Python pdb bug, followed by bug in bugs.python.org

[donald...@gmail.com]

I am I've developed an application in Python 3.3.1 (on an up-to-date 64-bit Arch Linux system) and am attempting to use pdb to debug it. I am getting incorrect stack traces. I've made up a little 10-line program that illustrates the problem and I attempted to register on the bug-tracker site, unsuccessfully, to file a bug report. I followed the link in the email and got a "broken form" error message when I attempted to log in to the bug-tracker. I then tried replying to the verification email, and got some other form of brokenness in reply.

Suggestions?

[Arnaud Delobelle]

On 9 April 2013 16:25, <donald...@gmail.com> wrote:
> I am I've developed an application in Python 3.3.1 (on an up-to-date 64-bit Arch Linux system) and am attempting to use pdb to debug it. I am getting incorrect stack traces. I've made up a little 10-line program that illustrates the problem and I attempted to register on the bug-tracker site, unsuccessfully, to file a bug report. I followed the link in the email and got a "broken form" error message when I attempted to log in to the bug-tracker. I then tried replying to the verification email, and got some other form of brokenness in reply.
>
> Suggestions?

Python is well known for being a bug-free language, so from your
account I think that the most likely explanation is that you may have
acquired by accident some special mutant powers that introduce bugs
into the software that you come in contact with. Therefore, my
suggestion is that you stay clear of any computer equipment from now
on, as you may cause hazards for yourself and others.

Others may be able to elaborate further.

--
Arnaud

[John Gordon]

In <a34e8251-85df-4486...@googlegroups.com> donald...@gmail.com writes:

> I am I've developed an application in Python 3.3.1 (on an up-to-date 64-bit=
> Arch Linux system) and am attempting to use pdb to debug it. I am getting =
> incorrect stack traces. I've made up a little 10-line program that illustra=
> tes the problem and I attempted to register on the bug-tracker site, unsucc=
> essfully, to file a bug report. I followed the link in the email and got a =
> "broken form" error message when I attempted to log in to the bug-tracker. =
> I then tried replying to the verification email, and got some other form of=
> brokenness in reply.

> Suggestions?

Post the 10-line program here, so others can verify whether it is a bug.

--
John Gordon A is for Amy, who fell down the stairs
gor...@panix.com B is for Basil, assaulted by bears
-- Edward Gorey, "The Gashlycrumb Tinies"

[Ned Deily]

In article <a34e8251-85df-4486...@googlegroups.com>,

I was just able to create a new user by through the link from the
bugs.python.org web page and then clicking on the link in the confirming
email (although it did take a couple of hours for the confirming email
to arrive). There is a meta tracker for problems with the Python
issuer tracker itself:

http://psf.upfronthosting.co.za/roundup/meta/

but you do have to register for that tracker (a separate registration).
If you are able to supply more details, we might be able to follow up on
the registration problem. And, as someone else suggested, you could
post the details of the pdb problem here. Note, there are already a
number of currently open issues with pdb reported on the bug tracker.
If you haven't already, you could search for "pdb" and see if your
problem has been reported. Thanks for bringing the problem(s) up!

--
Ned Deily,
n...@acm.org

[Gregory Ewing]

Ned Deily wrote:
> There is a meta tracker for problems with the Python
> issuer tracker itself:
>
> http://psf.upfronthosting.co.za/roundup/meta/
>
> but you do have to register for that tracker (a separate registration).

You mean there's no meta-meta-tracker for reporting
problems registering with the meta-tracker?

There should be meta-trackers all the way up!

--
Greg

[Ned Deily]

In article <asm8lh...@mid.individual.net>,

Alas, as is well known, Python does not support tail call elimination.

--
Ned Deily,
n...@acm.org

[donald...@gmail.com]

> > Suggestions?
>
>
>
> Post the 10-line program here, so others can verify whether it is a bug.

#! /usr/bin/env python3
import pdb
def foo(message):
print(message)
pdb.set_trace()
foo('first call')
foo('second call')

Stick this in an file with execute permission and run it. At the first breakpoint, the backtrace will be correct. Continue. At the second breakpoint, a backtrace will show the foo('first call') on the stack when, in fact, the call came from foo('second call'), as verified by the printed message.

[donald...@gmail.com]

On Tuesday, April 9, 2013 6:17:28 PM UTC-4, Ned Deily wrote:
> In article <a34e8251-85df-4486...@googlegroups.com>,

I just got registered successfully using a different username than I tried the other day. I haven't attempted to reproduce the problem I saw then, but I will and will report it if I can provide a proper description of the issue.

I just submitted a bug report on the pdb issue.

Thanks for the helpful reply.

/Don

>
>
>
> --
>
> Ned Deily,
>
> n...@acm.org

[donald...@gmail.com]

On Tuesday, April 9, 2013 6:17:28 PM UTC-4, Ned Deily wrote:
> In article <a34e8251-85df-4486...@googlegroups.com>,
>

[Ian Kelly]

On Thu, Apr 11, 2013 at 8:56 AM, <donald...@gmail.com> wrote:
> #! /usr/bin/env python3
> import pdb
> def foo(message):
> print(message)
> pdb.set_trace()
> foo('first call')
> foo('second call')
>
> Stick this in an file with execute permission and run it. At the first breakpoint, the backtrace will be correct. Continue. At the second breakpoint, a backtrace will show the foo('first call') on the stack when, in fact, the call came from foo('second call'), as verified by the printed message.

This is what I get using Python 3.3.1 in Windows:

C:\Users\ikelly\Desktop>c:\python33\python python_bug.py
first call
--Return--
> c:\users\ikelly\desktop\python_bug.py(7)foo()->None
-> pdb.set_trace()
(Pdb) c
second call
--Return--
> c:\users\ikelly\desktop\python_bug.py(7)foo()->None
-> pdb.set_trace()
(Pdb) c

[Robert Kern]

On 2013-04-11 23:11, Ian Kelly wrote:
> On Thu, Apr 11, 2013 at 8:56 AM, <donald...@gmail.com> wrote:

>> #! /usr/bin/env python3
>> import pdb
>> def foo(message):
>> print(message)
>> pdb.set_trace()
>> foo('first call')
>> foo('second call')
>>
>> Stick this in an file with execute permission and run it. At the first breakpoint, the backtrace will be correct. Continue. At the second breakpoint, a backtrace will show the foo('first call') on the stack when, in fact, the call came from foo('second call'), as verified by the printed message.
>
>

> This is what I get using Python 3.3.1 in Windows:
>
> C:\Users\ikelly\Desktop>c:\python33\python python_bug.py
> first call
> --Return--
>> c:\users\ikelly\desktop\python_bug.py(7)foo()->None
> -> pdb.set_trace()
> (Pdb) c
> second call
> --Return--
>> c:\users\ikelly\desktop\python_bug.py(7)foo()->None
> -> pdb.set_trace()
> (Pdb) c

Use `where` to see the problem:

[~/scratch]$ python3.3 pdbbug.py
first call
--Return--
> /Users/rkern/scratch/pdbbug.py(4)foo()->None
-> pdb.set_trace()
(Pdb) where
/Users/rkern/scratch/pdbbug.py(5)<module>()
-> foo('first call')
> /Users/rkern/scratch/pdbbug.py(4)foo()->None

-> pdb.set_trace()
(Pdb) c
second call
--Return--

> /Users/rkern/scratch/pdbbug.py(4)foo()->None
-> pdb.set_trace()
(Pdb) where
/Users/rkern/scratch/pdbbug.py(5)<module>()
-> foo('first call')
> /Users/rkern/scratch/pdbbug.py(4)foo()->None
-> pdb.set_trace()
(Pdb)

--
Robert Kern

"I have come to believe that the whole world is an enigma, a harmless enigma
that is made terrible by our own mad attempt to interpret it as though it had
an underlying truth."
-- Umberto Eco

[Ian Kelly]

On Thu, Apr 11, 2013 at 12:42 PM, Robert Kern <rober...@gmail.com> wrote:
> Use `where` to see the problem:

Ah. Then I can verify that the problem occurs in Windows as well:

C:\Users\ikelly\Desktop>c:\python33\python python_bug.py
first call
--Return--
> c:\users\ikelly\desktop\python_bug.py(7)foo()->None
-> pdb.set_trace()

(Pdb) w
c:\users\ikelly\desktop\python_bug.py(9)<module>()

-> foo('first call')

> c:\users\ikelly\desktop\python_bug.py(7)foo()->None
-> pdb.set_trace()
(Pdb) c
second call
--Return--
> c:\users\ikelly\desktop\python_bug.py(7)foo()->None
-> pdb.set_trace()

(Pdb) w
c:\users\ikelly\desktop\python_bug.py(9)<module>()

-> foo('first call')

[Piotr Dobrogost]

On Thursday, April 11, 2013 5:12:53 PM UTC+2, donald...@gmail.com wrote:
>
> I just submitted a bug report on the pdb issue.

It would be nice of you to share the link to this issue.

[Ned Deily]

In article <eba316f9-cf50-418f...@googlegroups.com>,

http://bugs.python.org/issue17697

--
Ned Deily,
n...@acm.org

[Piotr Dobrogost]

On Thursday, April 11, 2013 5:12:53 PM UTC+2, donald...@gmail.com wrote:
>

> I just submitted a bug report on the pdb issue.