Message 124694 - Python tracker

➜

This issue tracker has been migrated to GitHub , and is currently read-only.
For more information, see the GitHub FAQs in the Python's Developer Guide.

In-reply-to
Author	vstinner
Recipients	vstinner
Date	2010年12月27日.02:05:45
SpamBayes Score	1.3046609e-09
Marked as misclassified	No
Message-id	<1293415548.91.0.0358465969488.issue10779@psf.upfronthosting.co.za>

Content
PyErr_WarnExplicit() expects a filename encoded to UTF-8. This function is only called twice in the Python interpreter: compiler_assert() (with "assertion is always true, perhaps remove parentheses?") and symtable_warn() (eg. with "name 'xxx' is assigned to before global declaration"), and both functions pass a filename encoded to the filesystem encoding. PyErr_WarnExplicit() should use the filesystem encoding instead of UTF-8 to decode the filename. I already did the same change in issue #9713 and #10114 (r85569). Attached patch fixes this issue. See also issue #10778 (decoding_fgets() (tokenizer.c) decodes the filename from the wrong encoding).

Content

PyErr_WarnExplicit() expects a filename encoded to UTF-8. This function is only called twice in the Python interpreter: compiler_assert() (with "assertion is always true, perhaps remove parentheses?") and symtable_warn() (eg. with "name 'xxx' is assigned to before global declaration"), and both functions pass a filename encoded to the filesystem encoding.
PyErr_WarnExplicit() should use the filesystem encoding instead of UTF-8 to decode the filename. I already did the same change in issue #9713 and #10114 (r85569).
Attached patch fixes this issue.
See also issue #10778 (decoding_fgets() (tokenizer.c) decodes the filename from the wrong encoding).

History
Date	User	Action	Args
2010年12月27日 02:05:48	vstinner	set	recipients: + vstinner
2010年12月27日 02:05:48	vstinner	set	messageid: <1293415548.91.0.0358465969488.issue10779@psf.upfronthosting.co.za>
2010年12月27日 02:05:46	vstinner	link	issue10779 messages
2010年12月27日 02:05:45	vstinner	create

homepage