➜
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 2005-10-03 07:58 by nlhas, last changed 2022-04-11 14:56 by admin. This issue is now closed.
| Files | ||||
|---|---|---|---|---|
| File name | Uploaded | Description | Edit | |
| make.log | nlhas, 2005-10-03 07:58 | Output from make | ||
| script.txt | nlhas, 2005-10-10 18:17 | Script output nm and ar | ||
| Messages (6) | |||
|---|---|---|---|
| msg26470 - (view) | Author: Paul Mothersdill (nlhas) | Date: 2005-10-03 07:58 | |
Recommended to post this as a bug in comp.lang. python: ++++ Reply from Reinhold Birkenfeld > I have a standard Debian x86 system with Python 2.4.1 (compiled from > source). Attempts to compile 2.4.2 fail with references to Unicode -- > is there a basic system library that's missing? > Can you post a bug report to SourceForge and include the full output from make there? Also, have there been any unusual configure warnings? Reinhold +++++ No: nothing unusual that I can see in configure output -- I can post config.log if necessary. |
|||
| msg26471 - (view) | Author: Martin v. Löwis (loewis) *  |
Date: 2005-10-08 10:03 | |
Logged In: YES user_id=21627 Do you have an Objects/unicodeobject.o in your build tree? If yes, can you please do nm Objects/unicodeobject.o > unicodesyms.txt and attach the resulting file, as well as ar tv libpython2.4.a|grep unicode and report the output of this command? TIA. |
|||
| msg26472 - (view) | Author: Paul Mothersdill (nlhas) | Date: 2005-10-10 18:17 | |
Logged In: YES user_id=424542 There is no output from nm unicodeobject.o -- the file is there (441 bytes, no permissions problems). See script.txt. |
|||
| msg26473 - (view) | Author: Paul Mothersdill (nlhas) | Date: 2005-10-10 18:57 | |
Logged In: YES user_id=424542 The previous stuff was based on the bz2 source file. On a hunch I downloaded the tgz source, which compiled like a dream. I can only assume that there must be a difference between the two source files (at least as downloaded a couple of weeks ago immediately after release for bz2 and now for tgz). Unfortunately the original bz2 file has been deleted. Anyway: the problem has "gone away". |
|||
| msg26474 - (view) | Author: Martin v. Löwis (loewis) * |
Date: 2005-10-10 20:19 | |
Logged In: YES user_id=21627 Really strange... Can you please 1.'file Objects/unicodeobject.o' (report its output) 2. rm Objects/unicodeobject.o 3. make (report the precise command used to rebuild unicodeobject.o) 4. Invoke the command detected in 3), adding --save-temps 5. Attached a compressed version of the resulting unicodeobject.i |
|||
| msg26475 - (view) | Author: Martin v. Löwis (loewis) * |
Date: 2005-10-10 20:24 | |
Logged In: YES user_id=21627 Didn't notice your last message. Really strange, still... When I download both tar files, and uncompress them with their compressors (gzip 1.3.5 and bzip 1.0.2), the resulting tar files are byte-for-byte identical. So I assume something happened with your download. Closing as "works for me". |
|||
| History | |||
|---|---|---|---|
| Date | User | Action | Args |
| 2022-04-11 14:56:13 | admin | set | github: 42440 |
| 2005-10-03 07:58:22 | nlhas | create | |