - python-win32 mailing list archive at Nabble.com.


From keating at acm.org  Wed Jan  6 20:39:36 2010
From: keating at acm.org (Paul Keating)
Date: Wed, 06 Jan 2010 20:39:36 +0100
Subject: [python-win32] HKLM\Software\Python\PythonCore\2.6\PythonPath -
 does it serve any purpose?
Message-ID: <024601ca8f08$ca833ff0$4801a8c0@vuurpijlhuisje.local>

What is the registry key HKLM\Software\Python\PythonCore\2.6\PythonPath for? The docs (http://www.python.org/doc/2.6/using/windows.html) say

  Modifying the module search path can also be done through the Windows registry: Edit HKEY_LOCAL_MACHINE\SOFTWARE\Python\PythonCore\version\PythonPath

but this doesn't work. Judging by what I see in the registries of old servers at my site, I think it must have worked in Python 1.5, but it doesn't now, as the following shows:

Export the value of the key:
C:\>regedit /E  xxx  HKEY_LOCAL_MACHINE\SOFTWARE\Python\PythonCore\2.6\PythonPath

Display it:
C:\>type xxx
...
@="F:\\Development\\Python26\\Lib;F:\\Development\\Python26\\DLLs;F:\\Developmen
t\\Python26\\Lib\\lib-tk;F:\\Development\\Python26\\Lib\\site-packages\\win32;F:
\\Development\\Python26\\Lib\\site-packages\\pythonwin;c:\\temp"

Note the temporary addition of c:\temp to the end of the path. Now ask Python what it thinks:

C:\>python
Python 2.6.4 (r264:75708, Oct 26 2009, 08:23:19) [MSC v.1500 32 bit (Intel)] on
win32
Type "help", "copyright", "credits" or "license" for more information.
>>> import sys
>>> sys.path
['', 'C:\\WINNT\\system32\\python26.zip', 'F:\\Development\\Python26\\DLLs', 'F:
\\Development\\Python26\\lib', 'F:\\Development\\Python26\\lib\\plat-win', 'F:\\
Development\\Python26\\lib\\lib-tk', 'F:\\Development\\Python26', 'F:\\Developme
nt\\Python26\\lib\\site-packages', 'F:\\Development\\Python26\\lib\\site-package
s\\win32', 'F:\\Development\\Python26\\lib\\site-packages\\win32\\lib', 'F:\\Dev
elopment\\Python26\\lib\\site-packages\\Pythonwin']
>>>

Not only is c:\temp not in sys.path, even the folders that it has in common with the registry entry are in a different order (eg DLLs and lib), so it's hard to believe that the registry entry has any influence at all on sys.path. Also sys.path has lib\plat-win which I think was introduced about Python 1.5 and I think disappeared again about Python 2.0.

So, why is that registry key there at all? And why does sys.path have the nonexistent plat-win in it, even in Python 3.0? Is it just historical baggage? 

I would like to see the registry key once again working as advertised. I know I can set the pythonpath environment variable, and I also know that environment variables just live in another registry key, but changing that key strikes me as inelegant. Python stuff should live in Software\Python, not in System\CurrentControlSet. 

If the registry key turned out to be a bad idea, it would be illuminating to know why.

--
Paul Keating
The Hague
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-win32/attachments/20100106/73c6d6b8/attachment.htm>

From drobinow at gmail.com  Wed Jan  6 22:06:51 2010
From: drobinow at gmail.com (David Robinow)
Date: Wed, 6 Jan 2010 16:06:51 -0500
Subject: [python-win32] HKLM\Software\Python\PythonCore\2.6\PythonPath -
	does it serve any purpose?
In-Reply-To: <024601ca8f08$ca833ff0$4801a8c0@vuurpijlhuisje.local>
References: <024601ca8f08$ca833ff0$4801a8c0@vuurpijlhuisje.local>
Message-ID: <4eb0089f1001061306p2a5d2836o6a64c021c2a3aa7e@mail.gmail.com>

On Wed, Jan 6, 2010 at 2:39 PM, Paul Keating <keating at acm.org> wrote:
> What is the registry key HKLM\Software\Python\PythonCore\2.6\PythonPath for?
> The docs (http://www.python.org/doc/2.6/using/windows.html) say
>
> Modifying the module search path can also be done through the Windows
> registry: Edit
> HKEY_LOCAL_MACHINE\SOFTWARE\Python\PythonCore\version\PythonPath
>
> but this doesn't work.
...
> So, why is that registry key?there at all? And why?does sys.path?have?the
> nonexistent?plat-win in it, even in Python 3.0? Is?it just historical
> baggage?
>
> I would like to see?the registry key once again working as advertised. I
> know I can?set the pythonpath environment variable, and I also know that
> environment variables just live in another registry key, but changing that
> key strikes me as inelegant. Python stuff should live in Software\Python,
> not in System\CurrentControlSet.
>
> If the registry key?turned out to be a bad idea, it would be illuminating to
> know why.
> --
> Paul Keating
> The Hague
This is really the wrong list. The developers hang out at the python-dev list.
(They'll probably tell you to file a bug report!)

From drobinow at gmail.com  Thu Jan  7 00:17:31 2010
From: drobinow at gmail.com (David Robinow)
Date: Wed, 6 Jan 2010 18:17:31 -0500
Subject: [python-win32] HKLM\Software\Python\PythonCore\2.6\PythonPath -
	does it serve any purpose?
In-Reply-To: <024601ca8f08$ca833ff0$4801a8c0@vuurpijlhuisje.local>
References: <024601ca8f08$ca833ff0$4801a8c0@vuurpijlhuisje.local>
Message-ID: <4eb0089f1001061517n2e1db95che954f66e404a82f9@mail.gmail.com>

On Wed, Jan 6, 2010 at 2:39 PM, Paul Keating <keating at acm.org> wrote:
> What is the registry key HKLM\Software\Python\PythonCore\2.6\PythonPath for?
> The docs (http://www.python.org/doc/2.6/using/windows.html) say
>
> Modifying the module search path can also be done through the Windows
> registry: Edit
> HKEY_LOCAL_MACHINE\SOFTWARE\Python\PythonCore\version\PythonPath
>
>
> but this doesn't work. ...
>
Well, a little more research shows that we both have been editing the
wrong thing.
 See:
http://yuji.wordpress.com/2009/08/31/python-on-windows-setting-pythonpath-environment-variable/

for a description of the correct technique. It works for me.
This is definitely a documentation bug. I don't see how anybody could
be expected to figure this out from the current docs.

From skippy.hammond at gmail.com  Thu Jan  7 05:52:05 2010
From: skippy.hammond at gmail.com (Mark Hammond)
Date: Thu, 07 Jan 2010 15:52:05 +1100
Subject: [python-win32] mapi.ATTACH_EMBEDDED_MSG
In-Reply-To: <27026941.post@talk.nabble.com>
References: <27026941.post@talk.nabble.com>
Message-ID: <4B456875.7050809@gmail.com>

On 7/01/2010 6:15 AM, lduchesne wrote:
>
> Hi,
>
> I'm tying to embed a mail message in another mail using the following code:
>
> [...]
> imsg = draftFolder.CreateMessage(None, 0)
> message = outboxFolder.CreateMessage(None, 0)
> attach = message.CreateAttach(None, 0)
>
> attach[1].SetProps([(mapitags.PR_ATTACH_METHOD, mapi.ATTACH_EMBEDDED_MSG),
>                      (mapitags.PR_ATTACH_DATA_OBJ, imsg),
>                      (mapitags.PR_DISPLAY_NAME, filename),
>                      (mapitags.PR_ATTACH_LONG_FILENAME, filename)])

You will need something like:

attachMsg = imsg.OpenProperty(mapitags.PR_ATTACH_DATA_OBJ, 
mapi.IID_IMessage, 0, mapi.MAPI_CREATE)

After which you have an IMessage for the new item.  You can then 
probably use the CopyTo() method to copy from the source to the new one.

HTH,

Mark

From santosh.mce at gmail.com  Thu Jan  7 06:40:51 2010
From: santosh.mce at gmail.com (Santosh Mohan)
Date: Thu, 7 Jan 2010 11:10:51 +0530
Subject: [python-win32] Porting code from Python 2.x code to Python 3.x
In-Reply-To: <427239421001062137s43a677f3oeff48ec6e03d0cec@mail.gmail.com>
References: <427239421001062137s43a677f3oeff48ec6e03d0cec@mail.gmail.com>
Message-ID: <427239421001062140g19ad45c6hef94c01820c73157@mail.gmail.com>

HI,

I installed new version of Python 3.1. I tried to run my older programs, but
I couldn't run.

I had used "print" statements, in Newer version print statement syntax is
changed.

Python2.x
>>>x=20
>>>print x
>>> 20

Python 3.x
>>>x=20
>>> print x
  File "<stdin>", line 1
    print x
          ^
SyntaxError: invalid syntax
>>>print(x)
>>>20

Is there any utility for running the older version programs?


Regards,
Santosh M
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-win32/attachments/20100107/1745857e/attachment.htm>

From mvilas at gmail.com  Thu Jan  7 07:01:33 2010
From: mvilas at gmail.com (Mario Alejandro Vilas Jerez)
Date: Thu, 7 Jan 2010 03:01:33 -0300
Subject: [python-win32] Porting code from Python 2.x code to Python 3.x
In-Reply-To: <427239421001062140g19ad45c6hef94c01820c73157@mail.gmail.com>
References: <427239421001062137s43a677f3oeff48ec6e03d0cec@mail.gmail.com>
	<427239421001062140g19ad45c6hef94c01820c73157@mail.gmail.com>
Message-ID: <3fbf862f1001062201r6fa57c87h69b5e5c8265f857e@mail.gmail.com>

Try this script, it comes with your Python 3 installation:

http://docs.python.org/library/2to3.html

You may have to review the source code manually in non trivial cases,
though.

Regards,
-Mario

On Thu, Jan 7, 2010 at 2:40 AM, Santosh Mohan <santosh.mce at gmail.com> wrote:

> HI,
>
> I installed new version of Python 3.1. I tried to run my older programs,
> but I couldn't run.
>
> I had used "print" statements, in Newer version print statement syntax is
> changed.
>
> Python2.x
> >>>x=20
> >>>print x
> >>> 20
>
> Python 3.x
> >>>x=20
> >>> print x
>   File "<stdin>", line 1
>     print x
>           ^
> SyntaxError: invalid syntax
> >>>print(x)
> >>>20
>
> Is there any utility for running the older version programs?
>
>
> Regards,
> Santosh M
>
> _______________________________________________
> python-win32 mailing list
> python-win32 at python.org
> http://mail.python.org/mailman/listinfo/python-win32
>
>


-- 
HONEY: I want to? put some powder on my nose.
GEORGE: Martha, won?t you show her where we keep the euphemism?
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-win32/attachments/20100107/3b79a05a/attachment.htm>

From gklein at xs4all.nl  Thu Jan  7 11:26:34 2010
From: gklein at xs4all.nl (Gertjan Klein)
Date: Thu, 07 Jan 2010 11:26:34 +0100
Subject: [python-win32] Still looking for a method to get CPUID info ...
References: <36dec4ff1001051242q2ae88c0cx332ced7065b12121@mail.gmail.com>	<4B43B4F7.8070803@probo.com>
	<i9s8k55vh1kq0kcqdfr55joe3s0046u0kp@4ax.com>
	<4B44D726.6010101@probo.com>
Message-ID: <6qcbk59t7irtkkee0dul389qjrvgrko231@4ax.com>

Tim Roberts wrote:

>Very interesting.  I tried this on Python 2.5.2 on my Windows 7 64
>system (using 32-bit Python).  The C version works fine, but the
>assembler version gets a "DLL initialization error".  I fixed that by
>adding "mov eax, 1" to the "start" routine.  See if that helps your issue.

It doesn't, I get the same error (either 0x1020800 or 0x20800, the
former the most often, which makes me wonder if the error contains bit
flags).

Interestingly, if I place a ret after the mov eax, [style] instruction,
or even comment out the cpuid instruction, I don't get an exception.
This makes me wonder whether there is something permission-related that
is different between Python 2.5 and 2.6 -- perhaps the 2.5 executable is
not allowed to execute that instruction?

Gertjan.



From timr at probo.com  Thu Jan  7 18:36:09 2010
From: timr at probo.com (Tim Roberts)
Date: Thu, 07 Jan 2010 09:36:09 -0800
Subject: [python-win32] Still looking for a method to get CPUID info ...
In-Reply-To: <6qcbk59t7irtkkee0dul389qjrvgrko231@4ax.com>
References: <36dec4ff1001051242q2ae88c0cx332ced7065b12121@mail.gmail.com>	<4B43B4F7.8070803@probo.com>	<i9s8k55vh1kq0kcqdfr55joe3s0046u0kp@4ax.com>	<4B44D726.6010101@probo.com>
	<6qcbk59t7irtkkee0dul389qjrvgrko231@4ax.com>
Message-ID: <4B461B89.7040009@probo.com>

Gertjan Klein wrote:
> Interestingly, if I place a ret after the mov eax, [style] instruction,
> or even comment out the cpuid instruction, I don't get an exception.
> This makes me wonder whether there is something permission-related that
> is different between Python 2.5 and 2.6 -- perhaps the 2.5 executable is
> not allowed to execute that instruction?
>   

No, that doesn't make sense.  I don't think CPUID is a privileged
instruction.  Even if it were, that's talking about CPU privileges,
which do not vary from process to process.

However, I'll bet I know what it is.  CPUID trashes ebx, and in the x86
calling sequence, that's supposed to be saved and restored.  One of the
sets of code is probably storing an error code in ebx and assuming that
it survives the CPU cal.  Try changing the two function headers to this:

GetCpuidEcx   proc   public uses ebx, style:DWORD
...
GetCpuidEdx   proc   public uses ebx, style:DWORD

-- 
Tim Roberts, timr at probo.com
Providenza & Boekelheide, Inc.


From gareth at maptel.com.au  Fri Jan  8 04:25:15 2010
From: gareth at maptel.com.au (Gareth Walters)
Date: Fri, 8 Jan 2010 14:25:15 +1100
Subject: [python-win32]  'No such interface supported' question
Message-ID: <8EC4F97637845642809C79C97C0F5440757079B1D9@MAILMAN.maptel.local>

Hi Nick,



I have done some python work before but I new to the COM stuff. I am trying to Create a MXD and layers to it from python. I have been looking at this PythonNet stuff and it is all getting a bit blurry.



I was wondering if you could give me some pointers to basically get python recognizing the ArcMap class or do you have any code that would help me get to that point?



Look forward to hearing from you.



Cheers,



Gareth

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-win32/attachments/20100108/1c153955/attachment.htm>

From gklein at xs4all.nl  Fri Jan  8 07:02:53 2010
From: gklein at xs4all.nl (Gertjan Klein)
Date: Fri, 08 Jan 2010 07:02:53 +0100
Subject: [python-win32] Still looking for a method to get CPUID info ...
References: <36dec4ff1001051242q2ae88c0cx332ced7065b12121@mail.gmail.com>	<4B43B4F7.8070803@probo.com>	<i9s8k55vh1kq0kcqdfr55joe3s0046u0kp@4ax.com>	<4B44D726.6010101@probo.com>
	<6qcbk59t7irtkkee0dul389qjrvgrko231@4ax.com>
	<4B461B89.7040009@probo.com>
Message-ID: <7phdk5thgprdh9naaul3dodfq5l924athf@4ax.com>

Tim Roberts wrote:

>However, I'll bet I know what it is.  CPUID trashes ebx, and in the x86
>calling sequence, that's supposed to be saved and restored.

That's it! Now, both Python 2.5 and 2.6 (and 3.1, if I change the print
statement) run this code successfully. Does that mean there is a
difference in calling sequence between Python 2.5 and 2.6? Or do they
have different expectations of which registers are left untouched?

Thanks for sticking with this. I often save little bits of code from
this newsgroup that I might want to use later, and now I have a little
project that shows me how to create a DLL from assembly code and call it
from Python.

Gertjan.



From davea at ieee.org  Fri Jan  8 15:29:46 2010
From: davea at ieee.org (Dave Angel)
Date: Fri, 08 Jan 2010 09:29:46 -0500
Subject: [python-win32] Still looking for a method to get CPUID info ...
In-Reply-To: <7phdk5thgprdh9naaul3dodfq5l924athf@4ax.com>
References: <36dec4ff1001051242q2ae88c0cx332ced7065b12121@mail.gmail.com>	<4B43B4F7.8070803@probo.com>	<i9s8k55vh1kq0kcqdfr55joe3s0046u0kp@4ax.com>	<4B44D726.6010101@probo.com>	<6qcbk59t7irtkkee0dul389qjrvgrko231@4ax.com>	<4B461B89.7040009@probo.com>
	<7phdk5thgprdh9naaul3dodfq5l924athf@4ax.com>
Message-ID: <4B47415A.9020309@ieee.org>

Gertjan Klein wrote:
> Tim Roberts wrote:
>
>   
>> However, I'll bet I know what it is.  CPUID trashes ebx, and in the x86
>> calling sequence, that's supposed to be saved and restored.
>>     
>
> That's it! Now, both Python 2.5 and 2.6 (and 3.1, if I change the print
> statement) run this code successfully. Does that mean there is a
> difference in calling sequence between Python 2.5 and 2.6? Or do they
> have different expectations of which registers are left untouched?
>
> Thanks for sticking with this. I often save little bits of code from
> this newsgroup that I might want to use later, and now I have a little
> project that shows me how to create a DLL from assembly code and call it
> from Python.
>
> Gertjan.
>
>
>   
On general principle, I'd use
    push ebx
...
    pop ebx

pairs around the cpuid instruction, if trashing ebx is a problem.  Seems 
to me it's safer than making any assumptions about implicit 
saves/restores.  I haven't looked at the cpuid instruction for years, 
but I'd push/pop any register (other than EAX) it modifies, and in 
between them do the cpuid and an appropriate move of whichever register.

DaveA


From mail at timgolden.me.uk  Fri Jan  8 18:30:02 2010
From: mail at timgolden.me.uk (Tim Golden)
Date: Fri, 08 Jan 2010 17:30:02 +0000
Subject: [python-win32] Question concerning wmi c.CIM_DataFile
In-Reply-To: <4B425B33.2060003@co.marshall.ia.us>
References: <2E8EE617FF5865459F6EF9217BC466E2063E3C@opr-pr-exc-01.rgfc.main>	<4B423A29.9050305@probo.com>
	<4B425B33.2060003@co.marshall.ia.us>
Message-ID: <4B476B9A.4080009@timgolden.me.uk>

On 04/01/2010 21:18, Mike Driscoll wrote:
> The OP should check out Tim Golden's winshell module as it gives access
> to the common folders quite easily:
>
> http://timgolden.me.uk/python/winshell.html

Thanks for the plug, Mike. Problem is that the
OP is after running this code on multiple machines (ie
remotely) so the winshell approach won't work.

TJG

From mail at timgolden.me.uk  Fri Jan  8 18:37:04 2010
From: mail at timgolden.me.uk (Tim Golden)
Date: Fri, 08 Jan 2010 17:37:04 +0000
Subject: [python-win32] Exception: object has no attribute
	'Win32_Process'
In-Reply-To: <a0dfa4f11001040906m41009b57o214c58272768dedd@mail.gmail.com>
References: <a0dfa4f11001040906m41009b57o214c58272768dedd@mail.gmail.com>
Message-ID: <4B476D40.8030805@timgolden.me.uk>

On 04/01/2010 17:06, Wade Dyck wrote:
[...]

> At this point I want to use Win32_Process.create() to xcopy some files from
> a third server, so I switch to delegation since impersonate won't allow
> access to network resources:
>
>      remote = wmi.connect_server(server=self.server, user=r"%s\%s"
>              % (self.domain, self.user), password=self.password,
>              impersonation_level="delegate")
>      self.c = wmi.WMI(wmi=remote)
>
> But then when I try to create the process using:
>
>      pid, res = self.c.Win32_Process.Create('cmd.exe /c xcopy %s %s \
>                      /s /e /i /y /c /q' % (src, dest))

As a tiny aside here, I suspect that: (a) you don't need the cmd.exe
to run xcopy: it's actually an executable; and (b) you might to
double-quote the %s params to xcopy. Neither of these things are
going to be causing the error you report, but just to be helpful...


>
> I get the following exception:
>
> AttributeError: '<win32com.gen_py.Microsoft WMI Scripting V1.2
> Library.ISWbemSer
> vices instance at 0x12790040>' object has no attribute 'Win32_Process'
>
> However, this works fine if I don't set the userAccountControl attribute
> above using adsi. It also seems specific to actually setting the attribute,
> as the code is always using adsi to get DNs and the current value of
> userAccountControl.

It is peculiar, assuming you are running exactly the same code, only
with AD setting switched as you describe. ie there's no possibility
that self.c has become something other than a WMI instance?

I'll try to look it at it when I get back to work (which won't be
till Monday now; I'm just back from a week away). If you manage
to fix it in the meantime, please do post back.

TJG

From timr at probo.com  Fri Jan  8 19:04:36 2010
From: timr at probo.com (Tim Roberts)
Date: Fri, 08 Jan 2010 10:04:36 -0800
Subject: [python-win32] Still looking for a method to get CPUID info ...
In-Reply-To: <4B47415A.9020309@ieee.org>
References: <36dec4ff1001051242q2ae88c0cx332ced7065b12121@mail.gmail.com>	<4B43B4F7.8070803@probo.com>	<i9s8k55vh1kq0kcqdfr55joe3s0046u0kp@4ax.com>	<4B44D726.6010101@probo.com>	<6qcbk59t7irtkkee0dul389qjrvgrko231@4ax.com>	<4B461B89.7040009@probo.com>	<7phdk5thgprdh9naaul3dodfq5l924athf@4ax.com>
	<4B47415A.9020309@ieee.org>
Message-ID: <4B4773B4.2030006@probo.com>

Dave Angel wrote:
>>   
> On general principle, I'd use
>    push ebx
> ...
>    pop ebx
>
> pairs around the cpuid instruction, if trashing ebx is a problem. 
> Seems to me it's safer than making any assumptions about implicit
> saves/restores.

That's what the "uses ebx" clause that I added to the "proc" statement
does.  It pushes ebx on entry, and pops it at every "ret".  The
advantage of that is that I don't have to worry about saving it
repeatedly throughout the function.  I'm free to do whatever I want to
ebx internally.


> I haven't looked at the cpuid instruction for years, but I'd push/pop
> any register (other than EAX) it modifies, and in between them do the
> cpuid and an appropriate move of whichever register.

Well, cpuid is kind of an odd exception, because it modifies all of the
general purpose registers (eax, ebx, ecx, edx) in one fell swoop.

-- 
Tim Roberts, timr at probo.com
Providenza & Boekelheide, Inc.


From timr at probo.com  Fri Jan  8 19:04:39 2010
From: timr at probo.com (Tim Roberts)
Date: Fri, 08 Jan 2010 10:04:39 -0800
Subject: [python-win32] Still looking for a method to get CPUID info ...
In-Reply-To: <7phdk5thgprdh9naaul3dodfq5l924athf@4ax.com>
References: <36dec4ff1001051242q2ae88c0cx332ced7065b12121@mail.gmail.com>	<4B43B4F7.8070803@probo.com>	<i9s8k55vh1kq0kcqdfr55joe3s0046u0kp@4ax.com>	<4B44D726.6010101@probo.com>	<6qcbk59t7irtkkee0dul389qjrvgrko231@4ax.com>	<4B461B89.7040009@probo.com>
	<7phdk5thgprdh9naaul3dodfq5l924athf@4ax.com>
Message-ID: <4B4773B7.9060206@probo.com>

Gertjan Klein wrote:
> Tim Roberts wrote
>> However, I'll bet I know what it is.  CPUID trashes ebx, and in the x86
>> calling sequence, that's supposed to be saved and restored.
>>     
>
> That's it! Now, both Python 2.5 and 2.6 (and 3.1, if I change the print
> statement) run this code successfully. Does that mean there is a
> difference in calling sequence between Python 2.5 and 2.6? Or do they
> have different expectations of which registers are left untouched?
>   

This could be something as obscure as different compilers, or even
different optimization settings.  The generated code in 2.5 happened to
store something in ebx and expected it to survive across the API call (a
reasonable expectation!).  The generated code in 2.6 did not.

-- 
Tim Roberts, timr at probo.com
Providenza & Boekelheide, Inc.


From timr at probo.com  Fri Jan  8 19:13:11 2010
From: timr at probo.com (Tim Roberts)
Date: Fri, 08 Jan 2010 10:13:11 -0800
Subject: [python-win32] 'No such interface supported' question
In-Reply-To: <8EC4F97637845642809C79C97C0F5440757079B1D9@MAILMAN.maptel.local>
References: <8EC4F97637845642809C79C97C0F5440757079B1D9@MAILMAN.maptel.local>
Message-ID: <4B4775B7.8030206@probo.com>

Gareth Walters wrote:
>
> Hi Nick,
>
>  
>
> I have done some python work before but I new to the COM stuff. I am
> trying to Create a MXD and layers to it from python. I have been
> looking at this PythonNet stuff and it is all getting a bit blurry.
>
>  
>
> I was wondering if you could give me some pointers to basically get
> python recognizing the ArcMap class or do you have any code that would
> help me get to that point?
>

There are a number of web resources for this.  Google for "arcmap
python" and you should find some good hints.  It looks like there are
even some Python modules available for this, although I don't know if
they apply to your particular case.

If you do need to use COM, you'll need to know the name of the COM class
object to create.  You would then do something like:
    arc = win32com.client.Dispatch("ArcMap.Application")
Then you can use methods on the "arc" object.  Note that I'm just making
up that name -- I don't know what the class really is.

-- 
Tim Roberts, timr at probo.com
Providenza & Boekelheide, Inc.


From l at urent.org  Mon Jan 11 13:50:00 2010
From: l at urent.org (Laurent Duchesne)
Date: Mon, 11 Jan 2010 07:50:00 -0500
Subject: [python-win32] =?utf-8?q?mapi=2EATTACH=5FEMBEDDED=5FMSG?=
In-Reply-To: <4B456875.7050809@gmail.com>
References: <27026941.post@talk.nabble.com> <4B456875.7050809@gmail.com>
Message-ID: <fd7363ee3fcb26e8dae7d8b31292d838@localhost>


Thanks, it's working now :)

In case someone wants to know, I will be converting my .eml source file to
a message object using this call:
IConverterSession::MIMEToMAPI  (outlook 2003+)
http://msdn.microsoft.com/en-us/library/aa193062(office.11).aspx

Since not all calls are available using python, I will use a C++ wrapper
which looks something like this:
std::wstring ConvertMIMEToMAPI(const std::wstring& p_rSourceFilename);
Where the return value is a filename containing a compoung object
(CLSID_MailMessage) which I hope I will be able to read back in python.

Laurent

On Thu, 07 Jan 2010 15:52:05 +1100, Mark Hammond
<skippy.hammond at gmail.com>
wrote:
> On 7/01/2010 6:15 AM, lduchesne wrote:
>>
>> Hi,
>>
>> I'm tying to embed a mail message in another mail using the following
>> code:
>>
>> [...]
>> imsg = draftFolder.CreateMessage(None, 0)
>> message = outboxFolder.CreateMessage(None, 0)
>> attach = message.CreateAttach(None, 0)
>>
>> attach[1].SetProps([(mapitags.PR_ATTACH_METHOD,
>> mapi.ATTACH_EMBEDDED_MSG),
>>                      (mapitags.PR_ATTACH_DATA_OBJ, imsg),
>>                      (mapitags.PR_DISPLAY_NAME, filename),
>>                      (mapitags.PR_ATTACH_LONG_FILENAME, filename)])
> 
> You will need something like:
> 
> attachMsg = imsg.OpenProperty(mapitags.PR_ATTACH_DATA_OBJ, 
> mapi.IID_IMessage, 0, mapi.MAPI_CREATE)
> 
> After which you have an IMessage for the new item.  You can then 
> probably use the CopyTo() method to copy from the source to the new one.
> 
> HTH,
> 
> Mark

From mdriscoll at co.marshall.ia.us  Mon Jan 11 15:57:28 2010
From: mdriscoll at co.marshall.ia.us (Mike Driscoll)
Date: Mon, 11 Jan 2010 08:57:28 -0600
Subject: [python-win32] Question concerning wmi c.CIM_DataFile
In-Reply-To: <4B476B9A.4080009@timgolden.me.uk>
References: <2E8EE617FF5865459F6EF9217BC466E2063E3C@opr-pr-exc-01.rgfc.main>	<4B423A29.9050305@probo.com>	<4B425B33.2060003@co.marshall.ia.us>
	<4B476B9A.4080009@timgolden.me.uk>
Message-ID: <4B4B3C58.4050403@co.marshall.ia.us>

On 1:59 PM, Tim Golden wrote:
> On 04/01/2010 21:18, Mike Driscoll wrote:
>> The OP should check out Tim Golden's winshell module as it gives access
>> to the common folders quite easily:
>>
>> http://timgolden.me.uk/python/winshell.html
>
> Thanks for the plug, Mike. Problem is that the
> OP is after running this code on multiple machines (ie
> remotely) so the winshell approach won't work.
>
> TJG
>

I must have read the question wrong then. At my workplace, we use the 
winshell module in our login scripts, so it works fine on "remote" 
machines. Alas, I didn't realize he was wanting to do it from his own 
machine or I wouldn't have said that.

Anyway, thanks for the handy module. I use it a quite a bit for various 
manipulations of files on user accounts.

- Mike

From ross at biostat.ucsf.edu  Tue Jan 12 01:35:35 2010
From: ross at biostat.ucsf.edu (Ross Boylan)
Date: Mon, 11 Jan 2010 16:35:35 -0800
Subject: [python-win32] Guaranteed cleanup code execution?
Message-ID: <1263256535.15072.119.camel@corn.betterworld.us>

Is there a way to ensure that some cleanup code will get run whenever my
process terminates?

I've tried signal and try:finally:, while noting that, e.g., atexit says
it doesn't handle interrupts.  I can catch many circumstances, but if I
kill the process from task manager my cleanup doesn't happen (at least,
print "I'm cleaning up" doesn't come out).

I don't see anything in win32api that looks appropriate, and I don't
know if Windows provides any sure way to do what I want.  There is a
call to handle some very abnormal exits, but they are sufficiently bad
that, like SIGSEGV, I don't want to run the cleanup code (even if I
can).

The process will eventually run as a service.  I assume there is some
protocol for managing normal service control manager requests to start
and stop.

Thanks for any pointers.
Ross Boylan


From mvilas at gmail.com  Tue Jan 12 03:33:23 2010
From: mvilas at gmail.com (Mario Alejandro Vilas Jerez)
Date: Mon, 11 Jan 2010 23:33:23 -0300
Subject: [python-win32] Guaranteed cleanup code execution?
In-Reply-To: <1263256535.15072.119.camel@corn.betterworld.us>
References: <1263256535.15072.119.camel@corn.betterworld.us>
Message-ID: <3fbf862f1001111833s2030b7beuf3fca2cf5c41f902@mail.gmail.com>

If your process is being terminated forcefully (like from the task manager)
then you probably don't want to run any cleanup code. If you really must you
could code something in C, put it in a DLL library and have it executed in
DllMain when the library is unloaded. (And I think you'll still have
problems if someone with enough privileges calls TerminateProcess on your
process).

But what I think you're really looking for is the Service Control Manager:

http://msdn.microsoft.com/en-us/library/ms685150(VS.85).aspx

The ControlService API lets you stop services (among other things):

http://msdn.microsoft.com/en-us/library/ms682108(VS.85).aspx

And the StartService API lets you start a service manually:

http://msdn.microsoft.com/en-us/library/ms686321(VS.85).aspx

Hope this helps! :)
-Mario

On Mon, Jan 11, 2010 at 9:35 PM, Ross Boylan <ross at biostat.ucsf.edu> wrote:

> Is there a way to ensure that some cleanup code will get run whenever my
> process terminates?
>
> I've tried signal and try:finally:, while noting that, e.g., atexit says
> it doesn't handle interrupts.  I can catch many circumstances, but if I
> kill the process from task manager my cleanup doesn't happen (at least,
> print "I'm cleaning up" doesn't come out).
>
> I don't see anything in win32api that looks appropriate, and I don't
> know if Windows provides any sure way to do what I want.  There is a
> call to handle some very abnormal exits, but they are sufficiently bad
> that, like SIGSEGV, I don't want to run the cleanup code (even if I
> can).
>
> The process will eventually run as a service.  I assume there is some
> protocol for managing normal service control manager requests to start
> and stop.
>
> Thanks for any pointers.
> Ross Boylan
>
> _______________________________________________
> python-win32 mailing list
> python-win32 at python.org
> http://mail.python.org/mailman/listinfo/python-win32
>



-- 
HONEY: I want to? put some powder on my nose.
GEORGE: Martha, won?t you show her where we keep the euphemism?
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-win32/attachments/20100111/12f5cc6c/attachment.htm>

From ah.kode at gmail.com  Tue Jan 12 13:03:07 2010
From: ah.kode at gmail.com (a h)
Date: Tue, 12 Jan 2010 17:33:07 +0530
Subject: [python-win32] passing c structure into python function
Message-ID: <a4653d3d1001120403j47e7df24hdded4d74c6244b06@mail.gmail.com>

Hi
I have witten C code which call a function defined in python, passing an
structure, but couldn't know how to pass c structure into a python
function(data conversion) and how to write python script function which
takes an c structure and fill value into it (commented portion)

I have looked at struct module but couldn't understand that how it works for
below.

typedef struct employee
{
 char emp_name[10];
 int emp_ID;
}emp;

int main()
{
emp e = {0};
Pyobject *pName, *pModule, *pFunc, *pArgs,*pValue;
Py_Initialize();
pName = PyString_FromString("test");
pModule = PyImport_Import(pName);
Py_DECREF(pName);
if (pModule != NULL)
{
pFunc = PyObject_GetAttrString(pModule, "testStruct");
if(pFunc && PyCallable_Check(pFunc))
{
pArgs = e //******************logic to assign struct employee into pArgs
pValue = PyObject_CallObject(pFunc, pArgs);
}
}
Py_DECREF(pValue);
Py_DECREF(pModule);
Py_Finalize();
return 0;
}

and now python script test.py
def testStruct():    # **** need any parameter will be specify
#**** assign value into the cStructure
f=open(filename)
f.write(cStructure.Value) # **** value to write in file

Thanks in advance
regards
ah
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-win32/attachments/20100112/ae484ba7/attachment.htm>

From timr at probo.com  Tue Jan 12 18:19:24 2010
From: timr at probo.com (Tim Roberts)
Date: Tue, 12 Jan 2010 09:19:24 -0800
Subject: [python-win32] passing c structure into python function
In-Reply-To: <a4653d3d1001120403j47e7df24hdded4d74c6244b06@mail.gmail.com>
References: <a4653d3d1001120403j47e7df24hdded4d74c6244b06@mail.gmail.com>
Message-ID: <4B4CAF1C.8050304@probo.com>

a h wrote:
>
> I have witten C code which call a function defined in python, passing
> an structure, but couldn't know how to pass c structure into a python
> function(data conversion) and how to write python script
> function which takes an c structure and fill value into it (commented
> portion)
>  
> I have looked at struct module but couldn't understand that how it
> works for below.
>  
> typedef struct employee
> {
>  char emp_name[10];
>  int emp_ID;
> }emp;

I would pass the struct as a string of 14 bytes, then use
    emp_name, emp_ID = struct.unpack( '=10cI', incoming )

class employee:
    def __init__( self, name, id ):
        self.name = name
        self.id = id
    def to_file( self, f ):
        f.write( struct,pack( '=10cI', self.name, self.id )
    def from_string( self, s ):
        self.name, self.id = struct.unpack( '=10cI', s )

f = open( 'testing', 'w' )
e = employee( 'Testing', 923 )
e.to_file( f )

-- 
Tim Roberts, timr at probo.com
Providenza & Boekelheide, Inc.


From aahz at pythoncraft.com  Tue Jan 12 20:41:34 2010
From: aahz at pythoncraft.com (Aahz)
Date: Tue, 12 Jan 2010 11:41:34 -0800
Subject: [python-win32] passing c structure into python function
In-Reply-To: <4B4CAF1C.8050304@probo.com>
References: <a4653d3d1001120403j47e7df24hdded4d74c6244b06@mail.gmail.com>
	<4B4CAF1C.8050304@probo.com>
Message-ID: <20100112194134.GA27959@panix.com>

On Tue, Jan 12, 2010, Tim Roberts wrote:
>
> I would pass the struct as a string of 14 bytes, then use
>     emp_name, emp_ID = struct.unpack( '=10cI', incoming )
> 
> class employee:
>     def __init__( self, name, id ):
>         self.name = name
>         self.id = id
>     def to_file( self, f ):
>         f.write( struct,pack( '=10cI', self.name, self.id )
>     def from_string( self, s ):
>         self.name, self.id = struct.unpack( '=10cI', s )

Why are you using the asymmetric "to_file" and "from_string" methods?
Just for examples?

BTW, I think that in example code it's a Good Idea to follow PEP8...
-- 
Aahz (aahz at pythoncraft.com)           <*>         http://www.pythoncraft.com/

"If you think it's expensive to hire a professional to do the job, wait
until you hire an amateur."  --Red Adair

From timr at probo.com  Tue Jan 12 23:07:32 2010
From: timr at probo.com (Tim Roberts)
Date: Tue, 12 Jan 2010 14:07:32 -0800
Subject: [python-win32] passing c structure into python function
In-Reply-To: <20100112194134.GA27959@panix.com>
References: <a4653d3d1001120403j47e7df24hdded4d74c6244b06@mail.gmail.com>	<4B4CAF1C.8050304@probo.com>
	<20100112194134.GA27959@panix.com>
Message-ID: <4B4CF2A4.4000504@probo.com>

Aahz wrote:
> On Tue, Jan 12, 2010, Tim Roberts wrote:
>   
>> I would pass the struct as a string of 14 bytes, then use
>>     emp_name, emp_ID = struct.unpack( '=10cI', incoming )
>>
>> class employee:
>>     def __init__( self, name, id ):
>>         self.name = name
>>         self.id = id
>>     def to_file( self, f ):
>>         f.write( struct,pack( '=10cI', self.name, self.id )
>>     def from_string( self, s ):
>>         self.name, self.id = struct.unpack( '=10cI', s )
>>     
>
> Why are you using the asymmetric "to_file" and "from_string" methods?
> Just for examples?
>   

Yes.  That nagged me when I wrote it, and I decided to leave it as an
exercise for the reader...

-- 
Tim Roberts, timr at probo.com
Providenza & Boekelheide, Inc.


From ah.kode at gmail.com  Wed Jan 13 15:31:43 2010
From: ah.kode at gmail.com (a h)
Date: Wed, 13 Jan 2010 20:01:43 +0530
Subject: [python-win32] passing c structure into python function
In-Reply-To: <a4653d3d1001120403j47e7df24hdded4d74c6244b06@mail.gmail.com>
References: <a4653d3d1001120403j47e7df24hdded4d74c6244b06@mail.gmail.com>
Message-ID: <a4653d3d1001130631m4b11ee2awf18b48c9a9099aef@mail.gmail.com>

Hi
thanks for your response
in the last piece of c code i want to know how do i pass structure into
python function in c. I have tried to call a python function which takes
string as parameter.

i have passed string parameter as :
const char *str_pArgs = "hello";
pArgs = PyTuple_New(1);
pValue = PyString_FromString(str_pArgs);
PyTuple_SetItem(pArgs, 0,pValue);
pValue = PyObject_CallObject(pFunc, pArgs);

similarly how do i convert this structure for python function argument.

->pArgs = e //******************logic to assign struct employee into pArgs

regards
ah

On Tue, Jan 12, 2010 at 5:33 PM, a h <ah.kode at gmail.com> wrote:

> Hi
> I have witten C code which call a function defined in python, passing an
> structure, but couldn't know how to pass c structure into a python
> function(data conversion) and how to write python script function which
> takes an c structure and fill value into it (commented portion)
>
> I have looked at struct module but couldn't understand that how it works
> for below.
>
> typedef struct employee
> {
>  char emp_name[10];
>  int emp_ID;
> }emp;
>
> int main()
> {
> emp e = {0};
> Pyobject *pName, *pModule, *pFunc, *pArgs,*pValue;
> Py_Initialize();
> pName = PyString_FromString("test");
> pModule = PyImport_Import(pName);
> Py_DECREF(pName);
> if (pModule != NULL)
> {
> pFunc = PyObject_GetAttrString(pModule, "testStruct");
> if(pFunc && PyCallable_Check(pFunc))
> {
> pArgs = e //******************logic to assign struct employee into pArgs
> pValue = PyObject_CallObject(pFunc, pArgs);
> }
> }
> Py_DECREF(pValue);
> Py_DECREF(pModule);
> Py_Finalize();
> return 0;
> }
>
> and now python script test.py
> def testStruct():    # **** need any parameter will be specify
> #**** assign value into the cStructure
> f=open(filename)
> f.write(cStructure.Value) # **** value to write in file
>
> Thanks in advance
> regards
> ah
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-win32/attachments/20100113/892fd93b/attachment.htm>

From gagsl-py2 at yahoo.com.ar  Wed Jan 13 17:56:21 2010
From: gagsl-py2 at yahoo.com.ar (Gabriel Genellina)
Date: Wed, 13 Jan 2010 13:56:21 -0300
Subject: [python-win32] Guaranteed cleanup code execution?
References: <1263256535.15072.119.camel@corn.betterworld.us>
	<3fbf862f1001111833s2030b7beuf3fca2cf5c41f902@mail.gmail.com>
Message-ID: <op.u6hdb7alx6zn5v@gabriel.sleipnir-bsas>

En Mon, 11 Jan 2010 23:33:23 -0300, Mario Alejandro Vilas Jerez  
<mvilas at gmail.com> escribi?:

> But what I think you're really looking for is the Service Control  
> Manager:
>
> http://msdn.microsoft.com/en-us/library/ms685150(VS.85).aspx
>
> The ControlService API lets you stop services (among other things):
>
> http://msdn.microsoft.com/en-us/library/ms682108(VS.85).aspx
>
> And the StartService API lets you start a service manually:
>
> http://msdn.microsoft.com/en-us/library/ms686321(VS.85).aspx

All of which are covered in the win32service and win32serviceutil modules  
inside PyWin32. There are a couple demos too. It's quite easy to write a  
service using the classes provided.

-- 
Gabriel Genellina


From gagsl-py2 at yahoo.com.ar  Wed Jan 13 18:16:14 2010
From: gagsl-py2 at yahoo.com.ar (Gabriel Genellina)
Date: Wed, 13 Jan 2010 14:16:14 -0300
Subject: [python-win32] 'No such interface supported' question
References: <8EC4F97637845642809C79C97C0F5440757079B1D9@MAILMAN.maptel.local>
Message-ID: <op.u6hd9c0zx6zn5v@gabriel.sleipnir-bsas>

En Fri, 08 Jan 2010 00:25:15 -0300, Gareth Walters <gareth at maptel.com.au>  
escribi?:

> I have done some python work before but I new to the COM stuff. I am  
> trying to Create a MXD and layers to it from python. I have been looking  
> at this PythonNet stuff and it is all getting a bit blurry.
> I was wondering if you could give me some pointers to basically get  
> python recognizing the ArcMap class or do you have any code that would  
> help me get to that point?

Mark Hammond's book "Python Programming On Win32" covers using COM objects  
 from Python. You're mostly interested in chapter 12. I think this chapter  
is available for preview from the O'Reilly site.
http://oreilly.com/catalog/9781565926219/

-- 
Gabriel Genellina


From timr at probo.com  Wed Jan 13 19:35:20 2010
From: timr at probo.com (Tim Roberts)
Date: Wed, 13 Jan 2010 10:35:20 -0800
Subject: [python-win32] passing c structure into python function
In-Reply-To: <a4653d3d1001130631m4b11ee2awf18b48c9a9099aef@mail.gmail.com>
References: <a4653d3d1001120403j47e7df24hdded4d74c6244b06@mail.gmail.com>
	<a4653d3d1001130631m4b11ee2awf18b48c9a9099aef@mail.gmail.com>
Message-ID: <4B4E1268.4040203@probo.com>

a h wrote:
>
> thanks for your response
> in the last piece of c code i want to know how do i pass structure
> into python function in c. I have tried to call a python function
> which takes string as parameter.
>  
> i have passed string parameter as :
> const char *str_pArgs = "hello";
> pArgs = PyTuple_New(1);
> pValue = PyString_FromString(str_pArgs);
> PyTuple_SetItem(pArgs, 0,pValue);
> pValue = PyObject_CallObject(pFunc, pArgs);
>  
> similarly how do i convert this structure for python function argument.
>  
> ->pArgs = e //******************logic to assign struct employee into pArgs

There are several ways to do this.  If it were me, looking for the easy
way, I would just pass the structure as a string (by casting the address
of the struct to a (char *)), and unpack it with struct.unpack, as I
showed earlier.

There are many other ways, however.  You could create a tuple from the
struct elements, and pass the elements in their native types.  You could
even go to the trouble of creating a dict or an object in your C code,
populate it, and return it.  It depends on how often this will be used
and who is your audience.

-- 
Tim Roberts, timr at probo.com
Providenza & Boekelheide, Inc.


From lyeoh at pop.jaring.my  Wed Jan 13 19:32:09 2010
From: lyeoh at pop.jaring.my (Lincoln Yeoh)
Date: Thu, 14 Jan 2010 02:32:09 +0800
Subject: [python-win32] Getting a list of "task buttons" or tasks/windows in
 the "alt-tab" list?
Message-ID: <201001131832.o0DIWJPw019549@vsmtp6.jaring.my>

Hi,

With python win32 is there a way to get only the windows that are 
represented by the buttons of the Windows taskbar (or the windows 
listed in the "alt-tab" list)?

I have found various ways of doing so, but I can't seem to get them 
to work with Excel's windows.

 From what I observe (I may be wrong though), Excel by default 
appears to have a main window and some "dummy windows" 
(classname=MS-SDIb) representing the opened Excel documents. If you 
click on those "dummy windows" the relevant excel document gets 
foregrounded in the _main_ window (not the dummy windows).

If three Excel documents are opened, there are four top level Excel 
windows (or at least they appear to be top level to me) - the main, 
and three MS-SDIb ones. But the taskbar and alt-tab list only show 3 
items. What logic should one use to create the same list?

I've created a python program to dynamically map key combos to 
top-level windows[1], it kind of works (enough to be helpful to me), 
but I'm having difficulty handling even common stuff like the Excel 
case so it's not ready to be inflicted on the public yet :).

Thanks,

Link.

[1] Does something similar to some of the stuff I suggest here: 
http://www.mail-archive.com/usability at gnome.org/msg04993.html
and here: 
http://live.gnome.org/GnomeShell/DesignersPlayground/KeyboardShortcuts





From Andrew.MacIntyre at acma.gov.au  Thu Jan 14 00:27:32 2010
From: Andrew.MacIntyre at acma.gov.au (Andrew MacIntyre)
Date: Thu, 14 Jan 2010 10:27:32 +1100
Subject: [python-win32] 'No such interface supported' question
	[SEC=PERSONAL]
In-Reply-To: <8EC4F97637845642809C79C97C0F5440757079B1D9@MAILMAN.maptel.local>
References: <8EC4F97637845642809C79C97C0F5440757079B1D9@MAILMAN.maptel.local>
Message-ID: <7B01D7143C4AD54899EA079D4557562AFEF8F6@ACT01EXC02.internal.govt>

Gareth Walters wrote:

?  I have done some python work before but I new to the COM stuff. I am trying to Create a MXD and layers to it from python. I have been looking at this PythonNet stuff and it is all getting a bit blurry. 

 

?  I was wondering if you could give me some pointers to basically get python recognizing the ArcMap class or do you have any code that would help me get to that point?

 

You will need the comtypes package to access most of the ArcObjects classes.  If you have 

the Python 2.5 installed by the ArcGIS installer, you will have the requisite ctypes extension.

 

Sadly, most of the ArcObjects classes don't support the Dispatch interfaces required by PythonWin.

 

-------------------------> "These thoughts are mine alone!" <---------

Andrew MacIntyre           Operations Branch

tel:   +61 2 6219 5356     Communications Infrastructure Division

fax:   +61 2 6253 3277     Australian Communications & Media Authority

email: andrew.macintyre at acma.gov.au <mailto:andrew.macintyre at acma.gov.au>             http://www.acma.gov.au/ <http://www.acma.gov.au/> 

 

 


If you have received this email in error, please notify the sender immediately and erase all copies of the email and any attachments to it. The information contained in this email and any attachments may be private, confidential and legally privileged or the subject of copyright. If you are not the addressee it may be illegal to review, disclose, use, forward, or distribute this email and/or its contents.
 
Unless otherwise specified, the information in the email and any attachments is intended as a guide only and should not be relied upon as legal or technical advice or regarded as a substitute for legal or technical advice in individual cases. Opinions contained in this email or any of its attachments do not necessarily reflect the opinions of ACMA.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-win32/attachments/20100114/62489d9e/attachment-0001.htm>

From mehgcap at gmail.com  Fri Jan 15 06:05:56 2010
From: mehgcap at gmail.com (Alex Hall)
Date: Fri, 15 Jan 2010 00:05:56 -0500
Subject: [python-win32] WMI troubles!
Message-ID: <558CA385CC304B349F3FEB6D62A466DC@AlexPC>

Hello all,
I just joined this list. I am pretty new to Python, but I really like and 
hope to get more into it; right now, it is my second-favorite, second only 
to java. I am a computer science major in my third year and currently only 
program as a hobby/for something to do. Also, I am blind so I use JAWS for 
Windows (from http://www.freedomscientific.com) to do all my computer work; 
basically, this program reads me the screen. I only mention this so the 
projects I am doing make sense and so you understand why suggesting things 
like looking at a diagram or an animation will not work.

Anyway, onto the problem. I am working on a simple resource monitor (I have 
never found one that works well with screen readers, so I am writing one). I 
am using python2.6 with all the win32 libs installed. My monitor will use 
wmi to get all of its information, and it was going pretty well until a few 
hours ago, when I started receiving seemingly random errors. I can call my 
functions, such as getFreeRam or getLoad, with no problem as long as the 
call is hard-coded into the script. However, when I try to call said 
functions from a function monitoring keyboard input (using pyHooks), I get 
an error and the program crashes (at least it did until I put a try/except 
in there, but hitting the hotkey over and over always gives me an error). I 
will paste the entire file below. It is not very commented yet, so if 
something does not make sense, please let me know. My questions are:

1. Why am I getting these errors?

2. The getLoad function only returns one number (the counter is there to 
double check that the loop really only runs once). I have a dual core AMD, 
so I expected two numbers, a LoadPercentage for each core, not just one 
number. How do I get the load per core, not per physical processor?

3. Regarding pyHook: I press my win-` hotkey on the desktop and get 
something to happen in my script, which is good. However, I also get a 
Windows error sound. Is there a way to grab that keystroke before Windows 
can get it to avoid what seems to be passing the keystroke to the active app 
before my script can get it? I tried returning False in my keyUp function 
but that made no difference.

I am running Windows 7 Home Premium, 64 bit, on an AMD processor whose exact 
model escapes me. It is clocked at around 2.4ghz and I have 4gb of 800mhz 
ddr2 ram. The pc was built by my friend and I from Newegg parts, but has 
worked well for the last 18 months (seeing three different operating 
systems). Win7 was installed a few days ago, so I am pretty sure there is no 
problem with my installation as far as wmi goes. I tried to run the wmi 
diagnostic utility, but I got an error; I figure the utility cannot run in 
64 bit, but I am not sure. Anyway, any help here would be greatly 
appreciated. This is my first "real" project in python, so go easy on my 
stupid mistakes. :) Thanks!


CODE

import os, sys, wmi, win32com.client, pyHook, pythoncom

pythoncom.CoInitialize() #a forum post said I should put this...

win=0 #is either Windows key pressed?
ctrl=0 #either control key pressed?
alt=0 #what about either alt key?
#for speaking text with the user's active screenreader, or sapi if no reader
#http://www.empowermentzone.com/saysetup.exe
speaker=win32com.client.Dispatch("Say.Tools")
c=wmi.WMI("localhost") #for WMI access

def getLoad():
 i=0
 #get proc load, currently only first core for some reason
 for p in c.Win32_Processor():
  load=p.LoadPercentage
  i=i+1
 # end for
 speaker.say(str(load)+", looped "+str(i)+" times.")
 return(load)
#end def
#following works just fine here, but not when called later in keyUp method
getLoad()

def getFreeRam():
 #get total ram and ram available
 for info in c.Win32_OperatingSystem():
  ram=info.TotalVisibleMemorySize
  freeRam=info.FreePhysicalMemory
 #end for
 return(float(freeRam)/float(ram)*100) #return free ram as %
#end def

def toBiggestBytes(n):
 #returns a string where n is in the largest logical measure possible
 i=0 #counter
 units=[" bytes","kb","mb","gb","tb"]
 while(n>=1024):
  n=n/1024
  i=i+1
 return(str(n)+units[i])
#end def

#following does not work: why?
"""
for t in c.MSAcpi_ThermalZoneTemperature():
 temp=t.CurrentTemperature
# end for
"""

"""
#old interface used for early testing
end=0
while(end==0):
 lr=raw_input("\nr for ram, l for CPU load, e to exit.\n")
 if(lr=='r'):
  pRam=int(ram)/1024/1024
  speaker.say(str(pRam)+"mb total RAM")
 if(lr=='l'):
  speaker.say(str(load)+"% for core 1")
 if(lr=='e'):
  end=1 #exit loop
 #endif
#end while
"""


#Copied directly from example.py in the pyHook folder, then modified
def OnKeyDownEvent(event):
 global win
 global ctrl
 global alt
 if(event.KeyID==91 or event.KeyID==92):
  #user is pressing a Windows key, so set flag
  win=1
 #endif
 # return True to pass the event to other handlers
 # return False to stop the event from propagating
 return True
#end def

def OnKeyUpEvent(event):
 global win
 global ctrl
 global alt
 if(event.KeyID==91 or event.KeyID==92):
  #user released the Windows key, so unset flag
  win=0
 #endif
 if(win==1 and chr(event.Ascii)=='`'):
  try:
   speaker.say(str(getLoad()))
  except:
   speaker.say("Stupid errors!")
  #end except
 #endif
 return True
#end def

# create the hook mananger
hm = pyHook.HookManager()
# register two callbacks
hm.KeyDown = OnKeyDownEvent
hm.KeyUp = OnKeyUpEvent

# hook into the keyboard event
hm.HookKeyboard()

if __name__ == '__main__':
  import pythoncom
  pythoncom.PumpMessages()

#end copying/modifying

speaker=None; #release speaker object


#reference so I know what properties I can get
"""
  print 'MessageName:',event.MessageName
  print 'Message:',event.Message
  print 'Time:',event.Time
  print 'Window:',event.Window
  print 'WindowName:',event.WindowName
  print 'Ascii:', event.Ascii, chr(event.Ascii)
  print 'Key:', event.Key
  print 'KeyID:', event.KeyID
  print 'ScanCode:', event.ScanCode
  print 'Extended:', event.Extended
  print 'Injected:', event.Injected
  print 'Alt', event.Alt
  print 'Transition', event.Transition
  print '---'
"""

Have a great day,
Alex
Email: mehgcap at gmail.com 


From mail at timgolden.me.uk  Fri Jan 15 10:00:04 2010
From: mail at timgolden.me.uk (Tim Golden)
Date: Fri, 15 Jan 2010 09:00:04 +0000
Subject: [python-win32] WMI troubles!
In-Reply-To: <558CA385CC304B349F3FEB6D62A466DC@AlexPC>
References: <558CA385CC304B349F3FEB6D62A466DC@AlexPC>
Message-ID: <4B502E94.3010007@timgolden.me.uk>

On 15/01/2010 05:05, Alex Hall wrote:
> Hello all,
> I just joined this list. I am pretty new to Python, but I really like and
> hope to get more into it;

Welcome to the list. And to Python.

> Anyway, onto the problem. I am working on a simple resource monitor (I have
> never found one that works well with screen readers, so I am writing one). I
> am using python2.6 with all the win32 libs installed. My monitor will use
> wmi to get all of its information, and it was going pretty well until a few
> hours ago, when I started receiving seemingly random errors.

What would definitely help, and this is general advice for posting
to technical groups, is the exact error you received -- the traceback
which looks something like this:

Traceback (most recent call last):
   File "<stdin>", line 1, in <module>
   File "c:\work_in_progress\wmi\trunk\wmi.py", line 1124, in __getattr__
     return getattr (self._namespace, attribute)
   File "c:\python26\lib\site-packages\win32com\client\dynamic.py", line 512, in __getattr__
     raise AttributeError("%s.%s" % (self._username_, attr))
AttributeError: winmgmts:.Win32_Processx
>>>

In addition -- and I appreciate that this isn't always easy -- is
the smallest self-contained code example which will fail every time.
Apart from making it easier for people to help you out, it sometimes
makes the source of the error clearer even to you, the developer.

I don't have time now, but I will get to this later on today. The
likeliest cause is that PyHook is introducing a threading context
which is causing problems with the COM-based WMI stuff. Your early
CoInitialize is a nod in the right direction, but unfortunately
redundant where you've put it at the top of the main thread.

But sight of the traceback should help clarify things considerably.

TJG

From lyeoh at pop.jaring.my  Fri Jan 15 18:16:27 2010
From: lyeoh at pop.jaring.my (Lincoln Yeoh)
Date: Sat, 16 Jan 2010 01:16:27 +0800
Subject: [python-win32] Getting a list of "task buttons" or
 tasks/windows in the "alt-tab" list?
In-Reply-To: <201001131832.o0DIWJPw019549@vsmtp6.jaring.my>
References: <201001131832.o0DIWJPw019549@vsmtp6.jaring.my>
Message-ID: <201001151716.o0FHGWcO067512@vsmtp6.jaring.my>

At 02:32 AM 1/14/2010, Lincoln Yeoh wrote:

>If three Excel documents are opened, there are four top level Excel 
>windows (or at least they appear to be top level to me) - the main, 
>and three MS-SDIb ones. But the taskbar and alt-tab list only show 3 
>items. What logic should one use to create the same list?

OK I think it might be something to do with the "ITaskList_Deleted" 
window property that the main excel window sometimes has and 
sometimes doesn't (seems to depend on whether it has MS-SDIb windows or not).

Anyway, I'm now having difficulty checking for that window property 
using win32 python.

I can't seem to get user32.GetProp to work for me.

I tried:
rv=ctypes.windll.user32.GetProp(hwnd,'ITaskList_Deleted')

But I get:
   File "C:\Python26\lib\ctypes\__init__.py", line 366, in __getattr__
     func = self.__getitem__(name)
   File "C:\Python26\lib\ctypes\__init__.py", line 371, in __getitem__
     func = self._FuncPtr((name_or_ordinal, self))
AttributeError: function 'GetProp' not found

I tried:

rv=ctypes.windll.user32.GetPropA(hwnd,'ITaskList_Deleted')

But I get an ctypes.GetLastError()==2 after that.

Any tips?

Link.



From mehgcap at gmail.com  Fri Jan 15 18:58:18 2010
From: mehgcap at gmail.com (Alex Hall)
Date: Fri, 15 Jan 2010 12:58:18 -0500
Subject: [python-win32] WMI troubles!
In-Reply-To: <4B502E94.3010007@timgolden.me.uk>
References: <558CA385CC304B349F3FEB6D62A466DC@AlexPC>
	<4B502E94.3010007@timgolden.me.uk>
Message-ID: <F348C6022BCB405B976B22950AD7B6A2@AlexPC>

I am trying to put the stack trace into a file, since I cannot copy and 
paste from the cmd line output. Is there a way to do this? Also, if you run 
the file (assuming you have all the dependencies) the error happens as soon 
as you press windows-`. You can change "speaker.say" to "print" and comment 
out the line creating the speaker object so that you do not have to install 
sayTools.


Have a great day,
Alex
Email: mehgcap at gmail.com
----- Original Message ----- 
From: "Tim Golden" <mail at timgolden.me.uk>
Cc: <python-win32 at python.org>
Sent: Friday, January 15, 2010 4:00
Subject: Re: [python-win32] WMI troubles!


> On 15/01/2010 05:05, Alex Hall wrote:
>> Hello all,
>> I just joined this list. I am pretty new to Python, but I really like and
>> hope to get more into it;
>
> Welcome to the list. And to Python.
>
>> Anyway, onto the problem. I am working on a simple resource monitor (I 
>> have
>> never found one that works well with screen readers, so I am writing 
>> one). I
>> am using python2.6 with all the win32 libs installed. My monitor will use
>> wmi to get all of its information, and it was going pretty well until a 
>> few
>> hours ago, when I started receiving seemingly random errors.
>
> What would definitely help, and this is general advice for posting
> to technical groups, is the exact error you received -- the traceback
> which looks something like this:
>
> Traceback (most recent call last):
>   File "<stdin>", line 1, in <module>
>   File "c:\work_in_progress\wmi\trunk\wmi.py", line 1124, in __getattr__
>     return getattr (self._namespace, attribute)
>   File "c:\python26\lib\site-packages\win32com\client\dynamic.py", line 
> 512, in __getattr__
>     raise AttributeError("%s.%s" % (self._username_, attr))
> AttributeError: winmgmts:.Win32_Processx
>>>>
>
> In addition -- and I appreciate that this isn't always easy -- is
> the smallest self-contained code example which will fail every time.
> Apart from making it easier for people to help you out, it sometimes
> makes the source of the error clearer even to you, the developer.
>
> I don't have time now, but I will get to this later on today. The
> likeliest cause is that PyHook is introducing a threading context
> which is causing problems with the COM-based WMI stuff. Your early
> CoInitialize is a nod in the right direction, but unfortunately
> redundant where you've put it at the top of the main thread.
>
> But sight of the traceback should help clarify things considerably.
>
> TJG
> _______________________________________________
> python-win32 mailing list
> python-win32 at python.org
> http://mail.python.org/mailman/listinfo/python-win32 


From aahz at pythoncraft.com  Fri Jan 15 19:18:49 2010
From: aahz at pythoncraft.com (Aahz)
Date: Fri, 15 Jan 2010 10:18:49 -0800
Subject: [python-win32] WMI troubles!
In-Reply-To: <F348C6022BCB405B976B22950AD7B6A2@AlexPC>
References: <558CA385CC304B349F3FEB6D62A466DC@AlexPC>
	<4B502E94.3010007@timgolden.me.uk>
	<F348C6022BCB405B976B22950AD7B6A2@AlexPC>
Message-ID: <20100115181849.GA26279@panix.com>

On Fri, Jan 15, 2010, Alex Hall wrote:
>
> I am trying to put the stack trace into a file, since I cannot copy and  
> paste from the cmd line output. Is there a way to do this? Also, if you 
> run the file (assuming you have all the dependencies) the error happens 
> as soon as you press windows-`. You can change "speaker.say" to "print" 
> and comment out the line creating the speaker object so that you do not 
> have to install sayTools.

You probably will have difficulty doing this, but yes, you can
copy/paste command-line output using a mouse to select a screen area.

Actually, I just looked in my Windows virtual machine and you can do
alt-space, edit, select all, enter

You can also use try/except and the traceback module to capture an
exception and write it to a file.
-- 
Aahz (aahz at pythoncraft.com)           <*>         http://www.pythoncraft.com/

"If you think it's expensive to hire a professional to do the job, wait
until you hire an amateur."  --Red Adair

From mehgcap at gmail.com  Fri Jan 15 19:26:00 2010
From: mehgcap at gmail.com (Alex Hall)
Date: Fri, 15 Jan 2010 13:26:00 -0500
Subject: [python-win32] WMI troubles!
In-Reply-To: <20100115181849.GA26279@panix.com>
References: <558CA385CC304B349F3FEB6D62A466DC@AlexPC>
	<4B502E94.3010007@timgolden.me.uk>
	<F348C6022BCB405B976B22950AD7B6A2@AlexPC>
	<20100115181849.GA26279@panix.com>
Message-ID: <12E87B5F255043C084454902F42B8EAD@AlexPC>

What is the syntax to write the traceback to a text file? I have the file 
object and I can open and close the file, I just need the command, like 
"f.write(StackTrace())" or whatever the command is. I will try the alt-space 
idea as well.


Have a great day,
Alex
Email: mehgcap at gmail.com
----- Original Message ----- 
From: "Aahz" <aahz at pythoncraft.com>
To: "Alex Hall" <mehgcap at gmail.com>
Cc: <python-win32 at python.org>
Sent: Friday, January 15, 2010 13:18
Subject: Re: [python-win32] WMI troubles!


> On Fri, Jan 15, 2010, Alex Hall wrote:
>>
>> I am trying to put the stack trace into a file, since I cannot copy and
>> paste from the cmd line output. Is there a way to do this? Also, if you
>> run the file (assuming you have all the dependencies) the error happens
>> as soon as you press windows-`. You can change "speaker.say" to "print"
>> and comment out the line creating the speaker object so that you do not
>> have to install sayTools.
>
> You probably will have difficulty doing this, but yes, you can
> copy/paste command-line output using a mouse to select a screen area.
>
> Actually, I just looked in my Windows virtual machine and you can do
> alt-space, edit, select all, enter
>
> You can also use try/except and the traceback module to capture an
> exception and write it to a file.
> -- 
> Aahz (aahz at pythoncraft.com)           <*> 
> http://www.pythoncraft.com/
>
> "If you think it's expensive to hire a professional to do the job, wait
> until you hire an amateur."  --Red Adair 


From timr at probo.com  Fri Jan 15 19:33:26 2010
From: timr at probo.com (Tim Roberts)
Date: Fri, 15 Jan 2010 10:33:26 -0800
Subject: [python-win32] WMI troubles!
In-Reply-To: <558CA385CC304B349F3FEB6D62A466DC@AlexPC>
References: <558CA385CC304B349F3FEB6D62A466DC@AlexPC>
Message-ID: <4B50B4F6.1040704@probo.com>

Alex Hall wrote:
>
> Anyway, onto the problem. I am working on a simple resource monitor (I
> have never found one that works well with screen readers, so I am
> writing one). I am using python2.6 with all the win32 libs installed.
> My monitor will use wmi to get all of its information, and it was
> going pretty well until a few hours ago, when I started receiving
> seemingly random errors. I can call my functions, such as getFreeRam
> or getLoad, with no problem as long as the call is hard-coded into the
> script. However, when I try to call said functions from a function
> monitoring keyboard input (using pyHooks), I get an error and the
> program crashes (at least it did until I put a try/except in there,
> but hitting the hotkey over and over always gives me an error).

Windows hooks run in a unique environment.  When you install a Windows
hook, you are actually injecting a DLL into every process in the
system.  (It's not a low overhead operation!)  The code in the keyboard
hook runs as part of another processs.

Now, I have not looked into the pyHooks code to see if they are managing
that, by using some kind of inter-process communication.  If they are
not, then you are very limited in what you can do in your callback. 
"print" wouldn't work, because that process has a different standard
output.  You would need to send a signal back to your own process. 
Further, COM will not have been initialized in that other thread, so you
wouldn't be able to use a COM object, like "Say.Tools".  You can try
calling pythoncom.CoInitialize() in every callback, I suppose.


> I will paste the entire file below. It is not very commented yet, so
> if something does not make sense, please let me know. My questions are:
>
> 1. Why am I getting these errors?

What errors do you get?  I don't think you told us that.


> 2. The getLoad function only returns one number (the counter is there
> to double check that the loop really only runs once). I have a dual
> core AMD, so I expected two numbers, a LoadPercentage for each core,
> not just one number. How do I get the load per core, not per physical
> processor?

Well, it only returns one value because you overwrite the value of
"load" every time through the loop.  If you want to return multiple
values, you should add each new value to a list, and return the list.


> 3. Regarding pyHook: I press my win-` hotkey on the desktop and get
> something to happen in my script, which is good. However, I also get a
> Windows error sound. Is there a way to grab that keystroke before
> Windows can get it to avoid what seems to be passing the keystroke to
> the active app before my script can get it? I tried returning False in
> my keyUp function but that made no difference.

This shouldn't be an issue.  The hotkeys are handled by Explorer, which
sucks up the key after it launches your program.  I don't understand who
would trigger the sound.

-- 
Tim Roberts, timr at probo.com
Providenza & Boekelheide, Inc.


From timr at probo.com  Fri Jan 15 19:35:58 2010
From: timr at probo.com (Tim Roberts)
Date: Fri, 15 Jan 2010 10:35:58 -0800
Subject: [python-win32] WMI troubles!
In-Reply-To: <F348C6022BCB405B976B22950AD7B6A2@AlexPC>
References: <558CA385CC304B349F3FEB6D62A466DC@AlexPC>	<4B502E94.3010007@timgolden.me.uk>
	<F348C6022BCB405B976B22950AD7B6A2@AlexPC>
Message-ID: <4B50B58E.1020703@probo.com>

Alex Hall wrote:
> I am trying to put the stack trace into a file, since I cannot copy
> and paste from the cmd line output. 

Sure you can.  Alt-Space, Edit, Select-All.  Then press "Enter" to copy
the entire contents of the shell window to the clipboard.  If you were
sighted, I would suggest Alt-Space, Edit, Mark; then you can use your
mouse to draw a box around the text you want to clip, and again press
"Enter" to copy that region to the clipboard.

Alternatively, there are a couple of very handy tools called "gclip" and
"pclip" that are part of the "unxutils" collection on SourceForge. 
Anything piped into gclip is copied to the clipboard, and pclip prints
the clipboard to its standard output.

-- 
Tim Roberts, timr at probo.com
Providenza & Boekelheide, Inc.


From mehgcap at gmail.com  Fri Jan 15 19:39:49 2010
From: mehgcap at gmail.com (Alex Hall)
Date: Fri, 15 Jan 2010 13:39:49 -0500
Subject: [python-win32] WMI troubles!
In-Reply-To: <20100115181849.GA26279@panix.com>
References: <558CA385CC304B349F3FEB6D62A466DC@AlexPC>
	<4B502E94.3010007@timgolden.me.uk>
	<F348C6022BCB405B976B22950AD7B6A2@AlexPC>
	<20100115181849.GA26279@panix.com>
Message-ID: <29D27DFCDBCD4D4CBE97D79D652EFFBE@AlexPC>

Let's try this. Here is the stack trace when I ran my program. It ran fine, 
even calling the getLoad function once since that call is built into the 
code, not dependent on a keypress. This trace printed after I pressed win-`, 
the keypress I have assigned, through pyHook, to trigger a call to getLoad. 
If you want, you can change the call to getLoad to getFreeRam and a similar 
error occurs, pasted below this one.

Error from keypress calling getLoad():
C:\Users\Alex>python c:\prog\python\sys.py
Traceback (most recent call last):
  File "c:\python26\lib\site-packages\pyHook\HookManager.py", line 351, in 
Keybo
ardSwitch
    return func(event)
  File "c:\prog\python\sys.py", line 88, in OnKeyUpEvent
    speaker.say(str(getLoad()))
  File "c:\prog\python\sys.py", line 14, in getLoad
    for p in c.Win32_Processor():
  File "c:\python26\lib\site-packages\wmi.py", line 647, in query
    return self._namespace.query (wql, self, fields)
  File "c:\python26\lib\site-packages\wmi.py", line 835, in query
    return [ _wmi_object (obj, instance_of, fields) for obj in 
self._raw_query(w
ql) ]
  File "c:\python26\lib\site-packages\wmi.py", line 829, in _raw_query
    raise WMI_EXCEPTIONS.get (hresult, x_wmi (hresult))
wmi.x_wmi: -2147352567

Error from keypress calling getFreeRam():
C:\Users\Alex>python c:\prog\python\sys.py
Traceback (most recent call last):
  File "c:\python26\lib\site-packages\pyHook\HookManager.py", line 351, in 
Keybo
ardSwitch
    return func(event)
  File "c:\prog\python\sys.py", line 88, in OnKeyUpEvent
    speaker.say(str(getFreeRam()))
  File "c:\prog\python\sys.py", line 24, in getFreeRam
    for info in c.Win32_OperatingSystem():
  File "c:\python26\lib\site-packages\wmi.py", line 975, in __getattr__
    return getattr (self._namespace, attribute)
  File "c:\python26\lib\site-packages\win32com\client\dynamic.py", line 512, 
in
__getattr__
    raise AttributeError("%s.%s" % (self._username_, attr))
AttributeError: 
winmgmts:{impersonationLevel=impersonate,authenticationLevel=def
ault}//localhost/.Win32_OperatingSystem


Have a great day,
Alex
Email: mehgcap at gmail.com
----- Original Message ----- 
From: "Aahz" <aahz at pythoncraft.com>
To: "Alex Hall" <mehgcap at gmail.com>
Cc: <python-win32 at python.org>
Sent: Friday, January 15, 2010 13:18
Subject: Re: [python-win32] WMI troubles!


> On Fri, Jan 15, 2010, Alex Hall wrote:
>>
>> I am trying to put the stack trace into a file, since I cannot copy and
>> paste from the cmd line output. Is there a way to do this? Also, if you
>> run the file (assuming you have all the dependencies) the error happens
>> as soon as you press windows-`. You can change "speaker.say" to "print"
>> and comment out the line creating the speaker object so that you do not
>> have to install sayTools.
>
> You probably will have difficulty doing this, but yes, you can
> copy/paste command-line output using a mouse to select a screen area.
>
> Actually, I just looked in my Windows virtual machine and you can do
> alt-space, edit, select all, enter
>
> You can also use try/except and the traceback module to capture an
> exception and write it to a file.
> -- 
> Aahz (aahz at pythoncraft.com)           <*> 
> http://www.pythoncraft.com/
>
> "If you think it's expensive to hire a professional to do the job, wait
> until you hire an amateur."  --Red Adair 


From mehgcap at gmail.com  Fri Jan 15 19:49:30 2010
From: mehgcap at gmail.com (Alex Hall)
Date: Fri, 15 Jan 2010 13:49:30 -0500
Subject: [python-win32] WMI troubles!
In-Reply-To: <4B50B4F6.1040704@probo.com>
References: <558CA385CC304B349F3FEB6D62A466DC@AlexPC>
	<4B50B4F6.1040704@probo.com>
Message-ID: <3164DF6271584B1098783412B56698DB@AlexPC>

Regarding question 2, about only receiving one number for two cores, I have 
a counter in the loop which gets incremented each time the loop runs. This 
counter always ends up at 1, not 2, so it is only incremented once. When I 
first started this last week I did put any returns into a list, but the list 
was always just one element long. This seems to tell me that I am only 
getting one number, but I should be getting two. Is there a syntax thing 
that I am missing or confused about, or is wmi refusing me the information 
about my other core? Is there a different class I should be using to get the 
load per core?

As far as pyHook, it sounds like it is not going to work too well. Are you 
saying it could be the cause of the wmi troubles? If so, is there another 
library I should use instead? I want this to be a global thing, so no matter 
where in Windows you are, you need only press a hotkey to hear the status of 
your hardware, like a sighted person need only glance at their resource 
monitor gadget/sidebar app. For that reason I avoided pyGame and wx since, 
as I understand it, they only monitor keypresses in their windows, not 
through the entire system.

Finally, when you say "calling pythoncom.CoInitialize() in every callback", 
what do you mean by 'callback'?


Have a great day,
Alex
Email: mehgcap at gmail.com
----- Original Message ----- 
From: "Tim Roberts" <timr at probo.com>
To: "Python-Win32 List" <python-win32 at python.org>
Sent: Friday, January 15, 2010 13:33
Subject: Re: [python-win32] WMI troubles!


> Alex Hall wrote:
>>
>> Anyway, onto the problem. I am working on a simple resource monitor (I
>> have never found one that works well with screen readers, so I am
>> writing one). I am using python2.6 with all the win32 libs installed.
>> My monitor will use wmi to get all of its information, and it was
>> going pretty well until a few hours ago, when I started receiving
>> seemingly random errors. I can call my functions, such as getFreeRam
>> or getLoad, with no problem as long as the call is hard-coded into the
>> script. However, when I try to call said functions from a function
>> monitoring keyboard input (using pyHooks), I get an error and the
>> program crashes (at least it did until I put a try/except in there,
>> but hitting the hotkey over and over always gives me an error).
>
> Windows hooks run in a unique environment.  When you install a Windows
> hook, you are actually injecting a DLL into every process in the
> system.  (It's not a low overhead operation!)  The code in the keyboard
> hook runs as part of another processs.
>
> Now, I have not looked into the pyHooks code to see if they are managing
> that, by using some kind of inter-process communication.  If they are
> not, then you are very limited in what you can do in your callback.
> "print" wouldn't work, because that process has a different standard
> output.  You would need to send a signal back to your own process.
> Further, COM will not have been initialized in that other thread, so you
> wouldn't be able to use a COM object, like "Say.Tools".  You can try
> calling pythoncom.CoInitialize() in every callback, I suppose.
>
>
>> I will paste the entire file below. It is not very commented yet, so
>> if something does not make sense, please let me know. My questions are:
>>
>> 1. Why am I getting these errors?
>
> What errors do you get?  I don't think you told us that.
>
>
>> 2. The getLoad function only returns one number (the counter is there
>> to double check that the loop really only runs once). I have a dual
>> core AMD, so I expected two numbers, a LoadPercentage for each core,
>> not just one number. How do I get the load per core, not per physical
>> processor?
>
> Well, it only returns one value because you overwrite the value of
> "load" every time through the loop.  If you want to return multiple
> values, you should add each new value to a list, and return the list.
>
>
>> 3. Regarding pyHook: I press my win-` hotkey on the desktop and get
>> something to happen in my script, which is good. However, I also get a
>> Windows error sound. Is there a way to grab that keystroke before
>> Windows can get it to avoid what seems to be passing the keystroke to
>> the active app before my script can get it? I tried returning False in
>> my keyUp function but that made no difference.
>
> This shouldn't be an issue.  The hotkeys are handled by Explorer, which
> sucks up the key after it launches your program.  I don't understand who
> would trigger the sound.
>
> -- 
> Tim Roberts, timr at probo.com
> Providenza & Boekelheide, Inc.
>
> _______________________________________________
> python-win32 mailing list
> python-win32 at python.org
> http://mail.python.org/mailman/listinfo/python-win32 


From timr at probo.com  Fri Jan 15 20:20:41 2010
From: timr at probo.com (Tim Roberts)
Date: Fri, 15 Jan 2010 11:20:41 -0800
Subject: [python-win32] WMI troubles!
In-Reply-To: <3164DF6271584B1098783412B56698DB@AlexPC>
References: <558CA385CC304B349F3FEB6D62A466DC@AlexPC>
	<4B50B4F6.1040704@probo.com>
	<3164DF6271584B1098783412B56698DB@AlexPC>
Message-ID: <4B50C009.10300@probo.com>

Alex Hall wrote:
> Regarding question 2, about only receiving one number for two cores, I
> have a counter in the loop which gets incremented each time the loop
> runs. This counter always ends up at 1, not 2, so it is only
> incremented once. When I first started this last week I did put any
> returns into a list, but the list was always just one element long.
> This seems to tell me that I am only getting one number, but I should
> be getting two. Is there a syntax thing that I am missing or confused
> about, or is wmi refusing me the information about my other core? Is
> there a different class I should be using to get the load per core?

The number you are getting is the average load for the physical
processor, which includes all of the cores and hyperthreaded modules. 
If you have a dual core machine, you'll find that c.NumberOfCores is 2. 
Check this MSDN article:
    http://msdn.microsoft.com/en-us/library/aa394317.aspx

It describes the WMI Win32_PerfRawData_PerfOS_Processor class, which
includes information per logical processor, and describes how to figure
that out from the Win32_Processor information.


> As far as pyHook, it sounds like it is not going to work too well. Are
> you saying it could be the cause of the wmi troubles?

Well, you didn't say what the actual error was until just a few minutes
ago, so I didn't know the troubles were in WMI.  I assumed you were
having troubles reaching the "Say" class or doing a "print".  It's
possible that some of the processes you are hooking do not have
permission to access WMI.


> If so, is there another library I should use instead? I want this to
> be a global thing, so no matter where in Windows you are, you need
> only press a hotkey to hear the status of your hardware, like a
> sighted person need only glance at their resource monitor
> gadget/sidebar app. For that reason I avoided pyGame and wx since, as
> I understand it, they only monitor keypresses in their windows, not
> through the entire system.

In that case, why use a hook at all?  Why not just have a single script
that fetches and says the resource usage and then exits?  You could have
that tied to an Explorer hotkey, just like you do now.  Windows hooks
are a very heavyweight technology.  Your hook code gets executed for
every keystroke and every mouse motion.


> Finally, when you say "calling pythoncom.CoInitialize() in every
> callback", what do you mean by 'callback'?

Your OnKeyDown and OnKeyUp functions are being called in response to
external events.  You pass those functions to pyHooks, and it "calls you
back" when something happens.  Functions that you provide for someone
else to call are called "callbacks".

-- 
Tim Roberts, timr at probo.com
Providenza & Boekelheide, Inc.


From mehgcap at gmail.com  Fri Jan 15 20:38:05 2010
From: mehgcap at gmail.com (Alex Hall)
Date: Fri, 15 Jan 2010 14:38:05 -0500
Subject: [python-win32] WMI troubles!
In-Reply-To: <4B50C009.10300@probo.com>
References: <558CA385CC304B349F3FEB6D62A466DC@AlexPC><4B50B4F6.1040704@probo.com><3164DF6271584B1098783412B56698DB@AlexPC>
	<4B50C009.10300@probo.com>
Message-ID: <D2A9CA10622C4A43AEB049F3EF5AAD2A@AlexPC>

Thanks, I will look at the article!

I do not want a single script because, eventually, the plan is to have a 
keystroke for core 1, a keystroke for core 2, a keystroke for percent of ram 
used, a keystroke for hard disk space, and so on. While I could write each 
function in its own script and put a hotkey on each one, it seems easier and 
faster to just make one script and run it once, avoiding the delay I always 
get when this script runs for the first time from happening each time I 
press a hotkey.

Sorry for the lack of detail about the errors. I knew they were wmi errors, 
so when I read back through the messages I wrote it made sense to me, but of 
course I should have said what kind of errors were being generated.


Have a great day,
Alex
Email: mehgcap at gmail.com
----- Original Message ----- 
From: "Tim Roberts" <timr at probo.com>
To: "Python-Win32 List" <python-win32 at python.org>
Sent: Friday, January 15, 2010 14:20
Subject: Re: [python-win32] WMI troubles!


> Alex Hall wrote:
>> Regarding question 2, about only receiving one number for two cores, I
>> have a counter in the loop which gets incremented each time the loop
>> runs. This counter always ends up at 1, not 2, so it is only
>> incremented once. When I first started this last week I did put any
>> returns into a list, but the list was always just one element long.
>> This seems to tell me that I am only getting one number, but I should
>> be getting two. Is there a syntax thing that I am missing or confused
>> about, or is wmi refusing me the information about my other core? Is
>> there a different class I should be using to get the load per core?
>
> The number you are getting is the average load for the physical
> processor, which includes all of the cores and hyperthreaded modules.
> If you have a dual core machine, you'll find that c.NumberOfCores is 2.
> Check this MSDN article:
>    http://msdn.microsoft.com/en-us/library/aa394317.aspx
>
> It describes the WMI Win32_PerfRawData_PerfOS_Processor class, which
> includes information per logical processor, and describes how to figure
> that out from the Win32_Processor information.
>
>
>> As far as pyHook, it sounds like it is not going to work too well. Are
>> you saying it could be the cause of the wmi troubles?
>
> Well, you didn't say what the actual error was until just a few minutes
> ago, so I didn't know the troubles were in WMI.  I assumed you were
> having troubles reaching the "Say" class or doing a "print".  It's
> possible that some of the processes you are hooking do not have
> permission to access WMI.
>
>
>> If so, is there another library I should use instead? I want this to
>> be a global thing, so no matter where in Windows you are, you need
>> only press a hotkey to hear the status of your hardware, like a
>> sighted person need only glance at their resource monitor
>> gadget/sidebar app. For that reason I avoided pyGame and wx since, as
>> I understand it, they only monitor keypresses in their windows, not
>> through the entire system.
>
> In that case, why use a hook at all?  Why not just have a single script
> that fetches and says the resource usage and then exits?  You could have
> that tied to an Explorer hotkey, just like you do now.  Windows hooks
> are a very heavyweight technology.  Your hook code gets executed for
> every keystroke and every mouse motion.
>
>
>> Finally, when you say "calling pythoncom.CoInitialize() in every
>> callback", what do you mean by 'callback'?
>
> Your OnKeyDown and OnKeyUp functions are being called in response to
> external events.  You pass those functions to pyHooks, and it "calls you
> back" when something happens.  Functions that you provide for someone
> else to call are called "callbacks".
>
> -- 
> Tim Roberts, timr at probo.com
> Providenza & Boekelheide, Inc.
>
> _______________________________________________
> python-win32 mailing list
> python-win32 at python.org
> http://mail.python.org/mailman/listinfo/python-win32 


From timr at probo.com  Fri Jan 15 20:47:23 2010
From: timr at probo.com (Tim Roberts)
Date: Fri, 15 Jan 2010 11:47:23 -0800
Subject: [python-win32] WMI troubles!
In-Reply-To: <D2A9CA10622C4A43AEB049F3EF5AAD2A@AlexPC>
References: <558CA385CC304B349F3FEB6D62A466DC@AlexPC><4B50B4F6.1040704@probo.com><3164DF6271584B1098783412B56698DB@AlexPC>
	<4B50C009.10300@probo.com>
	<D2A9CA10622C4A43AEB049F3EF5AAD2A@AlexPC>
Message-ID: <4B50C64B.7070803@probo.com>

Alex Hall wrote:
>
> I do not want a single script because, eventually, the plan is to have
> a keystroke for core 1, a keystroke for core 2, a keystroke for
> percent of ram used, a keystroke for hard disk space, and so on. While
> I could write each function in its own script and put a hotkey on each
> one, it seems easier and faster to just make one script and run it
> once, avoiding the delay I always get when this script runs for the
> first time from happening each time I press a hotkey.

I guess that depends on how you define "easier" and "faster".  For
development time, it's certainly going to be faster to create a set of
small, self-contained, single-purpose scripts that start, run, and end,
than it is to figure out the twisty paths involved in Windows hooks. 
This is very much like the trade-off between writing CGI scripts and
writing a full-blown web application.  It's a lot quicker to get a CGI
script up and running.

Yes, there is a short delay when starting a Python app, but after the
interpreter has been loaded once, it should stay in the file cache, so
subsequent loads go faster.

-- 
Tim Roberts, timr at probo.com
Providenza & Boekelheide, Inc.


From mail at timgolden.me.uk  Fri Jan 15 21:19:37 2010
From: mail at timgolden.me.uk (Tim Golden)
Date: Fri, 15 Jan 2010 20:19:37 +0000
Subject: [python-win32] WMI troubles!
In-Reply-To: <12E87B5F255043C084454902F42B8EAD@AlexPC>
References: <558CA385CC304B349F3FEB6D62A466DC@AlexPC>	<4B502E94.3010007@timgolden.me.uk>	<F348C6022BCB405B976B22950AD7B6A2@AlexPC>	<20100115181849.GA26279@panix.com>
	<12E87B5F255043C084454902F42B8EAD@AlexPC>
Message-ID: <4B50CDD9.109@timgolden.me.uk>

On 15/01/2010 18:26, Alex Hall wrote:
> What is the syntax to write the traceback to a text file? I have the
> file object and I can open and close the file, I just need the command,
> like "f.write(StackTrace())" or whatever the command is. I will try the
> alt-space idea as well.

Just to answer this question on its own for the moment:
the following code is the kind of thing you can use
to trap an error and write the traceback to a file.
(You could also use the stdlib logging module):

<code>
import traceback

try:
   1 / 0 # or something more useful
except:
   with open ("error.log", "a") as f:
     traceback.print_exc (file=f)
     # or f.write (traceback.format_exc ())

</code>

TJG

From mail at timgolden.me.uk  Fri Jan 15 21:31:42 2010
From: mail at timgolden.me.uk (Tim Golden)
Date: Fri, 15 Jan 2010 20:31:42 +0000
Subject: [python-win32] WMI troubles!
In-Reply-To: <D2A9CA10622C4A43AEB049F3EF5AAD2A@AlexPC>
References: <558CA385CC304B349F3FEB6D62A466DC@AlexPC><4B50B4F6.1040704@probo.com><3164DF6271584B1098783412B56698DB@AlexPC>	<4B50C009.10300@probo.com>
	<D2A9CA10622C4A43AEB049F3EF5AAD2A@AlexPC>
Message-ID: <4B50D0AE.1060509@timgolden.me.uk>

On 15/01/2010 19:38, Alex Hall wrote:
> I do not want a single script because, eventually, the plan is to have a
> keystroke for core 1, a keystroke for core 2, a keystroke for percent of
> ram used, a keystroke for hard disk space, and so on. While I could
> write each function in its own script and put a hotkey on each one, it
> seems easier and faster to just make one script and run it once,
> avoiding the delay I always get when this script runs for the first time
> from happening each time I press a hotkey.

I've got slightly lost in this thread (not helped, I'm
afraid Alex, by your habit of top-posting ;) ) but in
an effort to be helpful, can I suggest the following.
Instead of fiddling with pyHooks which are -- as Tim R
points out -- somewhat dark and messy, consider adapting
the following example instead:

http://timgolden.me.uk/python/win32_how_do_i/catch_system_wide_hotkeys.html

This uses a straightforward Windows message loop to
provide to sort of global hotkeys I think you're after.
You might, for example, change the handle_win_f3
function to start a wmi instance (or use an existing
one) and then produce the output you were after.

TJG

From mehgcap at gmail.com  Fri Jan 15 21:57:28 2010
From: mehgcap at gmail.com (Alex Hall)
Date: Fri, 15 Jan 2010 15:57:28 -0500
Subject: [python-win32] WMI troubles!
In-Reply-To: <4B50D0AE.1060509@timgolden.me.uk>
References: <558CA385CC304B349F3FEB6D62A466DC@AlexPC><4B50B4F6.1040704@probo.com><3164DF6271584B1098783412B56698DB@AlexPC>	<4B50C009.10300@probo.com><D2A9CA10622C4A43AEB049F3EF5AAD2A@AlexPC>
	<4B50D0AE.1060509@timgolden.me.uk>
Message-ID: <C21516E2180744928E01574826968AD9@AlexPC>

Subject: Re: [python-win32] WMI troubles!


> On 15/01/2010 19:38, Alex Hall wrote:
>> I do not want a single script because, eventually, the plan is to have a
>> keystroke for core 1, a keystroke for core 2, a keystroke for percent of
>> ram used, a keystroke for hard disk space, and so on. While I could
>> write each function in its own script and put a hotkey on each one, it
>> seems easier and faster to just make one script and run it once,
>> avoiding the delay I always get when this script runs for the first time
>> from happening each time I press a hotkey.
>
> I've got slightly lost in this thread (not helped, I'm
> afraid Alex, by your habit of top-posting ;) ) but in
> an effort to be helpful, can I suggest the following.
> Instead of fiddling with pyHooks which are -- as Tim R
> points out -- somewhat dark and messy, consider adapting
> the following example instead:
>
> http://timgolden.me.uk/python/win32_how_do_i/catch_system_wide_hotkeys.html
>
> This uses a straightforward Windows message loop to
> provide to sort of global hotkeys I think you're after.
> You might, for example, change the handle_win_f3
> function to start a wmi instance (or use an existing
> one) and then produce the output you were after.

Thanks!! I "top post" since, when you use a screen reader, it is far easier 
to read a response straight away than it is to wade through a bunch of text 
you have already read, complete with many repetitions of "greater than". I 
will try to make in-line comments.
Anyway, I have copied the example you provided and am about to go modify it. 
If it works, it should help a lot! I just hope it will clear up the wmi 
errors I keep getting... Thanks for the link!
>
> TJG
> _______________________________________________
> python-win32 mailing list
> python-win32 at python.org
> http://mail.python.org/mailman/listinfo/python-win32 


From mehgcap at gmail.com  Sat Jan 16 01:26:17 2010
From: mehgcap at gmail.com (Alex Hall)
Date: Fri, 15 Jan 2010 19:26:17 -0500
Subject: [python-win32] syntax?
Message-ID: <5DA9BA03F5534642BBA4F69CAA91FB7B@AlexPC>

Hi all
I am getting a "syntax error" on the line
def sayUsedRam():
and I am not sure why. Nothing is wrong with that line, but I was playing 
with the preceeding function when this started happening. Does anyone see 
anything obvious about these two functions that would cause an error? I do 
not always trust errors with line numbers to give me the right line number. 
Thanks in advance for any help.

def sayLoad():
 i=0
 #get proc load, currently only first core for some reason
 load=[]
 for p in c.Win32_PerfFormattedData_PerfOS_Processor():
  load.push(p.PercentProcessorTime)
  #speaker.say(str(load))
  i=i+1
 #end for
 speaker.say(str(load[0])+", "+str(load[1])
#end def

def sayUsedRam():
 #get total ram and ram available
 #then speak used ram as a percent, rounded to 1 place
 for info in c.Win32_OperatingSystem():
  ram=info.TotalVisibleMemorySize
  freeRam=info.FreePhysicalMemory
 #end for
 #speaker.say(toBiggestBytes(freeRam,2)+". ")
 used=100-(float(freeRam)/float(ram)*100)
 speaker.say(str(round(used,1))+"% ram used.")
#end def


Have a great day,
Alex
Email: mehgcap at gmail.com 


From aahz at pythoncraft.com  Sat Jan 16 01:55:35 2010
From: aahz at pythoncraft.com (Aahz)
Date: Fri, 15 Jan 2010 16:55:35 -0800
Subject: [python-win32] syntax?
In-Reply-To: <5DA9BA03F5534642BBA4F69CAA91FB7B@AlexPC>
References: <5DA9BA03F5534642BBA4F69CAA91FB7B@AlexPC>
Message-ID: <20100116005535.GA6401@panix.com>

On Fri, Jan 15, 2010, Alex Hall wrote:
>
> I am getting a "syntax error" on the line
> def sayUsedRam():
> and I am not sure why. Nothing is wrong with that line, but I was playing 
> with the preceeding function when this started happening. Does anyone see 
> anything obvious about these two functions that would cause an error? I 
> do not always trust errors with line numbers to give me the right line 
> number. 

You probably have an indentation error, but I can't tell because the code
from your message had completely messed up indentation.
-- 
Aahz (aahz at pythoncraft.com)           <*>         http://www.pythoncraft.com/

"If you think it's expensive to hire a professional to do the job, wait
until you hire an amateur."  --Red Adair

From mehgcap at gmail.com  Sat Jan 16 02:53:28 2010
From: mehgcap at gmail.com (Alex Hall)
Date: Fri, 15 Jan 2010 20:53:28 -0500
Subject: [python-win32] letter and number key codes for win32con
Message-ID: <B9164389F21948CC8A32F2BE60155EB5@AlexPC>

Hello all,
I am successfully using hotkeys with win32con (thanks so much for that 
tip!!) but I want to use the numbers 1-0 (on the keyboard, not the numpad) 
instead of F keys. The code example I have says to use ord('x'), but I am 
not quite sure what this means or how to give the result of ord() to the 
hotkey dictionary. Does anyone know the codes to use? I have tried using 
win32con.49, win32con.VK_49, and all kinds of variants (using 1 instead of 
49, using ord() right in the dictionary, using single quotes...) Google is 
not much help on this; the only forum I found talking about this said that 
it could not be done.


Have a great day,
Alex
Email: mehgcap at gmail.com 


From timr at probo.com  Sat Jan 16 03:51:20 2010
From: timr at probo.com (Tim Roberts)
Date: Fri, 15 Jan 2010 18:51:20 -0800
Subject: [python-win32] letter and number key codes for win32con
In-Reply-To: <B9164389F21948CC8A32F2BE60155EB5@AlexPC>
References: <B9164389F21948CC8A32F2BE60155EB5@AlexPC>
Message-ID: <4B5129A8.1080108@probo.com>

Alex Hall wrote:
>
> I am successfully using hotkeys with win32con (thanks so much for that
> tip!!) but I want to use the numbers 1-0 (on the keyboard, not the
> numpad) instead of F keys. The code example I have says to use
> ord('x'), but I am not quite sure what this means or how to give the
> result of ord() to the hotkey dictionary. Does anyone know the codes
> to use? I have tried using win32con.49, win32con.VK_49, and all kinds
> of variants (using 1 instead of 49, using ord() right in the
> dictionary, using single quotes...) Google is not much help on this;
> the only forum I found talking about this said that it could not be done.

"ord" is a standard Python function that returns the numeric ASCII value
of a character.  The hotkey code for the 1 key is the ASCII value for
the character '1', which is the plain, ordinary integer 49.  You should
be able to use either ord('1') or 49.

-- 
Tim Roberts, timr at probo.com
Providenza & Boekelheide, Inc.


From mehgcap at gmail.com  Sat Jan 16 04:03:52 2010
From: mehgcap at gmail.com (Alex Hall)
Date: Fri, 15 Jan 2010 22:03:52 -0500
Subject: [python-win32] letter and number key codes for win32con
In-Reply-To: <4B5129A8.1080108@probo.com>
References: <B9164389F21948CC8A32F2BE60155EB5@AlexPC>
	<4B5129A8.1080108@probo.com>
Message-ID: <B6BD24B41F6D455FB585FAA67E6F6466@AlexPC>

Subject: Re: [python-win32] letter and number key codes for win32con


> Alex Hall wrote:
>>
>> I am successfully using hotkeys with win32con (thanks so much for that
>> tip!!) but I want to use the numbers 1-0 (on the keyboard, not the
>> numpad) instead of F keys. The code example I have says to use
>> ord('x'), but I am not quite sure what this means or how to give the
>> result of ord() to the hotkey dictionary. Does anyone know the codes
>> to use? I have tried using win32con.49, win32con.VK_49, and all kinds
>> of variants (using 1 instead of 49, using ord() right in the
>> dictionary, using single quotes...) Google is not much help on this;
>> the only forum I found talking about this said that it could not be done.
>
> "ord" is a standard Python function that returns the numeric ASCII value
> of a character.  The hotkey code for the 1 key is the ASCII value for
> the character '1', which is the plain, ordinary integer 49.  You should
> be able to use either ord('1') or 49.

Okay, the problem was that I was leaving on the "win32con." part, using 
"win32con.49" instead of just 49. It worked after I took off that win32con 
bit.
> -- 
> Tim Roberts, timr at probo.com
> Providenza & Boekelheide, Inc.
>
> _______________________________________________
> python-win32 mailing list
> python-win32 at python.org
> http://mail.python.org/mailman/listinfo/python-win32 


From mail at timgolden.me.uk  Sat Jan 16 08:07:02 2010
From: mail at timgolden.me.uk (Tim Golden)
Date: Sat, 16 Jan 2010 07:07:02 +0000
Subject: [python-win32] WMI troubles!
In-Reply-To: <C21516E2180744928E01574826968AD9@AlexPC>
References: <558CA385CC304B349F3FEB6D62A466DC@AlexPC><4B50B4F6.1040704@probo.com><3164DF6271584B1098783412B56698DB@AlexPC>	<4B50C009.10300@probo.com><D2A9CA10622C4A43AEB049F3EF5AAD2A@AlexPC>
	<4B50D0AE.1060509@timgolden.me.uk>
	<C21516E2180744928E01574826968AD9@AlexPC>
Message-ID: <4B516596.8040303@timgolden.me.uk>

On 15/01/2010 20:57, Alex Hall wrote:
 > Thanks!! I "top post" since, when you use a screen reader, it is far
 > easier to read a response straight away than it is to wade through a
 > bunch of text you have already read, complete with many repetitions of
 > "greater than".

I thought of that a few seconds after I sent the comment :)

The main problem for me was that I was getting emails
out of order (which happens often enough to be unsurprising).
That, combined with the rapid series of exchanges between you
and Tim Roberts all with a lot of text meant that I rather lost track.

 > Anyway, I have copied the example you provided and am about to go modify
 > it. If it works, it should help a lot! I just hope it will clear up the
 > wmi errors I keep getting... Thanks for the link!

Hope it helps

TJG

From narisipalli at gmail.com  Sun Jan 17 15:32:26 2010
From: narisipalli at gmail.com (Gowtham)
Date: Sun, 17 Jan 2010 09:32:26 -0500
Subject: [python-win32] syntax?
In-Reply-To: <5DA9BA03F5534642BBA4F69CAA91FB7B@AlexPC>
References: <5DA9BA03F5534642BBA4F69CAA91FB7B@AlexPC>
Message-ID: <b3281c491001170632m2fdcb3der458c99a3173a4d1a@mail.gmail.com>

Hi Alex,

I understand your code but its very difficult to read in python if the
indentation is not clear. Run through the Python style guide -
http://www.python.org/dev/peps/pep-0008/ or Google's pyguide -
http://google-styleguide.googlecode.com/svn/trunk/pyguide.html

Typical indentation is 4 spaces.

Cheers.

On Fri, Jan 15, 2010 at 7:26 PM, Alex Hall <mehgcap at gmail.com> wrote:

> Hi all
> I am getting a "syntax error" on the line
> def sayUsedRam():
> and I am not sure why. Nothing is wrong with that line, but I was playing
> with the preceeding function when this started happening. Does anyone see
> anything obvious about these two functions that would cause an error? I do
> not always trust errors with line numbers to give me the right line number.
> Thanks in advance for any help.
>
> def sayLoad():
>
> i=0
> #get proc load, currently only first core for some reason
> load=[]
> for p in c.Win32_PerfFormattedData_PerfOS_Processor():
>  load.push(p.PercentProcessorTime)
>  #speaker.say(str(load))
>
>  i=i+1
> #end for
> speaker.say(str(load[0])+", "+str(load[1])
> #end def
>
> def sayUsedRam():
>
> #get total ram and ram available
> #then speak used ram as a percent, rounded to 1 place
>
> for info in c.Win32_OperatingSystem():
>  ram=info.TotalVisibleMemorySize
>  freeRam=info.FreePhysicalMemory
> #end for
> #speaker.say(toBiggestBytes(freeRam,2)+". ")
> used=100-(float(freeRam)/float(ram)*100)
> speaker.say(str(round(used,1))+"% ram used.")
> #end def
>
>
>
> Have a great day,
> Alex
> Email: mehgcap at gmail.com
> _______________________________________________
> python-win32 mailing list
> python-win32 at python.org
> http://mail.python.org/mailman/listinfo/python-win32
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-win32/attachments/20100117/8f4642d5/attachment.htm>

From kevin.horn at gmail.com  Mon Jan 18 17:21:16 2010
From: kevin.horn at gmail.com (Kevin Horn)
Date: Mon, 18 Jan 2010 10:21:16 -0600
Subject: [python-win32] syntax?
In-Reply-To: <5DA9BA03F5534642BBA4F69CAA91FB7B@AlexPC>
References: <5DA9BA03F5534642BBA4F69CAA91FB7B@AlexPC>
Message-ID: <562bcc11001180821h1724eb9ama6cad4bc0e304da8@mail.gmail.com>

On Fri, Jan 15, 2010 at 6:26 PM, Alex Hall <mehgcap at gmail.com> wrote:

> Hi all
> I am getting a "syntax error" on the line
> def sayUsedRam():
> and I am not sure why. Nothing is wrong with that line, but I was playing
> with the preceeding function when this started happening. Does anyone see
> anything obvious about these two functions that would cause an error? I do
> not always trust errors with line numbers to give me the right line number.
> Thanks in advance for any help.
>
> def sayLoad():
> i=0
> #get proc load, currently only first core for some reason
> load=[]
> for p in c.Win32_PerfFormattedData_PerfOS_Processor():
>  load.push(p.PercentProcessorTime)
>  #speaker.say(str(load))
>  i=i+1
> #end for
> speaker.say(str(load[0])+", "+str(load[1])
> #end def
>
> def sayUsedRam():
> #get total ram and ram available
> #then speak used ram as a percent, rounded to 1 place
> for info in c.Win32_OperatingSystem():
>  ram=info.TotalVisibleMemorySize
>  freeRam=info.FreePhysicalMemory
> #end for
> #speaker.say(toBiggestBytes(freeRam,2)+". ")
> used=100-(float(freeRam)/float(ram)*100)
> speaker.say(str(round(used,1))+"% ram used.")
> #end def
>
>
> Have a great day,
> Alex
> Email: mehgcap at gmail.com
>

You appear to be missing a close paren on this line:

speaker.say(str(load[0])+", "+str(load[1])

in the sayLoad() function

Kevin Horn
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-win32/attachments/20100118/163933d4/attachment.htm>

From whateverwd at hotmail.com  Wed Jan 20 04:16:21 2010
From: whateverwd at hotmail.com (Wesley Overhults)
Date: Tue, 19 Jan 2010 21:16:21 -0600
Subject: [python-win32] pywin32 214 Problem
Message-ID: <SNT108-W408B6AA1D92598D590F16BD2640@phx.gbl>


So I have a new computer that runs Windows 7 and I have a program called Hellanzb that uses Python.  I installed Python 2.5 and I installed pywin-32 214 but when I try to use Hellanzb, it gives me this message.

 

"The program can't start because pywintypes25.dll is missing from your computer.  Try reinstalling the program to fix this problem."

 

I've reinstalled numerous times but nothing seems to work.  pywintypes25.dll is on my computer.  It's in the Python25 folder.  Where should it be to get this program to work?
 		 	   		  
_________________________________________________________________
Hotmail: Trusted email with powerful SPAM protection.
http://clk.atdmt.com/GBL/go/196390707/direct/01/
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-win32/attachments/20100119/b0e2c811/attachment.htm>

From metolone+gmane at gmail.com  Wed Jan 20 09:02:44 2010
From: metolone+gmane at gmail.com (Mark Tolonen)
Date: Wed, 20 Jan 2010 00:02:44 -0800
Subject: [python-win32] pywin32 214 Problem
References: <SNT108-W408B6AA1D92598D590F16BD2640@phx.gbl>
Message-ID: <hj6db1$1sm$1@ger.gmane.org>


"Wesley Overhults" <whateverwd at hotmail.com> wrote in message 
news:SNT108-W408B6AA1D92598D590F16BD2640 at phx.gbl...

> So I have a new computer that runs Windows 7 and I have a program called 
> Hellanzb that uses Python.  I installed Python 2.5 and I installed 
> pywin-32 214 but when I try to use Hellanzb, it gives me this message.

> "The program can't start because pywintypes25.dll is missing from your 
> computer.  Try reinstalling the program to fix this problem."

> I've reinstalled numerous times but nothing seems to work. 
> pywintypes25.dll is on my computer.  It's in the Python25 folder.  Where 
> should it be to get this program to work?

Sounds like a Windows manifest issue.  Try Process Monitor from SysInternals 
if you want to determine where your system is looking for and failing to 
find pywintypes25.dll.  It may give a clue what is wrong.

HTH,
Mark 



From skippy.hammond at gmail.com  Wed Jan 20 12:36:19 2010
From: skippy.hammond at gmail.com (Mark Hammond)
Date: Wed, 20 Jan 2010 22:36:19 +1100
Subject: [python-win32] pywin32 214 Problem
In-Reply-To: <SNT108-W408B6AA1D92598D590F16BD2640@phx.gbl>
References: <SNT108-W408B6AA1D92598D590F16BD2640@phx.gbl>
Message-ID: <4B56EAB3.2050702@gmail.com>

On 20/01/2010 2:16 PM, Wesley Overhults wrote:
> So I have a new computer that runs Windows 7 and I have a program called
> Hellanzb that uses Python. I installed Python 2.5 and I installed
> pywin-32 214 but when I try to use Hellanzb, it gives me this message.
>
> "The program can't start because pywintypes25.dll is missing from your
> computer. Try reinstalling the program to fix this problem."
>
> I've reinstalled numerous times but nothing seems to work.
> pywintypes25.dll is on my computer. It's in the Python25 folder. Where
> should it be to get this program to work?

I suspect you installed python "just for me" instead of for all users. 
try uninstalling python and reinstalling for all users, then reinstall 
pywin32.

Cheers,

Mark

From Olof.Rutgersson at xelerated.com  Wed Jan 20 15:57:53 2010
From: Olof.Rutgersson at xelerated.com (Olof Rutgersson)
Date: Wed, 20 Jan 2010 15:57:53 +0100
Subject: [python-win32] wexpect and running without a console
Message-ID: <987E17118F2F994D92C0EFD24784D10B1A34257840@xelwa4.xelerated.com>

Hello

I'm using wexpect to access various programs such as plink, cygwin and other command line tools in windows. The code works fine when the python interpreter is launched from a console window, but when I launch it through any other process (for instance, a scheduler for automation of certain tasks) I get an exception. The spawn call for creating a child process to the python process works (and really does spawn the child process), but as soon as I do an expect() call the following exception occurs:

   File "...\expect\wexpect.py", line 1330, in expect
     return self.expect_list(compiled_pattern_list, timeout, searchwindowsize)
   File "...\expect\wexpect.py", line 1344, in expect_list
     return self.expect_loop(searcher_re(pattern_list), timeout, searchwindowsize)
   File "...\expect\wexpect.py", line 1397, in expect_loop
     c = self.read_nonblocking (self.maxread, timeout)
   File "...\expect\wexpect.py", line 1771, in read_nonblocking
     s = self.wtty.read_nonblocking(timeout, size)
   File "...\expect\wexpect.py", line 2149, in read_nonblocking
     self.switchBack()
   File "...\expect\wexpect.py", line 1975, in switchBack
     FreeConsole()
 error: (87, 'FreeConsole', 'The parameter is incorrect.')

A closer look reveals that line 2149 is the exception handling of another error thrown:
Traceback (most recent call last):
  File "...\expect\wexpect.py", line 2135, in read_nonblocking
    self.switchBack()
  File "...\expect\wexpect.py", line 1976, in switchBack
    AttachConsole(self.__parentPid)
error: (6, 'AttachConsole', 'The handle is invalid.')

The invalid handle error seems to be related to the fact that I'm not launching python.exe from a console but from another process, can it be related to issue 1124861 (http://bugs.python.org/issue1124861)? Do I need to create different stdin and stdout or spawn the child process in any other way from wexpect? Are there any other ways of solving this problem?

The wexpect code used can be found at: http://wstein.org/home/goreckc/sage/wexpect/wexpect.py

If this is not the correct mailing list, I apologize and hope that you can show me where to post my question.

Olof Rutgersson
System Test Engineer

Xelerated AB
Olof Palmes Gata 29
SE-111 22 Stockholm
Sweden





From m.echavarriagregory at umiami.edu  Wed Jan 20 18:56:58 2010
From: m.echavarriagregory at umiami.edu (Echavarria Gregory, Maria Angelica)
Date: Wed, 20 Jan 2010 12:56:58 -0500
Subject: [python-win32] How to use data on disk in this generator
Message-ID: <2E95F75EDC25D54999387A1E2E6EF6274349045A57@MBX02.cgcent.miami.edu>

Good afternoon,

I use PythonWin32 in my XP machine. I am programming an application and have trouble with memory when using the sumComb list below, that has a size of 2,657,205 elements. I DO NEED this many elements because they are product of combinatorial math operations. To solve my memory problem I created a file on disk that stores the elements of the list as they are created. Now, I need to use the 'list' in a generator and have not been able to replicate the effect without actually calling/storing the list in memory, that is, using the data directly from the file on disk:

I need to mimic what the following generator does:
def  validList(sumComb, GammaPossible, valid):
    for l in sumComb:
       for m in sumComb:
         for n in sumComb:
                k = [l,m,n]
                i = 0
                while i < valid:
                     C = 0
                     j = 0
                     while j < 3:
                           C = C +((GammaPossible[i][j])*k[j])
                           j += 1
                     yield C
                     i += 1

By not using the list sumComb in memory, but having its elements in a file F writen using F = open("my_list.txt", "w") and where elements are stored with column "/n" separation. I have not been able to find the appropriate syntax for the nested 'for' loops to replace sumComb list by using F.readlines(), str.split(), etc.  Please help me with this little challenge.

Angelica.


From aahz at pythoncraft.com  Wed Jan 20 20:07:08 2010
From: aahz at pythoncraft.com (Aahz)
Date: Wed, 20 Jan 2010 11:07:08 -0800
Subject: [python-win32] How to use data on disk in this generator
In-Reply-To: <2E95F75EDC25D54999387A1E2E6EF6274349045A57@MBX02.cgcent.miami.edu>
References: <2E95F75EDC25D54999387A1E2E6EF6274349045A57@MBX02.cgcent.miami.edu>
Message-ID: <20100120190707.GA6662@panix.com>

On Wed, Jan 20, 2010, Echavarria Gregory, Maria Angelica wrote:
>
> I use PythonWin32 in my XP machine. I am programming an application
> and have trouble with memory when using the sumComb list below, that
> has a size of 2,657,205 elements. I DO NEED this many elements because
> they are product of combinatorial math operations. To solve my memory
> problem I created a file on disk that stores the elements of the list
> as they are created. Now, I need to use the 'list' in a generator
> and have not been able to replicate the effect without actually
> calling/storing the list in memory, that is, using the data directly
> from the file on disk:

This doesn't seem like a win32-specific issue -- you may get better
responses by posting to comp.lang.python.
-- 
Aahz (aahz at pythoncraft.com)           <*>         http://www.pythoncraft.com/

"If you think it's expensive to hire a professional to do the job, wait
until you hire an amateur."  --Red Adair

From m.echavarriagregory at umiami.edu  Wed Jan 20 20:22:19 2010
From: m.echavarriagregory at umiami.edu (Echavarria Gregory, Maria Angelica)
Date: Wed, 20 Jan 2010 14:22:19 -0500
Subject: [python-win32] How to use data on disk in this generator
In-Reply-To: <2E95F75EDC25D54999387A1E2E6EF6274349045A57@MBX02.cgcent.miami.edu>
References: <2E95F75EDC25D54999387A1E2E6EF6274349045A57@MBX02.cgcent.miami.edu>
Message-ID: <2E95F75EDC25D54999387A1E2E6EF6274349045A5A@MBX02.cgcent.miami.edu>


Good afternoon,

I use PythonWin32 in my XP machine. I am programming an application and have trouble with memory when using the sumComb list below, that has a size of 2,657,205 elements. I DO NEED this many elements because they are product of combinatorial math operations. To solve my memory problem I created a file on disk that stores the elements of the list as they are created. Now, I need to use the 'list' in a generator and have not been able to replicate the effect without actually calling/storing the list in memory, that is, using the data directly from the file on disk:

I need to mimic what the following generator does:
def  validList(sumComb, GammaPossible, valid):
    for l in sumComb:
       for m in sumComb:
         for n in sumComb:
                k = [l,m,n]
                i = 0
                while i < valid:
                     C = 0
                     j = 0
                     while j < 3:
                           C = C +((GammaPossible[i][j])*k[j])
                           j += 1
                     yield C
                     i += 1

By not using the list sumComb in memory, but having its elements in a file F writen using F = open("my_list.txt", "w") and where elements are stored with column "/n" separation. I have not been able to find the appropriate syntax for the nested 'for' loops to replace sumComb list by using F.readlines(), str.split(), etc.  Please help me with this little challenge.

Angelica.


From skippy.hammond at gmail.com  Wed Jan 20 23:22:10 2010
From: skippy.hammond at gmail.com (Mark Hammond)
Date: Thu, 21 Jan 2010 09:22:10 +1100
Subject: [python-win32] wexpect and running without a console
In-Reply-To: <987E17118F2F994D92C0EFD24784D10B1A34257840@xelwa4.xelerated.com>
References: <987E17118F2F994D92C0EFD24784D10B1A34257840@xelwa4.xelerated.com>
Message-ID: <4B578212.7020907@gmail.com>

On 21/01/2010 1:57 AM, Olof Rutgersson wrote:

> The invalid handle error seems to be related to the fact that I'm not launching python.exe from a console but from another process, can it be related to issue 1124861 (http://bugs.python.org/issue1124861)? Do I need to create different stdin and stdout or spawn the child process in any other way from wexpect? Are there any other ways of solving this problem?

The error is because there is no console, but wexpect is attempting to 
use one.  The solution is to fix the code that assumes a console always 
exists - it doesn't.  I think you need to talk to the wexpect guys and 
see if either (a) you are misusing wexpect or (b) it is a bug in wexpect.

Cheers,

Mark

From Olof.Rutgersson at xelerated.com  Thu Jan 21 11:46:58 2010
From: Olof.Rutgersson at xelerated.com (Olof Rutgersson)
Date: Thu, 21 Jan 2010 11:46:58 +0100
Subject: [python-win32] wexpect and running without a console
In-Reply-To: <4B578212.7020907@gmail.com>
References: <987E17118F2F994D92C0EFD24784D10B1A34257840@xelwa4.xelerated.com>
	<4B578212.7020907@gmail.com>
Message-ID: <987E17118F2F994D92C0EFD24784D10B1A342579F7@xelwa4.xelerated.com>

Seems reasonable to me, I don't believe there is an error in my code since it works perfectly running from console and this error occurs instantly when NOT running from console. There is no way when accessing Wexpect to instruct it that Python is running without a console, and there should be no need to either (IMHO). Wexpect doesn't seem to have that much of a development going on, so I guess it's up to me to fix it (or run it on Linux/cygwin and use Pexpect instead).

Thanks for the help!

//Olof Rutgersson
-----Original Message-----
From: Mark Hammond [mailto:skippy.hammond at gmail.com] 
Sent: den 20 januari 2010 23:22
To: Olof Rutgersson
Cc: python-win32 at python.org
Subject: Re: [python-win32] wexpect and running without a console

On 21/01/2010 1:57 AM, Olof Rutgersson wrote:

> The invalid handle error seems to be related to the fact that I'm not launching python.exe from a console but from another process, can it be related to issue 1124861 (http://bugs.python.org/issue1124861)? Do I need to create different stdin and stdout or spawn the child process in any other way from wexpect? Are there any other ways of solving this problem?

The error is because there is no console, but wexpect is attempting to 
use one.  The solution is to fix the code that assumes a console always 
exists - it doesn't.  I think you need to talk to the wexpect guys and 
see if either (a) you are misusing wexpect or (b) it is a bug in wexpect.

Cheers,

Mark



From planders at gmail.com  Fri Jan 22 22:31:04 2010
From: planders at gmail.com (Preston Landers)
Date: Fri, 22 Jan 2010 15:31:04 -0600
Subject: [python-win32] odbc module and Unicode string literals
Message-ID: <796c415c1001221331m33cf932xd0fa26320af6a15f@mail.gmail.com>

Hello all,

I'm curious why the odbc module included in PyWin32 (version 214) is
compiled without the Unicode flags/defines.  This prevents you from
using Unicode string literals in queries.  By that I mean the Unicode
literals are converted to the 8-bit codepage (cp1252) and any
untranslatable characters are replaced with ?.  Unicode works ok in
bind variables, just not as literals in the main query.

That is, unless I set the Unicode flag to True for odbc in PyWin32
setup.py around line 1441.  Then the resulting odbc module seems to
handle Unicode literals just fine.   The test script included below
should demonstrate that.

The ADO module (adodbapi) doesn't have this issue - Unicode literals
work ok.  But it suffers from a few other problems, the most basic one
is a performance gap of 25-30% compared to ODBC even after I generated
the bindings with makepy.  I assume this is due to general COM
overhead.

I can compile my own odbc module with the Unicode flag but I'm curious
if there was a reason it was turned off to begin with.  Maybe I'm
missing something or would run into trouble later if I did that.

thanks,
-Preston


# Note: the goal should be to get "unicode intact" twice.
# First time is using a bind variable, second time is a string literal.
# This assumes SQL Server backend - string literals are prefixed with N
import odbc

# Change to suit your server
CONNSTR = "dsn/username/password"

db = odbc.odbc(CONNSTR)
cursor = db.cursor()

# This has some extra-wide characters (surrogate pairs)
testStr = u"\u0669(\u0361\u0e4f\u032f\u0361\u0e4f)\u06f6
.\U0002a5ab\U0001d122\U00024b00"

def query(sql, params):
    cursor.execute(sql, params)
    recs = cursor.fetchall()

    print repr(sql), repr(params)
    print repr(recs)

    if recs[0][0] == testStr:
        print "unicode intact."
    else:
        print "unicode did not survive round trip"

    print
    return recs

# Test as bind variable
query1 = "select ? as 'test'"
query(query1, [testStr])

# Test as straight Unicode literal
query2 = "select N'%s' as 'test'" % (testStr,)
query(query2, [])

From whatyoulookin at yahoo.com  Fri Jan 22 23:03:53 2010
From: whatyoulookin at yahoo.com (Alec Bennett)
Date: Fri, 22 Jan 2010 14:03:53 -0800 (PST)
Subject: [python-win32] Problem accessing GetFontData in Windows API
Message-ID: <225358.35751.qm@web112604.mail.gq1.yahoo.com>

I'm trying to access the following function from the Windows API:

http://msdn.microsoft.com/en-us/library/dd144885(VS.85).aspx

I'm sure I'm drastically over simplifying, but I'm testing it like this:

import win32api
print win32api.GetFontData()

This of course errors out with "AttributeError: function 'GetFontData' not found".

Does anyone have any tips on how I might access this function? I need to resolve the font face name (for example, "Arial") as returned by wxPython, to the Windows filename (for example, c:\windows\fonts\arial.ttf) as required by the Image module, and it looks like GetFontData() from the Windows API is the solution. 





      

From planders at gmail.com  Fri Jan 22 23:40:32 2010
From: planders at gmail.com (Preston Landers)
Date: Fri, 22 Jan 2010 16:40:32 -0600
Subject: [python-win32] Problem accessing GetFontData in Windows API
Message-ID: <F3851465-3DE3-482E-97A5-42D6E899C35E@gmail.com>

> I'm trying to access the following function from the Windows API:
> 
> http://msdn.microsoft.com/en-us/library/dd144885(VS.85).aspx
> 
> I'm sure I'm drastically over simplifying, but I'm testing it like this:
> 
> import win32api
> print win32api.GetFontData()
> 
> This of course errors out with "AttributeError: function 'GetFontData' not found".

I'm not 100% sure but I think you're going to need to use ctypes to call that. 

It's a little complicated but this may help:

http://python.net/crew/theller/ctypes/tutorial.html

>>> from ctypes import *
>>> windll.gdi32
<WinDLL 'gdi32', handle fe340000 at 27b4cc0>
>>> windll.gdi32.GetFontData
<_FuncPtr object at 0x000000000273B2B8>


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-win32/attachments/20100122/f1a3b8a5/attachment.htm>

From timr at probo.com  Fri Jan 22 23:51:09 2010
From: timr at probo.com (Tim Roberts)
Date: Fri, 22 Jan 2010 14:51:09 -0800
Subject: [python-win32] Problem accessing GetFontData in Windows API
In-Reply-To: <225358.35751.qm@web112604.mail.gq1.yahoo.com>
References: <225358.35751.qm@web112604.mail.gq1.yahoo.com>
Message-ID: <4B5A2BDD.1070909@probo.com>

Alec Bennett wrote:
> I'm trying to access the following function from the Windows API:
>
> http://msdn.microsoft.com/en-us/library/dd144885(VS.85).aspx
>
> I'm sure I'm drastically over simplifying, but I'm testing it like this:
>
> import win32api
> print win32api.GetFontData()
>
> This of course errors out with "AttributeError: function 'GetFontData' not found".
>
> Does anyone have any tips on how I might access this function? I need to resolve the font face name (for example, "Arial") as returned by wxPython, to the Windows filename (for example, c:\windows\fonts\arial.ttf) as required by the Image module, and it looks like GetFontData() from the Windows API is the solution. 
>   

Although it IS possible to find this information using the API, it is
quite tedious.  You'd need to have a DC, create an HFONT, select it into
the DC, then run GetOutlineTextMetrics, which is not a Python-friendly
API.  You get back a long structure where some members include the
offset to the strings you need.

A much better solution is to search through the registry values in
HKLM\SOFTWARE\Microsoft\Windows NT\CurrentVersion\Fonts.  That maps each
font name to its corresponding file name.

-- 
Tim Roberts, timr at probo.com
Providenza & Boekelheide, Inc.


From whatyoulookin at yahoo.com  Sat Jan 23 03:02:17 2010
From: whatyoulookin at yahoo.com (Alec Bennett)
Date: Fri, 22 Jan 2010 18:02:17 -0800 (PST)
Subject: [python-win32] Problem accessing GetFontData in Windows API
In-Reply-To: <4B5A2BDD.1070909@probo.com>
Message-ID: <718852.22422.qm@web112608.mail.gq1.yahoo.com>

> A much better solution is to search through the registry
> values in HKLM\SOFTWARE\Microsoft\Windows

Three things concern me with this method: 

1) if I understand correctly, Windows appends the language to the font title. In other words, for German the Arial font is "arial german". Normally this wouldn't be the biggest issue in the world, but there are often multiple font names with the same string in them, one for bold, one for italic, etc., so its not simply a matter of returning any entry with the name "arial" in it.

2) under Vista 64-bit I've been getting permission denied errors when trying to read the registry using winreg. 

3) its a kludge and bound to be inaccurate on some systems.

But yeah, it looks like it might be my best option. 







      

From timr at probo.com  Sat Jan 23 03:40:34 2010
From: timr at probo.com (Tim Roberts)
Date: Fri, 22 Jan 2010 18:40:34 -0800
Subject: [python-win32] Problem accessing GetFontData in Windows API
In-Reply-To: <718852.22422.qm@web112608.mail.gq1.yahoo.com>
References: <718852.22422.qm@web112608.mail.gq1.yahoo.com>
Message-ID: <4B5A61A2.8020301@probo.com>

Alec Bennett wrote:
>> A much better solution is to search through the registry
>> values in HKLM\SOFTWARE\Microsoft\Windows
>>     
>
> Three things concern me with this method: 
>
> 1) if I understand correctly, Windows appends the language to the font title. In other words, for German the Arial font is "arial german". Normally this wouldn't be the biggest issue in the world, but there are often multiple font names with the same string in them, one for bold, one for italic, etc., so its not simply a matter of returning any entry with the name "arial" in it.
>   

There is no "Arial German".  That's the whole point of Unicode and
TrueType.  All language glyphs are in one file.  You would have to deal
with bold, italic, and bold italic, but those are in the listing.

If you have antique raster fonts, it's possible you could come across
this, but those are separate files, so I think that's what you want.

> 2) under Vista 64-bit I've been getting permission denied errors when trying to read the registry using winreg. 
>   

Even if you set read-only?  I thought HKEY_LOCAL_MACHINE was readable
but not writable.

> 3) its a kludge and bound to be inaccurate on some systems.
>   

Not really -- this is the same database used by the system's font
manager.  That data is not simply decorative.  It's no more kludgy than
using OUTLINETEXTMETRICS

-- 
Tim Roberts, timr at probo.com
Providenza & Boekelheide, Inc.


From whatyoulookin at yahoo.com  Sat Jan 23 04:19:41 2010
From: whatyoulookin at yahoo.com (Alec Bennett)
Date: Fri, 22 Jan 2010 19:19:41 -0800 (PST)
Subject: [python-win32] Problem accessing GetFontData in Windows API
In-Reply-To: <4B5A61A2.8020301@probo.com>
Message-ID: <276251.93560.qm@web112603.mail.gq1.yahoo.com>

Thanks for the tips.

Using that method I'm able to obtain the filename for about 75% of the fonts in my Windows directory, but I can't resolve fonts with names such as this:

- the font's name is Arno Pro. It shows up in font choosers with names like "Arno Pro", "Arno Pro Caption", "Arno Pro Display", etc.

- however, no such names appear in the Windows font directory/registry. This font appears as, for example, "ArnoPro-Regular", "ArnoPro-Caption", "ArnoPro-Display".

I don't imagine you can think of a way to obtain the file name for fonts with that sort of naming convention?

I have to leave now but I can post a sample app later.













      

From skippy.hammond at gmail.com  Sun Jan 24 03:27:52 2010
From: skippy.hammond at gmail.com (Mark Hammond)
Date: Sun, 24 Jan 2010 13:27:52 +1100
Subject: [python-win32] odbc module and Unicode string literals
In-Reply-To: <796c415c1001221331m33cf932xd0fa26320af6a15f@mail.gmail.com>
References: <796c415c1001221331m33cf932xd0fa26320af6a15f@mail.gmail.com>
Message-ID: <4B5BB028.6040208@gmail.com>

On 23/01/2010 8:31 AM, Preston Landers wrote:
> Hello all,
>
> I'm curious why the odbc module included in PyWin32 (version 214) is
> compiled without the Unicode flags/defines.

No good reason I am aware of.  If you submit a patch, it included some 
tests and no-one else who actually uses this module objects, I'll apply it.

Cheers,

Mark

From lyeoh at pop.jaring.my  Sun Jan 24 15:07:35 2010
From: lyeoh at pop.jaring.my (Lincoln Yeoh)
Date: Sun, 24 Jan 2010 22:07:35 +0800
Subject: [python-win32] Getting a list of "task buttons" or
 tasks/windows in the "alt-tab" list?
In-Reply-To: <201001131832.o0DIWJPw019549@vsmtp6.jaring.my>
References: <201001131832.o0DIWJPw019549@vsmtp6.jaring.my>
Message-ID: <201001241408.o0OE84YA044194@vsmtp3.jaring.my>

Anyway to answer my own question :).

Excel's main window has an "ITaskList_Deleted" window property if the 
dummy windows exist.

Powerpoint behaves very similarly. And Kingsoft Office appears to do 
something like that except without the ITaskList_Deleted window property.

Link.

At 02:32 AM 1/14/2010, Lincoln Yeoh wrote:
>Hi,
>
>With python win32 is there a way to get only the windows that are 
>represented by the buttons of the Windows taskbar (or the windows 
>listed in the "alt-tab" list)?
>
>I have found various ways of doing so, but I can't seem to get them 
>to work with Excel's windows.
>
> From what I observe (I may be wrong though), Excel by default 
> appears to have a main window and some "dummy windows" 
> (classname=MS-SDIb) representing the opened Excel documents. If you 
> click on those "dummy windows" the relevant excel document gets 
> foregrounded in the _main_ window (not the dummy windows).
>
>If three Excel documents are opened, there are four top level Excel 
>windows (or at least they appear to be top level to me) - the main, 
>and three MS-SDIb ones. But the taskbar and alt-tab list only show 3 
>items. What logic should one use to create the same list?



From lyeoh at pop.jaring.my  Sun Jan 24 15:31:56 2010
From: lyeoh at pop.jaring.my (Lincoln Yeoh)
Date: Sun, 24 Jan 2010 22:31:56 +0800
Subject: [python-win32] WMI troubles!
In-Reply-To: <4B50D0AE.1060509@timgolden.me.uk>
References: <558CA385CC304B349F3FEB6D62A466DC@AlexPC>
	<4B50B4F6.1040704@probo.com>
	<3164DF6271584B1098783412B56698DB@AlexPC>
	<4B50C009.10300@probo.com>
	<D2A9CA10622C4A43AEB049F3EF5AAD2A@AlexPC>
	<4B50D0AE.1060509@timgolden.me.uk>
Message-ID: <201001241432.o0OEWIOF035248@vsmtp6.jaring.my>

Hi,

Coincidentally I'm using the hotkey code from Tim's url and it works 
quite well.

But I'm not sure which is better:

user32.GetMessageA or user32.GetMessageW (along with the 
corresponding user32.DispatchMessageA or user32.DispatchMessageW of course).

Regards,

Link.


At 04:31 AM 1/16/2010, Tim Golden wrote:
>On 15/01/2010 19:38, Alex Hall wrote:
>>I do not want a single script because, eventually, the plan is to have a
>>keystroke for core 1, a keystroke for core 2, a keystroke for percent of
>>ram used, a keystroke for hard disk space, and so on. While I could
>>write each function in its own script and put a hotkey on each one, it
>>seems easier and faster to just make one script and run it once,
>>avoiding the delay I always get when this script runs for the first time
>>from happening each time I press a hotkey.
>
>I've got slightly lost in this thread (not helped, I'm
>afraid Alex, by your habit of top-posting ;) ) but in
>an effort to be helpful, can I suggest the following.
>Instead of fiddling with pyHooks which are -- as Tim R
>points out -- somewhat dark and messy, consider adapting
>the following example instead:
>
>http://timgolden.me.uk/python/win32_how_do_i/catch_system_wide_hotkeys.html
>
>This uses a straightforward Windows message loop to
>provide to sort of global hotkeys I think you're after.
>You might, for example, change the handle_win_f3
>function to start a wmi instance (or use an existing
>one) and then produce the output you were after.
>
>TJG
>_______________________________________________
>python-win32 mailing list
>python-win32 at python.org
>http://mail.python.org/mailman/listinfo/python-win32



From timr at probo.com  Mon Jan 25 18:01:13 2010
From: timr at probo.com (Tim Roberts)
Date: Mon, 25 Jan 2010 09:01:13 -0800
Subject: [python-win32] Problem accessing GetFontData in Windows API
In-Reply-To: <276251.93560.qm@web112603.mail.gq1.yahoo.com>
References: <276251.93560.qm@web112603.mail.gq1.yahoo.com>
Message-ID: <4B5DCE59.4030401@probo.com>

Alec Bennett wrote:
> Thanks for the tips.
>
> Using that method I'm able to obtain the filename for about 75% of the fonts in my Windows directory, but I can't resolve fonts with names such as this:
>
> - the font's name is Arno Pro. It shows up in font choosers with names like "Arno Pro", "Arno Pro Caption", "Arno Pro Display", etc.
>
> - however, no such names appear in the Windows font directory/registry. This font appears as, for example, "ArnoPro-Regular", "ArnoPro-Caption", "ArnoPro-Display".
>
> I don't imagine you can think of a way to obtain the file name for fonts with that sort of naming convention?
>   

Yecch.  Yes, this is ugly.  TrueType font have several names, including
a "full name" and a "PostScript name", where the PostScript name must be
ASCII and usually has no spaces.  I'm not sure I have an answer for this
problem.

In the worst case, I suppose you could do like GIMP does, and scan
through %SystemRoot%\Fonts on your own.  You could cache the information
so you only checked files that had changed.

The alternative is to go through the trouble of constructing a ctypes
wrapper for GetOutlineTextMetrics.

-- 
Tim Roberts, timr at probo.com
Providenza & Boekelheide, Inc.


From timr at probo.com  Mon Jan 25 19:06:52 2010
From: timr at probo.com (Tim Roberts)
Date: Mon, 25 Jan 2010 10:06:52 -0800
Subject: [python-win32] WMI troubles!
In-Reply-To: <201001241432.o0OEWIOF035248@vsmtp6.jaring.my>
References: <558CA385CC304B349F3FEB6D62A466DC@AlexPC>	<4B50B4F6.1040704@probo.com>	<3164DF6271584B1098783412B56698DB@AlexPC>	<4B50C009.10300@probo.com>	<D2A9CA10622C4A43AEB049F3EF5AAD2A@AlexPC>	<4B50D0AE.1060509@timgolden.me.uk>
	<201001241432.o0OEWIOF035248@vsmtp6.jaring.my>
Message-ID: <4B5DDDBC.5020403@probo.com>

Lincoln Yeoh wrote:
> Hi,
>
> Coincidentally I'm using the hotkey code from Tim's url and it works
> quite well.
>
> But I'm not sure which is better:
>
> user32.GetMessageA or user32.GetMessageW (along with the corresponding
> user32.DispatchMessageA or user32.DispatchMessageW of course).

That depends on the messages you're handling.  For some messages, WPARAM
and LPARAM point to structures that contain strings.  For those, you
need to know the character type.  If there are no strings, then it
doesn't matter which one you choose.

-- 
Tim Roberts, timr at probo.com
Providenza & Boekelheide, Inc.


From admin at genesisware.de  Wed Jan 27 09:03:07 2010
From: admin at genesisware.de (Stefan George)
Date: Wed, 27 Jan 2010 09:03:07 +0100
Subject: [python-win32] Visibility information
Message-ID: <000e01ca9f27$2984cb30$7c8e6190$@de>

Good morning PyWin32-users,

 

I am using the PyWin32 package and I am wondering if there is any way to get
visibility information of a button (I only know the handle).

Is that possible and is there any code snippet, you could pass to me?

 

I am looking for something like isVisible(hwnd) returns True or False.

 

Thank you!

 

Greetings from Germany,

 

Stefan George

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-win32/attachments/20100127/2ddfcedd/attachment.htm>

From efotinis at yahoo.com  Wed Jan 27 09:49:36 2010
From: efotinis at yahoo.com (Elias Fotinis)
Date: Wed, 27 Jan 2010 10:49:36 +0200
Subject: [python-win32] Visibility information
In-Reply-To: <000e01ca9f27$2984cb30$7c8e6190$@de>
References: <000e01ca9f27$2984cb30$7c8e6190$@de>
Message-ID: <745659D81FA14AC894C51C132EBDDC78@efcore>

From: Stefan George

> I am using the PyWin32 package and I am wondering if there is any way to 
> get visibility information of a button (I only know the handle).
> Is that possible and is there any code snippet, you could pass to me?
>
> I am looking for something like isVisible(hwnd) returns True or False.

It's pretty simple:

    import win32gui
    button_handle = ...
    is_visible = win32gui.IsWindowVisible(button_handle)

The win32gui contains all GUI related Windows API functions.

PyWin32's help file is great for looking up Windows functions and their 
respective modules. Just switch to the Index tab, type the function name and 
you'll find a help page with the parameters, return type and containing 
module of each API function.


From stef.mientki at gmail.com  Wed Jan 27 15:12:37 2010
From: stef.mientki at gmail.com (Stef Mientki)
Date: Wed, 27 Jan 2010 15:12:37 +0100
Subject: [python-win32] Maybe the wrong newsgroup to ask ?
Message-ID: <4B6049D5.6050407@gmail.com>

hello,

I'm trying to kill processes,
I started myself with subprocess.popen,
under windows XP, Python 2.6

Why are the subprocess.Popen methods kill() and terminate () not working,
while a simple suggestion from this newsgroup, shown below, works perfect ?

thanks,
Stef Mientki


     My_Process = subprocess.popen ( ...)
     handle = win32api.OpenProcess(win32con.PROCESS_TERMINATE, 0, 
My_Process.PID)
     try:
       win32api.TerminateProcess(handle,0)
       win32api.CloseHandle(handle)
     except:   #the process might already be closed by the user
       pass

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-win32/attachments/20100127/bf5f9015/attachment.htm>

From Sunny.Carter at metaswitch.com  Wed Jan 27 17:21:01 2010
From: Sunny.Carter at metaswitch.com (Sunny Carter)
Date: Wed, 27 Jan 2010 16:21:01 +0000
Subject: [python-win32] Boolean type changed in Python 3.0?
In-Reply-To: <4B42F11D.3010109@gmail.com>
References: <44EE1E31095AE349AC6C3C0B69FFBF02932E0B3D6F@ENFIMBOX1.ad.datcon.co.uk>
	<4B42F11D.3010109@gmail.com>
Message-ID: <44EE1E31095AE349AC6C3C0B69FFBF02932F2F7D43@ENFIMBOX1.ad.datcon.co.uk>

Mark - just wondered if you had any update on this?

I'm very happy to run some diagnostic tests for you on my setup, if this would help? If you need to send me debug versions or anything then please do, or let me know if I can provide any further info that would help to diagnose.

I can do my best to net down the problem and promise not to just spam you with any output - I spend my days debugging various other problems so am pretty familiar with providing useful info.

Thanks, appreciate it.

Sunny 

-----Original Message-----
From: Mark Hammond [mailto:skippy.hammond at gmail.com] 
Sent: 05 January 2010 07:58
To: Sunny Carter
Cc: python-win32 at python.org
Subject: Re: [python-win32] Boolean type changed in Python 3.0?

On 5/01/2010 3:51 AM, Sunny Carter wrote:
> Hi all,
> I'm trying my posting again with a different subject so that it is 
> more generic.
> I am having problems calling across the win32 API using a Boolean 
> argument in Python 3.0 (False in my python script) which is not 
> recognised as a Boolean (The error I get back from setattr is 'Boolean 
> value expected').
> Has the way that a Boolean is represented changed in Python 3.0? This 
> worked fine in Python 2.6. I cannot get it to work and have tried 
> values of False,True,0 and 1, all to no avail.
> It sounds like a bug in the win32com API for 3.0 to me.
> Further details below.
> Many thanks to anyone that can help,

I expect something subtle is going on with boolean conversions in 3.x, but the test suite does have coverage of booleans, so I'm really not sure what the problem could be.  I don't have photoshop so I'm unable to reproduce - I intend revisiting the tests to see if I can spot an edge case which isn't tested, but I'm yet to get to that after the holidays...

Cheers,

Mark

From skippy.hammond at gmail.com  Wed Jan 27 18:28:27 2010
From: skippy.hammond at gmail.com (Mark Hammond)
Date: Wed, 27 Jan 2010 09:28:27 -0800
Subject: [python-win32] Boolean type changed in Python 3.0?
In-Reply-To: <44EE1E31095AE349AC6C3C0B69FFBF02932F2F7D43@ENFIMBOX1.ad.datcon.co.uk>
References: <44EE1E31095AE349AC6C3C0B69FFBF02932E0B3D6F@ENFIMBOX1.ad.datcon.co.uk>
	<4B42F11D.3010109@gmail.com>
	<44EE1E31095AE349AC6C3C0B69FFBF02932F2F7D43@ENFIMBOX1.ad.datcon.co.uk>
Message-ID: <4B6077BB.5020109@gmail.com>

On 27/01/2010 8:21 AM, Sunny Carter wrote:
> Mark - just wondered if you had any update on this?

I believe Roger checked a fix in for this over the last few days.  I'm 
not sure when a new binary build will be available though, but probably 
not within a few weeks...

Cheers,

Mark

From Sunny.Carter at metaswitch.com  Wed Jan 27 19:34:04 2010
From: Sunny.Carter at metaswitch.com (Sunny Carter)
Date: Wed, 27 Jan 2010 18:34:04 +0000
Subject: [python-win32] Boolean type changed in Python 3.0?
In-Reply-To: <4B6077BB.5020109@gmail.com>
References: <44EE1E31095AE349AC6C3C0B69FFBF02932E0B3D6F@ENFIMBOX1.ad.datcon.co.uk>
	<4B42F11D.3010109@gmail.com>
	<44EE1E31095AE349AC6C3C0B69FFBF02932F2F7D43@ENFIMBOX1.ad.datcon.co.uk>
	<4B6077BB.5020109@gmail.com>
Message-ID: <44EE1E31095AE349AC6C3C0B69FFBF02932F2F7EEB@ENFIMBOX1.ad.datcon.co.uk>

Fabulous - thanks for the update.

Sunny 

-----Original Message-----
From: Mark Hammond [mailto:skippy.hammond at gmail.com] 
Sent: 27 January 2010 17:28
To: Sunny Carter
Cc: python-win32 at python.org
Subject: Re: [python-win32] Boolean type changed in Python 3.0?

On 27/01/2010 8:21 AM, Sunny Carter wrote:
> Mark - just wondered if you had any update on this?

I believe Roger checked a fix in for this over the last few days.  I'm not sure when a new binary build will be available though, but probably not within a few weeks...

Cheers,

Mark

From timr at probo.com  Wed Jan 27 20:07:57 2010
From: timr at probo.com (Tim Roberts)
Date: Wed, 27 Jan 2010 11:07:57 -0800
Subject: [python-win32] Maybe the wrong newsgroup to ask ?
In-Reply-To: <4B6049D5.6050407@gmail.com>
References: <4B6049D5.6050407@gmail.com>
Message-ID: <4B608F0D.3010607@probo.com>

Stef Mientki wrote:
>
> I'm trying to kill processes,
> I started myself with subprocess.popen,
> under windows XP, Python 2.6
>
> Why are the subprocess.Popen methods kill() and terminate () not working,
> while a simple suggestion from this newsgroup, shown below, works
> perfect ?
> ...
>
>     My_Process = subprocess.popen ( ...)
>     handle = win32api.OpenProcess(win32con.PROCESS_TERMINATE, 0,
> My_Process.PID)
>     try:
>       win32api.TerminateProcess(handle,0)
>       win32api.CloseHandle(handle)       
>     except:   #the process might already be closed by the user
>       pass

There shouldn't be any difference.  If you look at the code in
subprocess. it calls the same API.  The handle returned from
CreateProcess has all access rights, including PROCESS_TERMINATE.

Can you show us your original code?

-- 
Tim Roberts, timr at probo.com
Providenza & Boekelheide, Inc.


From stef.mientki at gmail.com  Wed Jan 27 20:45:56 2010
From: stef.mientki at gmail.com (Stef Mientki)
Date: Wed, 27 Jan 2010 20:45:56 +0100
Subject: [python-win32] Maybe the wrong newsgroup to ask ?
In-Reply-To: <4B608F0D.3010607@probo.com>
References: <4B6049D5.6050407@gmail.com> <4B608F0D.3010607@probo.com>
Message-ID: <4B6097F4.8010003@gmail.com>

On 27-01-2010 20:07, Tim Roberts wrote:
> Stef Mientki wrote:
>    
>> I'm trying to kill processes,
>> I started myself with subprocess.popen,
>> under windows XP, Python 2.6
>>
>> Why are the subprocess.Popen methods kill() and terminate () not working,
>> while a simple suggestion from this newsgroup, shown below, works
>> perfect ?
>> ...
>>
>>      My_Process = subprocess.popen ( ...)
>>      handle = win32api.OpenProcess(win32con.PROCESS_TERMINATE, 0,
>> My_Process.PID)
>>      try:
>>        win32api.TerminateProcess(handle,0)
>>        win32api.CloseHandle(handle)
>>      except:   #the process might already be closed by the user
>>        pass
>>      
> There shouldn't be any difference.  If you look at the code in
> subprocess. it calls the same API.  The handle returned from
> CreateProcess has all access rights, including PROCESS_TERMINATE.
>
> Can you show us your original code?
>
>    
Good Question Tim.
Showing the code is always difficult for me, because I encapsulate these 
complex things.
The code I gave above doesn't work either :-)
I use a kill method that also kills the children and seems to be the 
essential thing is this situation,
(btw why are there children ?)
thanks for the tip,
cheers,
Stef

import os
import subprocess
import time

if __name__ == '__main__':
   My_Path = os.path.split ( __file__ ) [0]
   cwd = os.path.join ( My_Path, '..',
       'Pylab_Works', 'pylab_works_programs', 'PyJamas_Test')
   filename = os.path.join ( cwd, 'Een_Vraag_Test.py' )

   Process =  subprocess.Popen ( [ 'python', filename ],
                                   cwd   = cwd ,
                                   shell =  ( os.name == 'nt') )

   print Process.pid

   time.sleep ( 5 )
   Process.kill ()
   #Process.terminate ()
   print 'Kill or terminate Doesnt succeed'

   time.sleep ( 2 )
   import win32api
   import win32con
   handle = win32api.OpenProcess(win32con.PROCESS_TERMINATE, 0, Process.pid)
   try:
     win32api.TerminateProcess(handle,0)
     win32api.CloseHandle(handle)        #close api
   except:   #the process might already be closed by the user
     pass
   print 'This still Doesnt works'


   time.sleep ( 2 )
   PID = Process.pid
   Children = []
   import win32com.client
   WMI = win32com.client.GetObject ( 'winmgmts:' )
   processes = WMI.InstancesOf ( 'Win32_Process' )
   for process in processes:
     pid    = process.Properties_ ( 'ProcessID' ).Value
     parent = process.Properties_ ( 'ParentProcessId' ).Value
     if parent == PID :
       Children.append ( pid )

   #Insert the parent at the top
   Children.insert ( 0, PID )

   import win32api
   import win32con
   for PID in Children :
     handle = win32api.OpenProcess(win32con.PROCESS_TERMINATE, 0, PID)
     try:
       win32api.TerminateProcess(handle,0)
       win32api.CloseHandle(handle)
     except:   #the process might already be closed by the user
       pass
   print 'Finally this works'


From twhitema at yahoo.com.au  Wed Jan 27 23:05:45 2010
From: twhitema at yahoo.com.au (Todd Whiteman)
Date: Wed, 27 Jan 2010 14:05:45 -0800
Subject: [python-win32] Maybe the wrong newsgroup to ask ?
In-Reply-To: <4B6049D5.6050407@gmail.com>
References: <4B6049D5.6050407@gmail.com>
Message-ID: <4B60B8B9.5010303@yahoo.com.au>

On 10-01-27 06:12 AM, Stef Mientki wrote:
> hello,
>
> I'm trying to kill processes,
> I started myself with subprocess.popen,
> under windows XP, Python 2.6
>
> Why are the subprocess.Popen methods kill() and terminate () not working,
> while a simple suggestion from this newsgroup, shown below, works perfect ?

Note that the kill method will not kill any child processes. If you want 
to be able to kill the launched process and all of it's child processes, 
see here:
http://benjamin.smedbergs.us/blog/2006-12-11/killableprocesspy/

Cheers,
Todd

From timr at probo.com  Thu Jan 28 00:08:18 2010
From: timr at probo.com (Tim Roberts)
Date: Wed, 27 Jan 2010 15:08:18 -0800
Subject: [python-win32] Maybe the wrong newsgroup to ask ?
In-Reply-To: <4B6097F4.8010003@gmail.com>
References: <4B6049D5.6050407@gmail.com> <4B608F0D.3010607@probo.com>
	<4B6097F4.8010003@gmail.com>
Message-ID: <4B60C762.7080401@probo.com>

Stef Mientki wrote:
>
> Showing the code is always difficult for me, because I encapsulate
> these complex things.
> The code I gave above doesn't work either :-)
> I use a kill method that also kills the children and seems to be the
> essential thing is this situation,
> (btw why are there children ?)

There are child processes because you are setting "shell" to true. 
Thus, subprocess actually launches cmd.exe, and cmd.exe launches Python.

Is there a reason you need a subshell, or was that superstition?

-- 
Tim Roberts, timr at probo.com
Providenza & Boekelheide, Inc.


From usfree74 at gmail.com  Thu Jan 28 00:19:49 2010
From: usfree74 at gmail.com (Xin Zhao)
Date: Wed, 27 Jan 2010 15:19:49 -0800
Subject: [python-win32] makepy or genpy issue: no get_accChildCount?
Message-ID: <cbd66b981001271519s23ff9ed6n77135181af8fd110@mail.gmail.com>

Hey,

I am trying to use pywin32 to do program automation. Basically, I need to
get an IAccessible object and call functions on it.

Here is what I did:

x = pythoncom.AccessibleObjectFromWindow(3803742, 0,
pythoncom.IID_IDispatch)
(3803742 is the HWND of an internet explorer window)

Then I do:
z = win32com.client.gencache.EnsureDispatch(x)

Now "z" is an IAccessiable object:
win32com.gen_py.1EA4DBF0-3C3B-11CF-810C-00AA00389B71x0x1x1.IAccessible.IAccessible
instance at 0x15279168

But when I did "dir(z)", I only see:
['CLSID', 'GetaccDefaultAction', 'GetaccDescription', 'GetaccHelp',
'GetaccHelpTopic', 'GetaccKeyboardShortcut', 'Getacc
Name', 'GetaccRole', 'GetaccState', 'GetaccValue', 'SetaccName',
'SetaccValue', '_ApplyTypes_', '__doc__', '__eq__', '__
getattr__', '__init__', '__module__', '__ne__', '__repr__', '__setattr__',
'_get_good_object_', '_get_good_single_object
_', '_oleobj_', '_prop_map_get_', '_prop_map_put_', 'accChild',
'accDoDefaultAction', 'accHitTest', 'accLocation', 'accN
avigate', 'accSelect', 'coclass_clsid']

Although MSDN lists function *get_accChildCount* as a function IAccessible
must support,
I don't see it in the method list of "z". Why?

Please help! Thanks a lot!

xin
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-win32/attachments/20100127/9f400d10/attachment-0001.htm>

From stef.mientki at gmail.com  Thu Jan 28 00:23:10 2010
From: stef.mientki at gmail.com (Stef Mientki)
Date: Thu, 28 Jan 2010 00:23:10 +0100
Subject: [python-win32] Maybe the wrong newsgroup to ask ?
In-Reply-To: <4B60C762.7080401@probo.com>
References: <4B6049D5.6050407@gmail.com>
	<4B608F0D.3010607@probo.com>	<4B6097F4.8010003@gmail.com>
	<4B60C762.7080401@probo.com>
Message-ID: <4B60CADE.4070908@gmail.com>

On 28-01-2010 00:08, Tim Roberts wrote:
> Stef Mientki wrote:
>    
>> Showing the code is always difficult for me, because I encapsulate
>> these complex things.
>> The code I gave above doesn't work either :-)
>> I use a kill method that also kills the children and seems to be the
>> essential thing is this situation,
>> (btw why are there children ?)
>>      
> There are child processes because you are setting "shell" to true.
> Thus, subprocess actually launches cmd.exe, and cmd.exe launches Python.
>
> Is there a reason you need a subshell, or was that superstition?
>
>    
thanks Tim,

I set Shell = True, to get rid of that ugly black dos box,
but on second thought pythonw and Shell=False is a better solution.

cheers,
Stef
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-win32/attachments/20100128/0faf7b1a/attachment.htm>

From blazej.kaczorowski at gmail.com  Thu Jan 28 12:30:57 2010
From: blazej.kaczorowski at gmail.com (=?ISO-8859-2?Q?B=B3a=BFej_Kaczorowski?=)
Date: Thu, 28 Jan 2010 12:30:57 +0100
Subject: [python-win32] Documentation for win32com for Worksheets
	Worksheets.Range and Worksheets.Cells
Message-ID: <2865bc641001280330h6a3cd799j3c286a50c7f5b16d@mail.gmail.com>

Hello,

I use python on win to deal with Excels and stuff at work. I was
trying to find proper documentation of fe. to Worksheets
Worksheets.Range and Worksheets.Cells objects but all I found on
google are some lame tutorials with how to write "Hello World" it is
simply ok for beginning but I need something more (fe. how to freeze
top row in sheet).

Can any1 point a place of full documentation for excel stuff in
win32com? I would be gratefull.

BR,
Blazej

From timr at probo.com  Thu Jan 28 20:36:05 2010
From: timr at probo.com (Tim Roberts)
Date: Thu, 28 Jan 2010 11:36:05 -0800
Subject: [python-win32] Documentation for win32com for
 Worksheets	Worksheets.Range and Worksheets.Cells
In-Reply-To: <2865bc641001280330h6a3cd799j3c286a50c7f5b16d@mail.gmail.com>
References: <2865bc641001280330h6a3cd799j3c286a50c7f5b16d@mail.gmail.com>
Message-ID: <4B61E725.8080100@probo.com>

B?a?ej Kaczorowski wrote:
> I use python on win to deal with Excels and stuff at work. I was
> trying to find proper documentation of fe. to Worksheets
> Worksheets.Range and Worksheets.Cells objects but all I found on
> google are some lame tutorials with how to write "Hello World" it is
> simply ok for beginning but I need something more (fe. how to freeze
> top row in sheet).
>
> Can any1 point a place of full documentation for excel stuff in
> win32com? I would be gratefull.
>   

There is no "full documentation".  Excel is just another COM object, so
there's nothing in win32com that specifically supports Excel.  It's just
like all of the other COM objects.

Microsoft has good documentation for the Excel object model, which
describes the properties and methods that Excel offers, usually showing
C++ or VBA code.  You have to translate that into Python (which is not
always easy).

The task you have chosen is not particularly easy.  To do the task you
have chosen, you need to "split" the window into two panes, then set the
size of the top pane to one row.
    xl = win32com.client.Dispatch("Excel.Application")
    xl.Visible = 1
    w = xl.Application.ActiveWindow
    w.Split = 1

I'm not entirely sure how to set the size of the individual panes.

-- 
Tim Roberts, timr at probo.com
Providenza & Boekelheide, Inc.


From waldemar.osuch at gmail.com  Thu Jan 28 21:03:42 2010
From: waldemar.osuch at gmail.com (Waldemar Osuch)
Date: Thu, 28 Jan 2010 13:03:42 -0700
Subject: [python-win32] Documentation for win32com for Worksheets
	Worksheets.Range and Worksheets.Cells
In-Reply-To: <4B61E725.8080100@probo.com>
References: <2865bc641001280330h6a3cd799j3c286a50c7f5b16d@mail.gmail.com>
	<4B61E725.8080100@probo.com>
Message-ID: <6fae95541001281203q1259f4e5w20475d94af4f0d75@mail.gmail.com>

What Tim said but if you are generating Excel files from
scratch consider using xlwt:
http://pypi.python.org/pypi/xlwt

The advantage is that the module is cross platform and you do
not need to have Excel installed.  In my experience it is also
quicker to generate spreadsheets


import xlwt

values = list('Witamy %s' % i for i in range(100))

hst = xlwt.easyxf('font: bold on, height 260;')

book = xlwt.Workbook(encoding='cp1252')
sh = book.add_sheet('hello')

sh.write(0, 0, 'Heading', hst)
for rx, val in enumerate(values, 1):
    sh.write(rx, 0, val)

# do the freeze
sh.panes_frozen = True
sh.horz_split_pos = 1
sh.remove_splits = True

book.save('hello.xls')


Waldemar

From mrkaczor at interia.pl  Thu Jan 28 21:18:30 2010
From: mrkaczor at interia.pl (=?ISO-8859-2?Q?B=B3a=BFej_Kaczorowski?=)
Date: Thu, 28 Jan 2010 21:18:30 +0100
Subject: [python-win32] Documentation for win32com for Worksheets
	Worksheets.Range and Worksheets.Cells
In-Reply-To: <4B61E725.8080100@probo.com>
References: <2865bc641001280330h6a3cd799j3c286a50c7f5b16d@mail.gmail.com>
	<4B61E725.8080100@probo.com>
Message-ID: <2865bc641001281218t1d3a3a8erd09c7d3f15cd09b4@mail.gmail.com>

A ok, thx, I found it and it is as you are saying thx it worked!

Blazej

On 28 January 2010 20:36, Tim Roberts <timr at probo.com> wrote:
> B?a?ej Kaczorowski wrote:
>> I use python on win to deal with Excels and stuff at work. I was
>> trying to find proper documentation of fe. to Worksheets
>> Worksheets.Range and Worksheets.Cells objects but all I found on
>> google are some lame tutorials with how to write "Hello World" it is
>> simply ok for beginning but I need something more (fe. how to freeze
>> top row in sheet).
>>
>> Can any1 point a place of full documentation for excel stuff in
>> win32com? I would be gratefull.
>>
>
> There is no "full documentation". ?Excel is just another COM object, so
> there's nothing in win32com that specifically supports Excel. ?It's just
> like all of the other COM objects.
>
> Microsoft has good documentation for the Excel object model, which
> describes the properties and methods that Excel offers, usually showing
> C++ or VBA code. ?You have to translate that into Python (which is not
> always easy).
>
> The task you have chosen is not particularly easy. ?To do the task you
> have chosen, you need to "split" the window into two panes, then set the
> size of the top pane to one row.
> ? ?xl = win32com.client.Dispatch("Excel.Application")
> ? ?xl.Visible = 1
> ? ?w = xl.Application.ActiveWindow
> ? ?w.Split = 1
>
> I'm not entirely sure how to set the size of the individual panes.
>
> --
> Tim Roberts, timr at probo.com
> Providenza & Boekelheide, Inc.
>
> _______________________________________________
> python-win32 mailing list
> python-win32 at python.org
> http://mail.python.org/mailman/listinfo/python-win32
>

From mrkaczor at interia.pl  Fri Jan 29 09:30:32 2010
From: mrkaczor at interia.pl (=?ISO-8859-2?Q?B=B3a=BFej_Kaczorowski?=)
Date: Fri, 29 Jan 2010 09:30:32 +0100
Subject: [python-win32] Documentation for win32com for Worksheets
	Worksheets.Range and Worksheets.Cells
In-Reply-To: <2865bc641001281218t1d3a3a8erd09c7d3f15cd09b4@mail.gmail.com>
References: <2865bc641001280330h6a3cd799j3c286a50c7f5b16d@mail.gmail.com>
	<4B61E725.8080100@probo.com>
	<2865bc641001281218t1d3a3a8erd09c7d3f15cd09b4@mail.gmail.com>
Message-ID: <2865bc641001290030q4644b61h1fe9729682b1c68a@mail.gmail.com>

xlwt - yes, but I have python 3.1 in my company and cant downgrade it :(

Blazej

2010/1/28 B?a?ej Kaczorowski <mrkaczor at interia.pl>:
> A ok, thx, I found it and it is as you are saying thx it worked!
>
> Blazej
>
> On 28 January 2010 20:36, Tim Roberts <timr at probo.com> wrote:
>> B?a?ej Kaczorowski wrote:
>>> I use python on win to deal with Excels and stuff at work. I was
>>> trying to find proper documentation of fe. to Worksheets
>>> Worksheets.Range and Worksheets.Cells objects but all I found on
>>> google are some lame tutorials with how to write "Hello World" it is
>>> simply ok for beginning but I need something more (fe. how to freeze
>>> top row in sheet).
>>>
>>> Can any1 point a place of full documentation for excel stuff in
>>> win32com? I would be gratefull.
>>>
>>
>> There is no "full documentation". ?Excel is just another COM object, so
>> there's nothing in win32com that specifically supports Excel. ?It's just
>> like all of the other COM objects.
>>
>> Microsoft has good documentation for the Excel object model, which
>> describes the properties and methods that Excel offers, usually showing
>> C++ or VBA code. ?You have to translate that into Python (which is not
>> always easy).
>>
>> The task you have chosen is not particularly easy. ?To do the task you
>> have chosen, you need to "split" the window into two panes, then set the
>> size of the top pane to one row.
>> ? ?xl = win32com.client.Dispatch("Excel.Application")
>> ? ?xl.Visible = 1
>> ? ?w = xl.Application.ActiveWindow
>> ? ?w.Split = 1
>>
>> I'm not entirely sure how to set the size of the individual panes.
>>
>> --
>> Tim Roberts, timr at probo.com
>> Providenza & Boekelheide, Inc.
>>
>> _______________________________________________
>> python-win32 mailing list
>> python-win32 at python.org
>> http://mail.python.org/mailman/listinfo/python-win32
>>
>

From mrkaczor at interia.pl  Fri Jan 29 09:32:07 2010
From: mrkaczor at interia.pl (=?ISO-8859-2?Q?B=B3a=BFej_Kaczorowski?=)
Date: Fri, 29 Jan 2010 09:32:07 +0100
Subject: [python-win32] Documentation for win32com for Worksheets
	Worksheets.Range and Worksheets.Cells
In-Reply-To: <2865bc641001290030q4644b61h1fe9729682b1c68a@mail.gmail.com>
References: <2865bc641001280330h6a3cd799j3c286a50c7f5b16d@mail.gmail.com>
	<4B61E725.8080100@probo.com>
	<2865bc641001281218t1d3a3a8erd09c7d3f15cd09b4@mail.gmail.com>
	<2865bc641001290030q4644b61h1fe9729682b1c68a@mail.gmail.com>
Message-ID: <2865bc641001290032n23e75ed4j5190f8ff86311222@mail.gmail.com>

o 0.7.2 looks like working, I will give a try thx

Blazej

2010/1/29 B?a?ej Kaczorowski <mrkaczor at interia.pl>:
> xlwt - yes, but I have python 3.1 in my company and cant downgrade it :(
>
> Blazej
>
> 2010/1/28 B?a?ej Kaczorowski <mrkaczor at interia.pl>:
>> A ok, thx, I found it and it is as you are saying thx it worked!
>>
>> Blazej
>>
>> On 28 January 2010 20:36, Tim Roberts <timr at probo.com> wrote:
>>> B?a?ej Kaczorowski wrote:
>>>> I use python on win to deal with Excels and stuff at work. I was
>>>> trying to find proper documentation of fe. to Worksheets
>>>> Worksheets.Range and Worksheets.Cells objects but all I found on
>>>> google are some lame tutorials with how to write "Hello World" it is
>>>> simply ok for beginning but I need something more (fe. how to freeze
>>>> top row in sheet).
>>>>
>>>> Can any1 point a place of full documentation for excel stuff in
>>>> win32com? I would be gratefull.
>>>>
>>>
>>> There is no "full documentation". ?Excel is just another COM object, so
>>> there's nothing in win32com that specifically supports Excel. ?It's just
>>> like all of the other COM objects.
>>>
>>> Microsoft has good documentation for the Excel object model, which
>>> describes the properties and methods that Excel offers, usually showing
>>> C++ or VBA code. ?You have to translate that into Python (which is not
>>> always easy).
>>>
>>> The task you have chosen is not particularly easy. ?To do the task you
>>> have chosen, you need to "split" the window into two panes, then set the
>>> size of the top pane to one row.
>>> ? ?xl = win32com.client.Dispatch("Excel.Application")
>>> ? ?xl.Visible = 1
>>> ? ?w = xl.Application.ActiveWindow
>>> ? ?w.Split = 1
>>>
>>> I'm not entirely sure how to set the size of the individual panes.
>>>
>>> --
>>> Tim Roberts, timr at probo.com
>>> Providenza & Boekelheide, Inc.
>>>
>>> _______________________________________________
>>> python-win32 mailing list
>>> python-win32 at python.org
>>> http://mail.python.org/mailman/listinfo/python-win32
>>>
>>
>

From mrkaczor at interia.pl  Fri Jan 29 09:46:43 2010
From: mrkaczor at interia.pl (=?ISO-8859-2?Q?B=B3a=BFej_Kaczorowski?=)
Date: Fri, 29 Jan 2010 09:46:43 +0100
Subject: [python-win32] Documentation for win32com for Worksheets
	Worksheets.Range and Worksheets.Cells
In-Reply-To: <2865bc641001290032n23e75ed4j5190f8ff86311222@mail.gmail.com>
References: <2865bc641001280330h6a3cd799j3c286a50c7f5b16d@mail.gmail.com>
	<4B61E725.8080100@probo.com>
	<2865bc641001281218t1d3a3a8erd09c7d3f15cd09b4@mail.gmail.com>
	<2865bc641001290030q4644b61h1fe9729682b1c68a@mail.gmail.com>
	<2865bc641001290032n23e75ed4j5190f8ff86311222@mail.gmail.com>
Message-ID: <2865bc641001290046l539baaeh8780a6a825037442@mail.gmail.com>

Not working, cant import library ... :( there is a way to run xlwt or
similar on python 3.1?

Blazej

2010/1/29 B?a?ej Kaczorowski <mrkaczor at interia.pl>:
> o 0.7.2 looks like working, I will give a try thx
>
> Blazej
>
> 2010/1/29 B?a?ej Kaczorowski <mrkaczor at interia.pl>:
>> xlwt - yes, but I have python 3.1 in my company and cant downgrade it :(
>>
>> Blazej
>>
>> 2010/1/28 B?a?ej Kaczorowski <mrkaczor at interia.pl>:
>>> A ok, thx, I found it and it is as you are saying thx it worked!
>>>
>>> Blazej
>>>
>>> On 28 January 2010 20:36, Tim Roberts <timr at probo.com> wrote:
>>>> B?a?ej Kaczorowski wrote:
>>>>> I use python on win to deal with Excels and stuff at work. I was
>>>>> trying to find proper documentation of fe. to Worksheets
>>>>> Worksheets.Range and Worksheets.Cells objects but all I found on
>>>>> google are some lame tutorials with how to write "Hello World" it is
>>>>> simply ok for beginning but I need something more (fe. how to freeze
>>>>> top row in sheet).
>>>>>
>>>>> Can any1 point a place of full documentation for excel stuff in
>>>>> win32com? I would be gratefull.
>>>>>
>>>>
>>>> There is no "full documentation". ?Excel is just another COM object, so
>>>> there's nothing in win32com that specifically supports Excel. ?It's just
>>>> like all of the other COM objects.
>>>>
>>>> Microsoft has good documentation for the Excel object model, which
>>>> describes the properties and methods that Excel offers, usually showing
>>>> C++ or VBA code. ?You have to translate that into Python (which is not
>>>> always easy).
>>>>
>>>> The task you have chosen is not particularly easy. ?To do the task you
>>>> have chosen, you need to "split" the window into two panes, then set the
>>>> size of the top pane to one row.
>>>> ? ?xl = win32com.client.Dispatch("Excel.Application")
>>>> ? ?xl.Visible = 1
>>>> ? ?w = xl.Application.ActiveWindow
>>>> ? ?w.Split = 1
>>>>
>>>> I'm not entirely sure how to set the size of the individual panes.
>>>>
>>>> --
>>>> Tim Roberts, timr at probo.com
>>>> Providenza & Boekelheide, Inc.
>>>>
>>>> _______________________________________________
>>>> python-win32 mailing list
>>>> python-win32 at python.org
>>>> http://mail.python.org/mailman/listinfo/python-win32
>>>>
>>>
>>
>

From mail at timgolden.me.uk  Fri Jan 29 10:05:45 2010
From: mail at timgolden.me.uk (Tim Golden)
Date: Fri, 29 Jan 2010 09:05:45 +0000
Subject: [python-win32] Documentation for win32com for
 Worksheets	Worksheets.Range and Worksheets.Cells
In-Reply-To: <2865bc641001290046l539baaeh8780a6a825037442@mail.gmail.com>
References: <2865bc641001280330h6a3cd799j3c286a50c7f5b16d@mail.gmail.com>	<4B61E725.8080100@probo.com>	<2865bc641001281218t1d3a3a8erd09c7d3f15cd09b4@mail.gmail.com>	<2865bc641001290030q4644b61h1fe9729682b1c68a@mail.gmail.com>	<2865bc641001290032n23e75ed4j5190f8ff86311222@mail.gmail.com>
	<2865bc641001290046l539baaeh8780a6a825037442@mail.gmail.com>
Message-ID: <4B62A4E9.4090703@timgolden.me.uk>

On 29/01/2010 08:46, B?a?ej Kaczorowski wrote:
> Not working, cant import library ... :( there is a way to run xlwt or
> similar on python 3.1?

If you haven't already, you really want to look over here:

   http://www.python-excel.org/

and in particular at the mailing list referred to at the bottom
of the page.

TJG

From mrkaczor at interia.pl  Fri Jan 29 10:48:27 2010
From: mrkaczor at interia.pl (=?ISO-8859-2?Q?B=B3a=BFej_Kaczorowski?=)
Date: Fri, 29 Jan 2010 10:48:27 +0100
Subject: [python-win32] Documentation for win32com for Worksheets
	Worksheets.Range and Worksheets.Cells
In-Reply-To: <4B62A4E9.4090703@timgolden.me.uk>
References: <2865bc641001280330h6a3cd799j3c286a50c7f5b16d@mail.gmail.com>
	<4B61E725.8080100@probo.com>
	<2865bc641001281218t1d3a3a8erd09c7d3f15cd09b4@mail.gmail.com>
	<2865bc641001290030q4644b61h1fe9729682b1c68a@mail.gmail.com>
	<2865bc641001290032n23e75ed4j5190f8ff86311222@mail.gmail.com>
	<2865bc641001290046l539baaeh8780a6a825037442@mail.gmail.com>
	<4B62A4E9.4090703@timgolden.me.uk>
Message-ID: <2865bc641001290148k7e9fbe6ekf7d8272fc5c7e5f4@mail.gmail.com>

non works in 3.1 :/

Blazej

2010/1/29 Tim Golden <mail at timgolden.me.uk>:
> On 29/01/2010 08:46, B?a?ej Kaczorowski wrote:
>>
>> Not working, cant import library ... :( there is a way to run xlwt or
>> similar on python 3.1?
>
> If you haven't already, you really want to look over here:
>
> ?http://www.python-excel.org/
>
> and in particular at the mailing list referred to at the bottom
> of the page.
>
> TJG
> _______________________________________________
> python-win32 mailing list
> python-win32 at python.org
> http://mail.python.org/mailman/listinfo/python-win32
>

From waldemar.osuch at gmail.com  Fri Jan 29 18:02:36 2010
From: waldemar.osuch at gmail.com (Waldemar Osuch)
Date: Fri, 29 Jan 2010 10:02:36 -0700
Subject: [python-win32] Documentation for win32com for Worksheets
	Worksheets.Range and Worksheets.Cells
In-Reply-To: <2865bc641001290148k7e9fbe6ekf7d8272fc5c7e5f4@mail.gmail.com>
References: <2865bc641001280330h6a3cd799j3c286a50c7f5b16d@mail.gmail.com>
	<4B61E725.8080100@probo.com>
	<2865bc641001281218t1d3a3a8erd09c7d3f15cd09b4@mail.gmail.com>
	<2865bc641001290030q4644b61h1fe9729682b1c68a@mail.gmail.com>
	<2865bc641001290032n23e75ed4j5190f8ff86311222@mail.gmail.com>
	<2865bc641001290046l539baaeh8780a6a825037442@mail.gmail.com>
	<4B62A4E9.4090703@timgolden.me.uk>
	<2865bc641001290148k7e9fbe6ekf7d8272fc5c7e5f4@mail.gmail.com>
Message-ID: <6fae95541001290902s6e753571y4f4f9b3347fbb316@mail.gmail.com>

2010/1/29 B?a?ej Kaczorowski <mrkaczor at interia.pl>:
> non works in 3.1 :/
>
The maintainer of xlwt, John Machin is thinking about the port to python 3.x
http://groups.google.com/group/python-excel/browse_thread/thread/e3fe92bd08aa1cdd#
but I do not think he is in a rush.

From timr at probo.com  Fri Jan 29 18:39:05 2010
From: timr at probo.com (Tim Roberts)
Date: Fri, 29 Jan 2010 09:39:05 -0800
Subject: [python-win32] Documentation for win32com for Worksheets
 Worksheets.Range and Worksheets.Cells
In-Reply-To: <2865bc641001290030q4644b61h1fe9729682b1c68a@mail.gmail.com>
References: <2865bc641001280330h6a3cd799j3c286a50c7f5b16d@mail.gmail.com>	
	<4B61E725.8080100@probo.com>	
	<2865bc641001281218t1d3a3a8erd09c7d3f15cd09b4@mail.gmail.com>
	<2865bc641001290030q4644b61h1fe9729682b1c68a@mail.gmail.com>
Message-ID: <4B631D39.3040207@probo.com>

B?a?ej Kaczorowski wrote:
> xlwt - yes, but I have python 3.1 in my company and cant downgrade it :(
>   

That is an amazing statement.  I can't imagine any company going
mission-critical with Python 3 yet.  There are just too many third-party
modules that have not been ported.

However, you don't have to downgrade.  Python 2 and Python 3 are able to
coexist on a single computer without any problems.  Just install them both.

-- 
Tim Roberts, timr at probo.com
Providenza & Boekelheide, Inc.


From timr at probo.com  Fri Jan 29 19:33:13 2010
From: timr at probo.com (Tim Roberts)
Date: Fri, 29 Jan 2010 10:33:13 -0800
Subject: [python-win32] Documentation for win32com for Worksheets
 Worksheets.Range and Worksheets.Cells
In-Reply-To: <2865bc641001291019q7adcaff5q176d42b7c829aa61@mail.gmail.com>
References: <2865bc641001280330h6a3cd799j3c286a50c7f5b16d@mail.gmail.com>	
	<4B61E725.8080100@probo.com>	
	<2865bc641001281218t1d3a3a8erd09c7d3f15cd09b4@mail.gmail.com>	
	<2865bc641001290030q4644b61h1fe9729682b1c68a@mail.gmail.com>	
	<4B631D39.3040207@probo.com>
	<2865bc641001291019q7adcaff5q176d42b7c829aa61@mail.gmail.com>
Message-ID: <4B6329E9.3070005@probo.com>

B?a?ej Kaczorowski wrote:
> I know, but I have app that runs with python3 .. and I would need to
> rewrite it to python2 to use xlwt and still win32com works ok (but
> slower) I probably wait for port ...
>   

OK, that's fine.  Just for your information, with appropriate use of
"from __future__" statements, most Python 3 code will run in Python 2.6
(or later).

-- 
Tim Roberts, timr at probo.com
Providenza & Boekelheide, Inc.


From mrkaczor at interia.pl  Fri Jan 29 19:40:28 2010
From: mrkaczor at interia.pl (=?ISO-8859-2?Q?B=B3a=BFej_Kaczorowski?=)
Date: Fri, 29 Jan 2010 19:40:28 +0100
Subject: [python-win32] Documentation for win32com for Worksheets
	Worksheets.Range and Worksheets.Cells
In-Reply-To: <4B6329E9.3070005@probo.com>
References: <2865bc641001280330h6a3cd799j3c286a50c7f5b16d@mail.gmail.com>
	<4B61E725.8080100@probo.com>
	<2865bc641001281218t1d3a3a8erd09c7d3f15cd09b4@mail.gmail.com>
	<2865bc641001290030q4644b61h1fe9729682b1c68a@mail.gmail.com>
	<4B631D39.3040207@probo.com>
	<2865bc641001291019q7adcaff5q176d42b7c829aa61@mail.gmail.com>
	<4B6329E9.3070005@probo.com>
Message-ID: <2865bc641001291040r75528c2cye112b6543caa7963@mail.gmail.com>

probably yes, maybe I will check that in monday :) thx

Blazej

2010/1/29 Tim Roberts <timr at probo.com>:
> B?a?ej Kaczorowski wrote:
>> I know, but I have app that runs with python3 .. and I would need to
>> rewrite it to python2 to use xlwt and still win32com works ok (but
>> slower) I probably wait for port ...
>>
>
> OK, that's fine. ?Just for your information, with appropriate use of
> "from __future__" statements, most Python 3 code will run in Python 2.6
> (or later).
>
> --
> Tim Roberts, timr at probo.com
> Providenza & Boekelheide, Inc.
>
> _______________________________________________
> python-win32 mailing list
> python-win32 at python.org
> http://mail.python.org/mailman/listinfo/python-win32
>

From stef.mientki at gmail.com  Sun Jan 31 00:35:33 2010
From: stef.mientki at gmail.com (Stef Mientki)
Date: Sun, 31 Jan 2010 00:35:33 +0100
Subject: [python-win32] How to fetch windows messages from another
	application ?
Message-ID: <4B64C245.4000104@gmail.com>

hello,

I've a foreign application (written in Delphi),
which I can dock in my wxPython application,
and I'm able to send message from the wxPython application to the 
foreign application.
So I'm able to send my own handle to the foreign application.

Now the last part is receiving messages from the foreign application in 
the main wxPython application.

How do I fetch a windows message from the foreign application ?

thanks,
Stef Mientki
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-win32/attachments/20100131/4d215068/attachment.htm>

From chef at ghum.de  Sun Jan 31 10:15:51 2010
From: chef at ghum.de (Massa, Harald Armin)
Date: Sun, 31 Jan 2010 10:15:51 +0100
Subject: [python-win32] How to fetch windows messages from another
	application ?
In-Reply-To: <4B64C245.4000104@gmail.com>
References: <4B64C245.4000104@gmail.com>
Message-ID: <e3e180dc1001310115t275bbf9p698938858205c6f3@mail.gmail.com>

Stef,

I do not have a readymade solution. But some words to help your googling:
what you are looking for is called "hooks" and "intercepting".

So please google for "intercept windows messages" or "Windows message hooks"
and "windows event hooks".

(As the message "mouse entered window" is also called "event")

Also have a look at tools like auto-it and pywinauto.

Best wishes,

Harald


-- 
GHUM Harald Massa
persuadere et programmare
Harald Armin Massa
Spielberger Stra?e 49
70435 Stuttgart
0173/9409607
no fx, no carrier pigeon
-
%s is too gigantic of an industry to bend to the whims of reality
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-win32/attachments/20100131/b4227aa7/attachment.htm>

From stef.mientki at gmail.com  Sun Jan 31 14:36:31 2010
From: stef.mientki at gmail.com (Stef Mientki)
Date: Sun, 31 Jan 2010 14:36:31 +0100
Subject: [python-win32] How to fetch windows messages from another
 application ?
In-Reply-To: <e3e180dc1001310115t275bbf9p698938858205c6f3@mail.gmail.com>
References: <4B64C245.4000104@gmail.com>
	<e3e180dc1001310115t275bbf9p698938858205c6f3@mail.gmail.com>
Message-ID: <4B65875F.9080202@gmail.com>

Thanks Harald,

Although "message" is a crime to search on google,
with you search words I found a workable solution
If others are interested:
http://blog.csdn.net/dahubaobao/archive/2009/09/10/4539324.aspx

cheers,
Stef

On 31-01-2010 10:15, Massa, Harald Armin wrote:
> Stef,
>
> I do not have a readymade solution. But some words to help your 
> googling: what you are looking for is called "hooks" and "intercepting".
>
> So please google for "intercept windows messages" or "Windows message 
> hooks" and "windows event hooks".
>
> (As the message "mouse entered window" is also called "event")
>
> Also have a look at tools like auto-it and pywinauto.
>
> Best wishes,
>
> Harald
>
>
> -- 
> GHUM Harald Massa
> persuadere et programmare
> Harald Armin Massa
> Spielberger Stra?e 49
> 70435 Stuttgart
> 0173/9409607
> no fx, no carrier pigeon
> -
> %s is too gigantic of an industry to bend to the whims of reality