Created on 2010-08-04 15:41 by brian.curtin, last changed 2010-08-04 16:23 by brian.curtin. This issue is now closed.
| Files | ||||
|---|---|---|---|---|
| File name | Uploaded | Description | Edit | |
| test_mp.diff | brian.curtin, 2010-08-04 15:41 | py3k | ||
| Messages (3) | |||
|---|---|---|---|
| msg112822 - (view) | Author: Brian Curtin (brian.curtin) *  |
Date: 2010-08-04 15:41 | |
I just realized test_multiprocessing is being skipped on Windows because a few relative imports of _multiprocessing are failing in win32 specific code blocks. Attached is a trivial patch to remove the relative import, enabling the tests to run and succeed on both dev and installed environments. Jesse: There don't appear to be any side effects here, but I get the feeling I might be overlooking some reason this might have been done. |
|||
| msg112823 - (view) | Author: Jesse Noller (jnoller) * |
Date: 2010-08-04 15:45 | |
The relative imports have been more of a pain then they've helped. I'm fine with nuking them so long as the test suite passes. |
|||
| msg112831 - (view) | Author: Brian Curtin (brian.curtin) * |
Date: 2010-08-04 16:23 | |
Fixed in r83722 (py3k) and r837274 (release31-maint). |
|||
| History | |||
|---|---|---|---|
| Date | User | Action | Args |
| 2010-08-04 16:23:35 | brian.curtin | set | status: open -> closed versions: - Python 2.7 messages: + msg112831 resolution: fixed stage: patch review -> committed/rejected |
| 2010-08-04 15:45:10 | jnoller | set | messages: + msg112823 |
| 2010-08-04 15:41:14 | brian.curtin | create | |