[issue24252] IDLE removes elements from tracebacks.

[Terry J. Reedy]

Change by Terry J. Reedy :


--
versions: +Python 3.10 -Python 3.6, Python 3.7

___
Python tracker 

___
___
Python-bugs-list mailing list
Unsubscribe: 
https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com

[issue24252] IDLE removes elements from tracebacks.

[Terry J. Reedy]

Terry J. Reedy added the comment:

#26627 was closed as an apparent duplicate of this.  It should be rechecked if 
this is fixed.

--
assignee:  -> terry.reedy
versions: +Python 3.7 -Python 2.7, Python 3.4, Python 3.5

___
Python tracker 

___
___
Python-bugs-list mailing list
Unsubscribe: 
https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com

[issue24252] IDLE removes elements from tracebacks.

[ppperry]

ppperry added the comment:

Your proposed patch does not work in its current form on my IDLE, but it does 
if I change tr[0][:-6] to tr[0][-6:] in the pdb checking code. 

Additionally, my proposed cleanup function did distinguish the difference 
between the unix and windows directory seperators by using os.path.dirname.

--

___
Python tracker rep...@bugs.python.org
http://bugs.python.org/issue24252
___
___
Python-bugs-list mailing list
Unsubscribe: 
https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com

[issue24252] IDLE removes elements from tracebacks.

[ppperry]

ppperry added the comment:

I mean if someone starts a new copy of idle from within the IDLE python shell 
itself by The only situation that this function would behave wrongly is when 
someone launches IDLE from the shell (Why would they do that?).

--

___
Python tracker rep...@bugs.python.org
http://bugs.python.org/issue24252
___
___
Python-bugs-list mailing list
Unsubscribe: 
https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com

[issue24252] IDLE removes elements from tracebacks.

[Terry J. Reedy]

Terry J. Reedy added the comment:

Did you test my proposed patch?  I am reluctant to make major changes in the 
absence of either documentation for why the code is as it is or test cases.  So 
I do not know about trimming top versus bottom and the examples intended to 
show both are needed.  I am not even sure that only top and bottom are 
'correct'.

That said, paying attention to the complete file name, or at least the prior 
directory ('idlelib' for idle files), seems plausible.  But we have to attend 
to '/' versus Windows' '\'.

The only situation that this function would behave wrongly is when someone 
launches IDLE from the shell (Why would they do that?)

Which shell?  I sometimes start Idle from both command line shell and python 
shell, for different reasons.

I was already aware that pdb and Idle might not be the only things to subclass 
bdb (with bugs). I think the patch should be 'inverted' to only add bdb to the 
exclude list if idlelib/RemoteDebugger.py (or ...\...) is present in the 
traceback. 

The current code was moved from Executive.runcode 2003-05-08 by KBK.
2. run.py: move exception printing to toplevel to allow access from main()
A couple of months before, the exclude list was shorter:
   exclude = (run.py, rpc.py, RemoteDebugger.py, bdb.py)
I did not find the reason for adding threading and queue.
A little before, the exception printing code was in rpc.py. I will do more 
history tracing later.

--

___
Python tracker rep...@bugs.python.org
http://bugs.python.org/issue24252
___
___
Python-bugs-list mailing list
Unsubscribe: 
https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com

[issue24252] IDLE removes elements from tracebacks.

[ppperry]

ppperry added the comment:

Another example of this overzealous removing is when you create a module named 
rpc, run, RemoteDebugger, or bdb. 
For example (in this environment, a file in the current directory named rpc.py 
exists and refers to the undefined name bar):
import rpc
Traceback (most recent call last):
  File pyshell#x, line 1, in module
import rpc
NameError: name 'bar' is not defined

The correct traceback is:
Traceback (most recent call last):
  File pyshell#x, line 1, in module
  File rpc.py, line 2, in module
bar
NameError: name 'bar' is not defined

The IDLE debugger also refuses to debug code in the above file (only if it is 
named rpc not any of the other names).

--

___
Python tracker rep...@bugs.python.org
http://bugs.python.org/issue24252
___
___
Python-bugs-list mailing list
Unsubscribe: 
https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com

[issue24252] IDLE removes elements from tracebacks.

[ppperry]

ppperry added the comment:

Is there any reason why the end of the traceback, rather then just the 
beginning, needs to be pruned in the first place?

Additionally, the search for pdb in the tb method will still undesirably 
prune the traceback if someone invents there own buggy debugger that subclasses 
from bdb.Bdb.

Would this cleanup function work?
def cleanup_traceback(tb):
idle_directory = os.path.dirname(__file__)
while os.path.dirname(tb[0][0]) == idle_directory:
del tb[0]
[rest of code not having to do with pruning tracebacks]
The only situation that this function would behave wrongly is when someone 
launches IDLE from the shell (Why would they do that?)

--

___
Python tracker rep...@bugs.python.org
http://bugs.python.org/issue24252
___
___
Python-bugs-list mailing list
Unsubscribe: 
https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com

[issue24252] IDLE removes elements from tracebacks.

[Terry J. Reedy]

Terry J. Reedy added the comment:

Your message arrived as I was just starting on this issue, wondering whether 
the exclusion of threading and queue blocks is proper and wishing for an easy 
test example, such as you provided. It verifies the issue in 3.x.  It also 
answers the question: trimming queue blocks is not acceptible.  (Many threading 
functions also raise various exceptions.) Informative tracebacks are a major 
python feature.  So Idle should be conservative about trimming tracebacks.

The attached 3.4 patch solves queue and threading problems by deleting them 
from the exclude list.  It attempts to solve your pdb problem by conditionally 
removing bdb.py also.  Could you patch your Idle, by hand if necessary, and 
test?

--
keywords: +patch
stage: needs patch - patch review
versions: +Python 3.4, Python 3.5, Python 3.6
Added file: http://bugs.python.org/file39600/tb-trim.diff

___
Python tracker rep...@bugs.python.org
http://bugs.python.org/issue24252
___
___
Python-bugs-list mailing list
Unsubscribe: 
https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com

[issue24252] IDLE removes elements from tracebacks.

[ppperry]

ppperry added the comment:

This problem also occurs in other situati, such as when trying to get items 
from an empty queue
 import Queue
 Q = Queue.Queue()
 q.get_nowait()
Traceback (most recent call last):
  File pyshell#16, line 1, in module
q.get_nowait()
Empty
In that case, it doesn't really matter, though.

--

___
Python tracker rep...@bugs.python.org
http://bugs.python.org/issue24252
___
___
Python-bugs-list mailing list
Unsubscribe: 
https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com

[issue24252] IDLE removes elements from tracebacks.

[Terry J. Reedy]

Terry J. Reedy added the comment:

The two additional entries are the last two.  In this case, they are needed 
because KeyError pertains to the dict lookup in the last line 
self.breaks[filename].

Mistaken in this particular case or not, the deletion is intentional. 
Run.print_exception has these two lines:
exclude = (run.py, rpc.py, threading.py, Queue.py,
   RemoteDebugger.py, bdb.py)
cleanup_traceback(tbe, exclude)
The docstring for cleanup_traceback is Remove excluded traces from 
beginning/end of tb; get cached lines.

The intent of the first 4 excludes is to make the tracebacks displayed in Idle 
for normal (non-debug) code the same as with Python, without Idle.  The intent 
of the last 2 is to continue the equivalence when using the Idle debugger 
(which inherits from bdb.py).

Using pdb with Idle is known to be somewhat problematical because Idle revises 
the execution environment.  In particular, identical tracebacks cannot be 
guaranteed.  (I tried to find the issue bug the buggy tracker search did not 
return the issue.)

A fix for this particular example would be to omit/remove bdb from exclude 
(which could be a set) when the first item contains 'pdb'.  I would consider a 
patch to do that.

--
stage:  - needs patch

___
Python tracker rep...@bugs.python.org
http://bugs.python.org/issue24252
___
___
Python-bugs-list mailing list
Unsubscribe: 
https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com

[issue24252] IDLE removes elements from tracebacks.

[ppperry]

New submission from ppperry:

When reproducing issue24160 in IDLE, I saw this traceback:
Traceback (most recent call last):
  File pyshell#14, line 1, in module
pdb.run(reload(test3))
  File C:\Python27\lib\pdb.py, line 1238, in run
Pdb().run(statement, globals, locals)
  File C:\Python27\lib\bdb.py, line 400, in run
exec cmd in globals, locals
  File string, line 1, in module
  File test3.py, line 1, in module
[contents of test3.py]
  File C:\Python27\lib\bdb.py, line 51, in trace_dispatch
return self.dispatch_call(frame, arg)
  File C:\Python27\lib\bdb.py, line 80, in dispatch_call
self.user_call(frame, arg)
  File C:\Python27\lib\pdb.py, line 148, in user_call
self.interaction(frame, None)
  File C:\Python27\lib\pdb.py, line 210, in interaction
self.cmdloop()
  File C:\Python27\lib\cmd.py, line 142, in cmdloop
stop = self.onecmd(line)
  File C:\Python27\lib\pdb.py, line 279, in onecmd
return cmd.Cmd.onecmd(self, line)
  File C:\Python27\lib\cmd.py, line 221, in onecmd
return func(arg)
  File C:\Python27\lib\pdb.py, line 622, in do_clear
err = self.clear_bpbynumber(i)
KeyError: 'c:\\documents and 
settings\\perry\\desktop\\coding_projects\\python\\test3.py'

Reproducing the same bug in the standard interpreter produces a more detailed 
traceback:
Traceback (most recent call last):
  File pyshell#592, line 1, in module
pdb.run(reload(test3))
  File C:\Python27\lib\pdb.py, line 1238, in run
Pdb().run(statement, globals, locals)
  File C:\Python27\lib\bdb.py, line 400, in run
exec cmd in globals, locals
  File string, line 1, in module
  File test3.py, line 1, in module
foo = 7789
  File C:\Python27\lib\bdb.py, line 51, in trace_dispatch
return self.dispatch_call(frame, arg)
  File C:\Python27\lib\bdb.py, line 80, in dispatch_call
self.user_call(frame, arg)
  File C:\Python27\lib\pdb.py, line 148, in user_call
self.interaction(frame, None)
  File C:\Python27\lib\pdb.py, line 210, in interaction
self.cmdloop()
  File C:\Python27\lib\cmd.py, line 142, in cmdloop
stop = self.onecmd(line)
  File C:\Python27\lib\pdb.py, line 279, in onecmd
return cmd.Cmd.onecmd(self, line)
  File C:\Python27\lib\cmd.py, line 221, in onecmd
return func(arg)
  File C:\Python27\lib\pdb.py, line 622, in do_clear
err = self.clear_bpbynumber(i)
  File C:\Python27\lib\bdb.py, line 297, in clear_bpbynumber
self._prune_breaks(bp.file, bp.line)
  File C:\Python27\lib\bdb.py, line 268, in _prune_breaks
self.breaks[filename].remove(lineno)
KeyError: 'c:\\documents and 
settings\\perry\\desktop\\coding_projects\\python\\test3.py'

The actual error message is irrelevant, other than the fact that in the 
standard interpreter, the traceback contains two more entries.

--
components: IDLE
messages: 243709
nosy: kbk, ppperry, roger.serwy, terry.reedy
priority: normal
severity: normal
status: open
title: IDLE removes elements from tracebacks.
type: behavior
versions: Python 2.7

___
Python tracker rep...@bugs.python.org
http://bugs.python.org/issue24252
___
___
Python-bugs-list mailing list
Unsubscribe: 
https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com