From taleinat at gmail.com  Thu Sep  7 13:57:18 2006
From: taleinat at gmail.com (Tal Einat)
Date: Thu, 7 Sep 2006 14:57:18 +0300
Subject: [Idle-dev] IDLE debugger
Message-ID: <7afdee2f0609070457ib8c373bg2ba129ba7b64eca@mail.gmail.com>

On Mar 5 2002, Edward K. Ream posted this on c.l.py:

> I am beginning to think that IDLE's debugger suffers from a vicious
> cycle:
>
> 1. Nobody uses IDLE's debugger, so
> 2. Nobody complains about its glaring deficiencies, so
> 3. Nothing gets fixed, so (back to one)
>
I think this is still true. I use IDLE all the time, but when I need to
debug I just use pdb. This is silly, GUI can make debugging so much better!
I think this is the #1 lacking feature in IDLE today, and I would like to
get to work at this.

How do you guys feel? Is anyone out there using IDLE's debugger? I'd love to
hear what people like / dislike about it.

I haven't much experience with the newer IDEs out there (for .NET, Java,
...), some of which I have heard have good graphical debuggers. I'd really
love to hear what people like (or dislike) about other debuggers, to see if
such features (or better ones) can be implemented for IDLE.

- Tal
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/idle-dev/attachments/20060907/0ff829d1/attachment.htm 

From frikker at gmail.com  Thu Sep  7 17:09:45 2006
From: frikker at gmail.com (Blaine Booher)
Date: Thu, 7 Sep 2006 11:09:45 -0400
Subject: [Idle-dev] IDLE debugger
In-Reply-To: <7afdee2f0609070457ib8c373bg2ba129ba7b64eca@mail.gmail.com>
References: <7afdee2f0609070457ib8c373bg2ba129ba7b64eca@mail.gmail.com>
Message-ID: <9a3d5a1b0609070809n3c51f1bas1e3752eed98aa393@mail.gmail.com>

I'll say this: I use IDLE exclusively, but I have not looked into any
"advanced features".  I have always been very minimal when it comes to
debugging, so pdb is a saving grace to me.  I wasn't aware there was a
debugger in IDLE, but as I said I haven't looked too hard.  What kind of
deficiencies does it have?  Does it use pdb or is it completely seperate?

Blaine


On 9/7/06, Tal Einat <taleinat at gmail.com> wrote:
>
>  On Mar 5 2002, Edward K. Ream posted this on c.l.py:
>
> > I am beginning to think that IDLE's debugger suffers from a vicious
> > cycle:
> >
> > 1. Nobody uses IDLE's debugger, so
> > 2. Nobody complains about its glaring deficiencies, so
> > 3. Nothing gets fixed, so (back to one)
> >
> I think this is still true. I use IDLE all the time, but when I need to
> debug I just use pdb. This is silly, GUI can make debugging so much better!
> I think this is the #1 lacking feature in IDLE today, and I would like to
> get to work at this.
>
> How do you guys feel? Is anyone out there using IDLE's debugger? I'd love
> to hear what people like / dislike about it.
>
> I haven't much experience with the newer IDEs out there (for .NET, Java,
> ...), some of which I have heard have good graphical debuggers. I'd really
> love to hear what people like (or dislike) about other debuggers, to see if
> such features (or better ones) can be implemented for IDLE.
>
> - Tal
>
> _______________________________________________
> IDLE-dev mailing list
> IDLE-dev at python.org
> http://mail.python.org/mailman/listinfo/idle-dev
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/idle-dev/attachments/20060907/b54274e8/attachment.html 

From faramir2 at gazeta.pl  Tue Sep 19 13:45:57 2006
From: faramir2 at gazeta.pl (Marek Nowicki)
Date: Tue, 19 Sep 2006 13:45:57 +0200
Subject: [Idle-dev] IDLE Hung up after open script by command line
Message-ID: <op.tf4iyvolb4ynqs@buster.mshome.net>

Hello,

I wrote that code in python and saved as prx.py:
--- CUT ---
 from BaseHTTPServer import HTTPServer, BaseHTTPRequestHandler
 from time import strftime, gmtime
import urllib2
import thread
 from sys import stdout
class RequestHandler(BaseHTTPRequestHandler):
     def serve(self):
         print "%s %s %s\r\n%s" % (self.command, self.path,  
self.request_version, self.headers)
         header={}
         header["content-length"]=0
         for i in str(self.headers).split("\r\n"):
             j=i.split(":", 1)
             if len(j)==2:
                 header[j[0].strip().lower()] = j[1].strip()
         content=self.rfile.read(int(header["content-length"]))
         print content
         url="http://faramir2.prv.pl"
         u=urllib2.urlopen(url)
         for i,j in u.info().items():
             print "%s: %s" % (i,j)
         self.server_version = "Apache"
         self.sys_version = ""
         self.send_response(200)
         self.send_header("Content-type", "text/html; charset=ISO-8859-2")
         self.send_header("Connectin", "close")
         self.end_headers()
     def do_POST(self): self.serve()
     def do_HEAD(self): self.serve()
     def do_GET(self): self.serve()
# from sys import exit
# exit(1)
address = ("", 80)
server = HTTPServer(address, RequestHandler)
thread.start_new_thread(server.serve_forever, () )
--- CUT ---

When I right click on that file and select "Edit with IDLE" it opens. Then  
when I push F5 the script is running. *Python Shell* is restarting. But  
when I try to connect by browser to http://localhost:80/ IDLE Hung-up. I  
don't see that hung ups when I open IDLE from shortcut and then in IDLE  
open file prx.py and run it works normally - good. IDLE does't hung up.

I don't know why it works like that, but I think that it's bug..

Python version: 2.5c2
Tk version: 8.4
IDLE version: 1.2c2

regards,
Marek



From ronaldoussoren at mac.com  Wed Sep 20 07:58:56 2006
From: ronaldoussoren at mac.com (Ronald Oussoren)
Date: Wed, 20 Sep 2006 07:58:56 +0200
Subject: [Idle-dev] IDLE Hung up after open script by command line
In-Reply-To: <op.tf4iyvolb4ynqs@buster.mshome.net>
References: <op.tf4iyvolb4ynqs@buster.mshome.net>
Message-ID: <5CCCD2DF-0395-4CFD-B55A-59D77265663C@mac.com>

Hi,

Thanks for the detailed bug report, although to nit-pick it would  
have been better to file the report at http://www.sf.net/projects/ 
python ;-)

I cannot reproduce your problem with Python 2.5 (final) and OSX 10.4.  
IDLE keeps responsive throughout, although it does take a while  
before the output of your script actually appears in the *Python  
Shell* window.

What OS are you using?

Ronald

-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/pkcs7-signature
Size: 3562 bytes
Desc: not available
Url : http://mail.python.org/pipermail/idle-dev/attachments/20060920/38271ee5/attachment.bin 

From faramir2 at gazeta.pl  Wed Sep 20 13:01:44 2006
From: faramir2 at gazeta.pl (Marek Nowicki)
Date: Wed, 20 Sep 2006 13:01:44 +0200
Subject: [Idle-dev] IDLE Hung up after open script by command line
In-Reply-To: <5CCCD2DF-0395-4CFD-B55A-59D77265663C@mac.com>
References: <op.tf4iyvolb4ynqs@buster.mshome.net>
	<5CCCD2DF-0395-4CFD-B55A-59D77265663C@mac.com>
Message-ID: <op.tf6bk7apb4ynqs@buster.mshome.net>

Dnia 20-09-2006 o 07:58:56 Ronald Oussoren <ronaldoussoren at mac.com>  
napisa?:
> Thanks for the detailed bug report, although to nit-pick it would
> have been better to file the report at http://www.sf.net/projects/
> python ;-)

I think that this isn't problem with python but with IDLE and I think that  
IDLE is independent project.

> I cannot reproduce your problem with Python 2.5 (final) and OSX 10.4.
> IDLE keeps responsive throughout, although it does take a while
> before the output of your script actually appears in the *Python
> Shell* window.
>
> What OS are you using?

Microsoft Windows XP Professional with ServicePack 2.

Marek



From ronaldoussoren at mac.com  Wed Sep 20 13:53:45 2006
From: ronaldoussoren at mac.com (Ronald Oussoren)
Date: Wed, 20 Sep 2006 13:53:45 +0200
Subject: [Idle-dev] IDLE Hung up after open script by command line
In-Reply-To: <op.tf6bk7apb4ynqs@buster.mshome.net>
References: <op.tf4iyvolb4ynqs@buster.mshome.net>
	<5CCCD2DF-0395-4CFD-B55A-59D77265663C@mac.com>
	<op.tf6bk7apb4ynqs@buster.mshome.net>
Message-ID: <7883070.1158753225497.JavaMail.ronaldoussoren@mac.com>

 
On Wednesday, September 20, 2006, at 01:18PM, Marek Nowicki <faramir2 at gazeta.pl> wrote:

>Dnia 20-09-2006 o 07:58:56 Ronald Oussoren <ronaldoussoren at mac.com>  
>napisa?:
>> Thanks for the detailed bug report, although to nit-pick it would
>> have been better to file the report at http://www.sf.net/projects/
>> python ;-)
>
>I think that this isn't problem with python but with IDLE and I think that  
>IDLE is independent project.

But IDLE does use the python bug tracker and is hosted in the python repository. Having a bugreport in the tracker keeps us from forgetting that issues exist.

>
>> I cannot reproduce your problem with Python 2.5 (final) and OSX 10.4.
>> IDLE keeps responsive throughout, although it does take a while
>> before the output of your script actually appears in the *Python
>> Shell* window.
>>
>> What OS are you using?
>
>Microsoft Windows XP Professional with ServicePack 2.

I was afraid you were going to say that. Hopefully someone that uses windows will drop by, I try to stay away from windows as much as possible.

Ronald


From faramir2 at gazeta.pl  Wed Sep 20 15:14:42 2006
From: faramir2 at gazeta.pl (Marek Nowicki)
Date: Wed, 20 Sep 2006 15:14:42 +0200
Subject: [Idle-dev] IDLE Hung up after open script by command line
In-Reply-To: <7883070.1158753225497.JavaMail.ronaldoussoren@mac.com>
References: <op.tf4iyvolb4ynqs@buster.mshome.net>
	<5CCCD2DF-0395-4CFD-B55A-59D77265663C@mac.com>
	<op.tf6bk7apb4ynqs@buster.mshome.net>
	<7883070.1158753225497.JavaMail.ronaldoussoren@mac.com>
Message-ID: <op.tf6hqsxab4ynqs@buster.mshome.net>

Dnia 20-09-2006 o 13:53:45 Ronald Oussoren <ronaldoussoren at mac.com>  
napisa?:
>>> Thanks for the detailed bug report, although to nit-pick it would
>>> have been better to file the report at http://www.sf.net/projects/
>>> python ;-)
>>
>> I think that this isn't problem with python but with IDLE and I think  
>> that
>> IDLE is independent project.
>
> But IDLE does use the python bug tracker and is hosted in the python  
> repository. Having a bugreport in the tracker keeps us from forgetting  
> that issues exist.

Ok.. thanks for informations. I just posted bugreport.

>>> I cannot reproduce your problem with Python 2.5 (final) and OSX 10.4.
>>> IDLE keeps responsive throughout, although it does take a while
>>> before the output of your script actually appears in the *Python
>>> Shell* window.
>>>
>>> What OS are you using?
>>
>> Microsoft Windows XP Professional with ServicePack 2.
>
> I was afraid you were going to say that. Hopefully someone that uses  
> windows will drop by, I try to stay away from windows as much as  
> possible.

Windows isn't so bad as you think :P... but ok.. I don't use MacOS :)

thanks for help,
Marek



From taleinat at gmail.com  Wed Sep 20 21:43:59 2006
From: taleinat at gmail.com (Tal Einat)
Date: Wed, 20 Sep 2006 15:43:59 -0400
Subject: [Idle-dev] IDLE Hung up after open script by command line
In-Reply-To: <op.tf4iyvolb4ynqs@buster.mshome.net>
References: <op.tf4iyvolb4ynqs@buster.mshome.net>
Message-ID: <7afdee2f0609201243g5463143fl7db31cb713ef5df9@mail.gmail.com>

Hi all, another WinXP SP2 user here.

First of all, this should be reproducible on any platform by running IDLE
with the "-n" flag.

On windows, the Python installer adds a "Edit with IDLE" context menu entry.
The OP mentioned that the problem only occurred when he opened the file in
this method and ran it with F5. Currently, the "Edit with IDLE" option opens
IDLE *without* a subprocess (with the "-n" flag), while the IDLE shortcut
runs IDLE *with* a subprocess (without the "-n" flag). In general it is much
better to run IDLE with a subprocess.

In this case what's happening is that the Python interpreter is constantly
busy, and (almost) never gets around to running IDLE. This is because IDLE
and the shell are actually running in the same process. On the other hand,
having IDLE and the shell run in separate processes allows IDLE to continue
functioning even while the shell is blocking or busy.

IDLE isn't broken, on the contrary! In IDLE the shell can be separated from
the GUI, which is a great feature, not found in most other Python
interpreters.

- Tal

P.S. The reason "Edit with IDLE" opens IDLE without a subprocess is that in
the current version there can only be one instance of IDLE with a
subprocess. I have submitted a patch to workaround this, and which allows
multiple version of IDLE to run in parallel, each with a subprocess of its
own. The patch must be more thoroughly tested and tweaked, but IMO is worth
giving a try.


On 9/19/06, Marek Nowicki <faramir2 at gazeta.pl> wrote:
>
> Hello,
>
> I wrote that code in python and saved as prx.py:
> --- CUT ---
> from BaseHTTPServer import HTTPServer, BaseHTTPRequestHandler
> from time import strftime, gmtime
> import urllib2
> import thread
> from sys import stdout
> class RequestHandler(BaseHTTPRequestHandler):
>     def serve(self):
>         print "%s %s %s\r\n%s" % (self.command, self.path,
> self.request_version, self.headers)
>         header={}
>         header["content-length"]=0
>         for i in str(self.headers).split("\r\n"):
>             j=i.split(":", 1)
>             if len(j)==2:
>                 header[j[0].strip().lower()] = j[1].strip()
>         content=self.rfile.read(int(header["content-length"]))
>         print content
>         url="http://faramir2.prv.pl"
>         u=urllib2.urlopen(url)
>         for i,j in u.info().items():
>             print "%s: %s" % (i,j)
>         self.server_version = "Apache"
>         self.sys_version = ""
>         self.send_response(200)
>         self.send_header("Content-type", "text/html; charset=ISO-8859-2")
>         self.send_header("Connectin", "close")
>         self.end_headers()
>     def do_POST(self): self.serve()
>     def do_HEAD(self): self.serve()
>     def do_GET(self): self.serve()
> # from sys import exit
> # exit(1)
> address = ("", 80)
> server = HTTPServer(address, RequestHandler)
> thread.start_new_thread(server.serve_forever, () )
> --- CUT ---
>
> When I right click on that file and select "Edit with IDLE" it opens. Then
> when I push F5 the script is running. *Python Shell* is restarting. But
> when I try to connect by browser to http://localhost:80/ IDLE Hung-up. I
> don't see that hung ups when I open IDLE from shortcut and then in IDLE
> open file prx.py and run it works normally - good. IDLE does't hung up.
>
> I don't know why it works like that, but I think that it's bug..
>
> Python version: 2.5c2
> Tk version: 8.4
> IDLE version: 1.2c2
>
> regards,
> Marek
>
>
> _______________________________________________
> IDLE-dev mailing list
> IDLE-dev at python.org
> http://mail.python.org/mailman/listinfo/idle-dev
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/idle-dev/attachments/20060920/f0974846/attachment.html 

From ronaldoussoren at mac.com  Wed Sep 20 22:13:49 2006
From: ronaldoussoren at mac.com (Ronald Oussoren)
Date: Wed, 20 Sep 2006 22:13:49 +0200
Subject: [Idle-dev] IDLE Hung up after open script by command line
In-Reply-To: <7afdee2f0609201243g5463143fl7db31cb713ef5df9@mail.gmail.com>
References: <op.tf4iyvolb4ynqs@buster.mshome.net>
	<7afdee2f0609201243g5463143fl7db31cb713ef5df9@mail.gmail.com>
Message-ID: <203247F1-D498-4B5C-8DBE-2566DE74E885@mac.com>


On Sep 20, 2006, at 9:43 PM, Tal Einat wrote:
>
>
> P.S. The reason "Edit with IDLE" opens IDLE without a subprocess is  
> that in the current version there can only be one instance of IDLE  
> with a subprocess. I have submitted a patch to workaround this, and  
> which allows multiple version of IDLE to run in parallel, each with  
> a subprocess of its own. The patch must be more thoroughly tested  
> and tweaked, but IMO is worth giving a try.

Is that really necessary? I've added 'editor with IDLE' functionality  
to IDLE on OSX and that just opens all files you open in the same  
copy of IDLE. IMHO that is more useful that starting several copies  
of IDLE when you open multiple files.

The disadvantage is that you can have only one shell window open.  
Your patch could be used to fix that (given your description of the  
patch, I haven't looked at the actual patch).

Ronald

-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/pkcs7-signature
Size: 3562 bytes
Desc: not available
Url : http://mail.python.org/pipermail/idle-dev/attachments/20060920/ab568534/attachment.bin 

From ronaldoussoren at mac.com  Wed Sep 20 22:49:48 2006
From: ronaldoussoren at mac.com (Ronald Oussoren)
Date: Wed, 20 Sep 2006 22:49:48 +0200
Subject: [Idle-dev] IDLE Hung up after open script by command line
In-Reply-To: <200609201527.46695.john.zelle@wartburg.edu>
References: <op.tf4iyvolb4ynqs@buster.mshome.net>
	<7afdee2f0609201243g5463143fl7db31cb713ef5df9@mail.gmail.com>
	<203247F1-D498-4B5C-8DBE-2566DE74E885@mac.com>
	<200609201527.46695.john.zelle@wartburg.edu>
Message-ID: <FAF2D985-0B9A-476F-8D2F-456C1E194176@mac.com>


On Sep 20, 2006, at 10:27 PM, John Zelle wrote:

> To me, the issue is not so much a single user editing multiple  
> files, but
> rather the situation when you have a shared machine with multiple
> (simultaneous) users, as is frequently the case in a lab  
> environment. With a
> fixed port, remote and local users cannot both be running IDLE. If  
> this patch
> addresses that, then I think it should be given high priority.

That sounds like a good reason for using this patch, applying it  
because you can then open multiple files using 'Edit with IDLE' is  
IMHO not because it fixes the wrong issue.

If added a comment to patch #1201569 (which isn't Tal's patch but  
something simular), it seems that patch was rejected for now because  
of vague socket-rebinding problems on Windows, which may be caused by  
odd behaviour of windows (see my comment on the patch for more  
information)

Ronald
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/pkcs7-signature
Size: 3562 bytes
Desc: not available
Url : http://mail.python.org/pipermail/idle-dev/attachments/20060920/fadb3ecc/attachment-0001.bin 

From john.zelle at wartburg.edu  Wed Sep 20 22:27:46 2006
From: john.zelle at wartburg.edu (John Zelle)
Date: Wed, 20 Sep 2006 15:27:46 -0500
Subject: [Idle-dev] IDLE Hung up after open script by command line
In-Reply-To: <203247F1-D498-4B5C-8DBE-2566DE74E885@mac.com>
References: <op.tf4iyvolb4ynqs@buster.mshome.net>
	<7afdee2f0609201243g5463143fl7db31cb713ef5df9@mail.gmail.com>
	<203247F1-D498-4B5C-8DBE-2566DE74E885@mac.com>
Message-ID: <200609201527.46695.john.zelle@wartburg.edu>

To me, the issue is not so much a single user editing multiple files, but 
rather the situation when you have a shared machine with multiple 
(simultaneous) users, as is frequently the case in a lab environment. With a 
fixed port, remote and local users cannot both be running IDLE. If this patch 
addresses that, then I think it should be given high priority.

--John

On Wednesday 20 September 2006 3:13 pm, Ronald Oussoren wrote:
> On Sep 20, 2006, at 9:43 PM, Tal Einat wrote:
> > P.S. The reason "Edit with IDLE" opens IDLE without a subprocess is
> > that in the current version there can only be one instance of IDLE
> > with a subprocess. I have submitted a patch to workaround this, and
> > which allows multiple version of IDLE to run in parallel, each with
> > a subprocess of its own. The patch must be more thoroughly tested
> > and tweaked, but IMO is worth giving a try.
>
> Is that really necessary? I've added 'editor with IDLE' functionality
> to IDLE on OSX and that just opens all files you open in the same
> copy of IDLE. IMHO that is more useful that starting several copies
> of IDLE when you open multiple files.
>
> The disadvantage is that you can have only one shell window open.
> Your patch could be used to fix that (given your description of the
> patch, I haven't looked at the actual patch).
>
> Ronald

-- 
John M. Zelle, Ph.D.             Wartburg College
Professor of Computer Science    Waverly, IA     
john.zelle at wartburg.edu          (319) 352-8360  

From mackay at aims.ac.za  Wed Sep 27 17:56:20 2006
From: mackay at aims.ac.za (David MacKay)
Date: Wed, 27 Sep 2006 17:56:20 +0200 (SAST)
Subject: [Idle-dev] C-k and C-y keybinding
Message-ID: <20060927155620.F413612A960@localhost.localdomain>

Hello, a newbie question - 
I have selected the "unix-like" keybinding. 
I love to move or duplicate text by 
using C-k (C-k C-k ...)
then using C-y (C-y).

But in IDLE with unix-like keybinding, 
C-k successfully kills the line, but does not put it into 
the cut-buffer, so C-y does not retrieve the killed object.

How can I modify IDLE to that C-k is the correct "kill and 
put in the cut/paste buffer"?

-- And another question:   I see that "C-w" is "cut"; 
but how can I create and extend the region to be cut, using 
just the keyboard?  In emacs, we use C-SPACE then C-w 
to set mark and cut region.   Is there an equivalent to 
C-SPACE in idle? 

Thanks very much

David  [using idle-python2.4 on an ubuntu machine running X windows]