New submission from David D. Riddle:
test_linecache.py reads from three files namely inspect_fodder.py,
inspect_fodder2.py, and mapping_tests.py. It reads the py files directly as
text files. This patch copies these files to linecache_fodder,
linecache_fodder2, and linecache_mapping_fodder
David D. Riddle added the comment:
I have made the changes you suggested. Please look at lines 87-97. I am not
certain if that is correct behavior. I was surprised that a SyntaxError is
thrown in addition to the UnicodeDecodeError when linecache.getlines takes a
file with bad unicode. Please
David D. Riddle added the comment:
I have made the changes you suggested. Please tell me if any further changes
are needed.
> The unclosed file suggests a cleanup bug in linecache which we should
> fix for hygiene, but is separate :)
Should I file a bug report?
--
Added file
David D. Riddle added the comment:
Here is another patch. I fixed a bug in GetLineTestsBadData.test_getline and
refactored the code. I think everything is good now.
As for the ResourceWarnings they are not caused by linecache. The problem is
with tokenize.open (See http://bugs.python.org