3.5 (a2) traceback regression snarls Idle

[terryjreedy]

BPO	23631
Nosy	@gvanrossum, @birkenfeld, @terryjreedy, @ncoghlan, @larryhastings, @rbtcollins
Files	issue-23631-1.patch

^{Note: these values reflect the state of the issue at the time it was migrated and might not reflect the current state.}

Show more details

GitHub fields:

assignee = None
closed_at = <Date 2015-03-17.20:34:25.061>
created_at = <Date 2015-03-10.21:25:28.627>
labels = ['type-bug', 'release-blocker']
title = '3.5 (a2) traceback regression snarls Idle'
updated_at = <Date 2015-03-17.21:07:30.988>
user = 'https://github.com/terryjreedy'

bugs.python.org fields:

activity = <Date 2015-03-17.21:07:30.988>
actor = 'berker.peksag'
assignee = 'none'
closed = True
closed_date = <Date 2015-03-17.20:34:25.061>
closer = 'rbcollins'
components = []
creation = <Date 2015-03-10.21:25:28.627>
creator = 'terry.reedy'
dependencies = []
files = ['38503']
hgrepos = []
issue_num = 23631
keywords = ['patch', '3.3regression']
message_count = 8.0
messages = ['237810', '237836', '238170', '238171', '238172', '238173', '238177', '238336']
nosy_count = 7.0
nosy_names = ['gvanrossum', 'georg.brandl', 'terry.reedy', 'ncoghlan', 'larry', 'rbcollins', 'python-dev']
pr_nums = []
priority = 'release blocker'
resolution = 'fixed'
stage = 'resolved'
status = 'closed'
superseder = None
type = 'behavior'
url = 'https://bugs.python.org/issue23631'
versions = ['Python 3.5']

[terryjreedy]

3.5.0a2, my Win 7 and Guido's new Win laptop: error tracebacks are mangled in a way that makes Idle severely less functional, hence 'release blocker'. First, the normal behavior.

Python 3.5.0a2 (v3.5.0a2:0337bd7ebcb6+, Mar  9 2015, 10:29:45) [MSC v.1900 64 bi
t (AMD64)] on win32
>>> 1/0
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
ZeroDivisionError: division by zero

Idle 3.4.3 and previous give same, with the helpful addition of the code line ('1/0'), as in batch mode tracebacks, since it uses the traceback module. But Idle 3.5.0a2 started from the installed icon gives

>>> 1/0
Traceback (most recent call last):
Traceback (most recent call last):
Traceback (most recent call last):
  File "C:\Program Files\Python 3.5\lib\idlelib\run.py", line 353, in runcode
    exec(code, self.locals)
  File "<pyshell#1>", line 1, in <module>
ZeroDivisionError: division by zero

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "C:\Program Files\Python 3.5\lib\idlelib\run.py", line 126, in main
    ret = method(*args, **kwargs)
  File "C:\Program Files\Python 3.5\lib\idlelib\run.py", line 365, in runcode
    print_exception()
  File "C:\Program Files\Python 3.5\lib\idlelib\run.py", line 216, in print_exception
    print_exc(typ, val, tb)
  File "C:\Program Files\Python 3.5\lib\idlelib\run.py", line 211, in print_exc
    traceback.print_list(tbe, file=efile)
  File "C:\Program Files\Python 3.5\lib\traceback.py", line 22, in print_list
    for item in StackSummary.from_list(extracted_list).format():
  File "C:\Program Files\Python 3.5\lib\traceback.py", line 370, in format
    frame.filename, frame.lineno, frame.name))
AttributeError: 'tuple' object has no attribute 'filename'

>> ================================ RESTART ================================
>>

run.py (as should be more or less everything in idlelib) is identical in 3.4 and 3.5. Idle started from a command line (py -3 -m idlelib) gives an even worse result. The Idle Shell just shows

>>> 1/0
Traceback (most recent call last):
Traceback (most recent call last):

>> ================================ RESTART ================================
>>

while the following appears in the command window (Command Prompt or PowerShell).

Traceback (most recent call last):
  File "C:\Program Files\Python 3.5\lib\idlelib\run.py", line 353, in runcode
    exec(code, self.locals)
  File "<pyshell#0>", line 1, in <module>
ZeroDivisionError: division by zero

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "C:\Program Files\Python 3.5\lib\idlelib\run.py", line 126, in main
    ret = method(*args, **kwargs)
  File "C:\Program Files\Python 3.5\lib\idlelib\run.py", line 365, in runcode
    print_exception()
  File "C:\Program Files\Python 3.5\lib\idlelib\run.py", line 216, in print_exception
    print_exc(typ, val, tb)
  File "C:\Program Files\Python 3.5\lib\idlelib\run.py", line 211, in print_exc
    traceback.print_list(tbe, file=efile)
  File "C:\Program Files\Python 3.5\lib\traceback.py", line 22, in print_list
    for item in StackSummary.from_list(extracted_list).format():
  File "C:\Program Files\Python 3.5\lib\traceback.py", line 370, in format
    frame.filename, frame.lineno, frame.name))
AttributeError: 'tuple' object has no attribute 'filename'
Exception in Tkinter callback
Traceback (most recent call last):
  File "C:\Program Files\Python 3.5\lib\idlelib\rpc.py", line 359, in pollpacket
    s = self.sock.recv(BUFSIZE)
ConnectionResetError: [WinError 10054] An existing connection was forcibly closed by the r

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "C:\Program Files\Python 3.5\lib\idlelib\rpc.py", line 432, in pollresponse
    message = self.pollmessage(wait)
  File "C:\Program Files\Python 3.5\lib\idlelib\rpc.py", line 384, in pollmessage
    packet = self.pollpacket(wait)
  File "C:\Program Files\Python 3.5\lib\idlelib\rpc.py", line 361, in pollpacket
    raise EOFError
EOFError

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "C:\Program Files\Python 3.5\lib\idlelib\PyShell.py", line 565, in poll_subprocess
    response = clt.pollresponse(self.active_seq, wait=0.05)
  File "C:\Program Files\Python 3.5\lib\idlelib\rpc.py", line 436, in pollresponse
    self.handle_EOF()
  File "C:\Program Files\Python 3.5\lib\idlelib\PyShell.py", line 383, in handle_EOF
    raise EOFError
EOFError

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
ValueError: invalid literal for int() with base 10: '??'

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "C:\Program Files\Python 3.5\lib\tkinter\__init__.py", line 1534, in __call__
    args = self.subst(*args)
  File "C:\Program Files\Python 3.5\lib\tkinter\__init__.py", line 1252, in _substitute
    e.num = getint_event(b)
  File "C:\Program Files\Python 3.5\lib\tkinter\__init__.py", line 1231, in getint_event
    return int(s)
SystemError: result with error in PyObject_Call

This needs to be checked on another OS, but the traceback problem does not strike me as Windows-specific.

[gvanrossum]

Yeah, the stack track makes me suspect http://bugs.python.org/issue17911

[rbtcollins]

I suspect that this is due to a list being passed in that wasn't created by traceback, in the older tuple-only format. That was meant to work, but possibly is being short circuited somewhere. Shall fix asap.

[rbtcollins]

Ah, idle is being somewhat naughty. It's taking the original traceback and then mangling the contents in-place, which is preserving the type information, and throwing off StackSummary.from_list. We can and should make the new code deal with this in case other folk are doing it. In future it would be good to subsume cleanup_traceback's guts into the traceback module, but we don't need to do that now.

[rbtcollins]

And here is a patch, since this is a regression I'll apply it tomorrow (or sooner if it gets reviews :))

[ncoghlan]

Patch looks good to me.

[python-dev]

New changeset ea3cc128ce35 by Robert Collins in branch 'default':
Issue bpo-23631: Fix traceback.format_list when a traceback has been mutated.
https://hg.python.org/cpython/rev/ea3cc128ce35

[rbtcollins]

Closing, though ideally Terry can confirm it is fully fixed for him.