traceback.print_last fails

[tzot]

BPO	1860
Nosy	@amauryfa
Files	test_tb_print_last.py: A test displaying the bad behaviour traceback_patch.diff: Patch for traceback.print_last()

^{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 2008-01-17.13:59:20.050>
created_at = <Date 2008-01-17.13:26:57.555>
labels = ['invalid', 'type-bug', 'library']
title = 'traceback.print_last fails'
updated_at = <Date 2008-01-17.13:59:20.038>
user = 'https://bugs.python.org/tzot'

bugs.python.org fields:

activity = <Date 2008-01-17.13:59:20.038>
actor = 'amaury.forgeotdarc'
assignee = 'none'
closed = True
closed_date = <Date 2008-01-17.13:59:20.050>
closer = 'amaury.forgeotdarc'
components = ['Library (Lib)']
creation = <Date 2008-01-17.13:26:57.555>
creator = 'tzot'
dependencies = []
files = ['9193', '9194']
hgrepos = []
issue_num = 1860
keywords = []
message_count = 3.0
messages = ['60028', '60029', '60030']
nosy_count = 2.0
nosy_names = ['tzot', 'amaury.forgeotdarc']
pr_nums = []
priority = 'normal'
resolution = 'not a bug'
stage = None
status = 'closed'
superseder = None
type = 'behavior'
url = 'https://bugs.python.org/issue1860'
versions = ['Python 2.6', 'Python 2.5']

[tzot]

traceback.print_last() depends on the existence of sys.last_type,
sys.last_value, sys.last_traceback, which don't always exist when
called. See attached example file. I will shortly send the patch for
Lib/traceback.py

[tzot]

I haven't submitted a patch since the transition from sf.net to
bugs.python.org; I assume that I don't have to open a new patch for
this, but if I have to, please let me know and I will gladly do it.

The unified diff is attached; the test example I issued works with the
patched version.

[amauryfa]

traceback.print_last() depends on the existence of sys.last_type
Yes, that is exactly as documented:

>>> help(traceback.print_last)
Help on function print_last in module traceback:
print_last(limit=None, file=None)
    This is a shorthand for 'print_exception(sys.last_type,
    sys.last_value, sys.last_traceback, limit, file)'.

Furthermore, in the doc:
http://docs.python.org/dev/library/sys.html#sys.last_type
"... they are set when an exception is not handled ..."
In your script, you *are* handling the exception.
Your script really should use traceback.print_exc() instead.
BTW, your patch basically makes print_last() identical to print_exc()...