Summary: | dev-python/subvertpy fails tests (subversion assert failed) | ||
---|---|---|---|
Product: | Gentoo Linux | Reporter: | Diego Elio Pettenò (RETIRED) <flameeyes> |
Component: | New packages | Assignee: | Python Gentoo Team <python> |
Status: | RESOLVED FIXED | ||
Severity: | normal | CC: | bazaar+obsolete, mgorny |
Priority: | Normal | ||
Version: | unspecified | ||
Hardware: | All | ||
OS: | Linux | ||
URL: | https://tinderboxlogs.s3.amazonaws.com/tbamd64.excelsior.flameeyes.eu/dev-python%3Asubvertpy-0.8.10%3A20120624-171057.html | ||
Whiteboard: | |||
Package list: | Runtime testing required: | --- |
Description
Diego Elio Pettenò (RETIRED)
2012-06-24 18:04:00 UTC
Curious; In file 'subversion/libsvn_wc/adm_ops.c' line 189 a C file, not a .py file. Could not replicate archtester subvertpy # USE_PYTHON="2.7" ebuild subvertpy-0.8.9.ebuild clean test * Testing of dev-python/subvertpy-0.8.9 with CPython 2.7... .................................................................................................................................................................... ---------------------------------------------------------------------- Ran 169 tests in 87.017s OK well, it seems the bug is a valid 1 since I now can replicate it. However, I have to add rm -f $PN/tests/test_client.py to the ebuild just to get to it. That means there are at least 2 separate tests that can't pass. The sad news is that https://bugs.launchpad.net/subvertpy/+bug/1022754 was files already 09 07 2012, just a couple of weeks after this one. subvertpy is not a highly used package, I'll attempt to keep an eye on that filed bug but after 3.5 months I am not expecting too much. Hopefully the next version of subvert.py will pass all again. I think this bug has been fixed upstream? See https://bugs.launchpad.net/bugs/1062247 and https://bugs.launchpad.net/subvertpy/+bug/1022754. (In reply to comment #3) > I think this bug has been fixed upstream? See > https://bugs.launchpad.net/bugs/1062247 and > https://bugs.launchpad.net/subvertpy/+bug/1022754. Well, first of all, this is a mess. It seems to me that the minor launchpad branches are not active, and the launchpad 'trunk' is in sync with git.samba.org repo. Therefore, I will be referring to the git repo. I'm thinking that 'rev 2456' may be just a random rev user tried, and the fix may have been earlier. But he is mentioning svn1.6 as well which is unlikely to fix new issues with svn1.7. One patch addressing test_info is this one [1]. Not sure how much it does; as far as I can tell, it seems that the package needs code fixes wrt svn1.7, not just test fixes. I'm thinking of adding a live ebuild to see if HEAD is at least semi-useful. If it is not, there's no point in tracking patches which won't help. [1]:http://gitweb.samba.org/?p=jelmer/subvertpy.git;a=commit;h=b596eae2e469 Can't repro with new 0.10.1. |