➜
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.
Created on 2006-01-27 15:50 by ronaldoussoren, last changed 2022-04-11 14:56 by admin. This issue is now closed.
| Files | ||||
|---|---|---|---|---|
| File name | Uploaded | Description | Edit | |
| python-2.4-no-posix-source-for-panther.patch | ronaldoussoren, 2006-01-27 15:50 | |||
| Messages (6) | |||
|---|---|---|---|
| msg49415 - (view) | Author: Ronald Oussoren (ronaldoussoren) *  |
Date: 2006-01-27 15:50 | |
This patch makes sure that _POSIX_SOURCE and friends won't get defined when you build on Panther. Defining those have no effect on Panther, but will make it impossible to build extensions on a Tiger system when Python itself was build on Panther. |
|||
| msg49416 - (view) | Author: Georg Brandl (georg.brandl) * |
Date: 2006-02-19 15:54 | |
Logged In: YES user_id=1188172 Can any OS X expert confirm that not defining _XOPEN_SOURCE and _POSIX_C_SOURCE hasn't a negative effect on OSX 10.3? |
|||
| msg49417 - (view) | Author: Ronald Oussoren (ronaldoussoren) * |
Date: 2006-02-21 09:23 | |
Logged In: YES user_id=580910 I don't suppose this helps, but yes this patch is save. I use this in production and haven't found an extension yet that wouldn't work when this patch is applied. I've also looked at the platform header files to determine if the patch is save. We (Bob Ippolitto and myself) are using a simular patch in our ongoing work to create an universal build of python. |
|||
| msg49418 - (view) | Author: Neal Norwitz (nnorwitz) * |
Date: 2006-03-20 04:11 | |
Logged In: YES user_id=33168 Ronald, this looks fine, go ahead and checkin. |
|||
| msg49419 - (view) | Author: Georg Brandl (georg.brandl) * |
Date: 2006-04-01 08:51 | |
Logged In: YES user_id=849994 Committed in 43536, 43537. |
|||
| msg49420 - (view) | Author: Martin v. Löwis (loewis) * |
Date: 2006-04-04 06:09 | |
Logged In: YES user_id=21627 This patch cheated in a really dangerous way: the configure fragment was gained through editing, not through running autoconf. This was really bad, as it broke two months and a release later. The problem is that, in configure.in, [ does not denote a shell square bracket - it denotes an m4 quoted string. The square bracket is denoted with a quadrigraph. When committing configure.in, please always ignore configure fragments, and run autoconf/autoheader yourself. I fixed that in 43617 and 43618. |
|||
| History | |||
|---|---|---|---|
| Date | User | Action | Args |
| 2022-04-11 14:56:15 | admin | set | github: 42844 |
| 2006-01-27 15:50:08 | ronaldoussoren | create | |