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.

Author pjenvey
Recipients Ronan.Lamy, larry, pitrou, pjenvey, serhiy.storchaka, vstinner
Date 2016-04-18.21:34:55
SpamBayes Score -1.0
Marked as misclassified Yes
Message-id <1461015295.39.0.246147143774.issue26800@psf.upfronthosting.co.za>
In-reply-to
Content
Basically a reopen of the older issue8485 with the same name. It was decided there to drop support for bytearray filenames -- partly because of the complexity of handling buffers but it was also deemed to just not make much sense.

This regressed or crept back into the posix module with the big path_converter changes for 3.3:

https://hg.python.org/cpython/file/ee9921b29fd8/Lib/test/test_posix.py#l411

IMHO this functionality should be deprecated/removed per the original discussion, or does someone want to reopen the debate?

The os module docs (and path_converter's own docs) explicitly advertise handling of str or bytes, not bytearrays or buffers. Even os.fsencode rejects bytearrays/buffers.

Related to issue26754 -- further inconsistencies around filename handling
History
Date User Action Args
2016-04-18 21:34:55pjenveysetrecipients: + pjenvey, pitrou, vstinner, larry, serhiy.storchaka, Ronan.Lamy
2016-04-18 21:34:55pjenveysetmessageid: <1461015295.39.0.246147143774.issue26800@psf.upfronthosting.co.za>
2016-04-18 21:34:55pjenveylinkissue26800 messages
2016-04-18 21:34:55pjenveycreate