27 Oct
2011
27 Oct
'11
8:19 a.m.
Le 26/10/2011 20:02, Chris Nighswonger a écrit : > Now, let us say I discover bug 'bar' in rel_3_8 and in rel_3_6. > > 1. I open a bug report against 'rel_3_8' for bug 'bar.' mmm... I think there are 2 sub-cases here: - library A, that uses Koha 3.6, find the bug and open it in rel_3_6 - hacker B, that uses master, find the bug and open it in master (or in rel_3_8 ? that's where there can be a confusion between master and rel_3_8) > 2. I fix 'bar' and attach a patch to the bug. > 3. The fix passes QA. > 4. The RM pushes the patch and moves it to 'rel_3_6' if applicable. (But > he has to have this knowledge somehow.) right, this point is the problem. And I feel it's not solved by this proposed workflow (not by the "need backporting as well") > That is ok as long as it works out in practice. not sure it will work well then... I think we haven't completly found the "perfect" workflow > It would probably be better to have a way for the bug reporter to > indicate *all* branches to which the bug applies, but I don't know if > this is possible. Perhaps Ian can comment on that. I've investigated bugzilla, and can't find a way to have a multiple version selection. OTH, it's easy to have another field "backport needed", that could be used for that. I don't know if it's a good idea, I just throw it... -- Paul POULAIN http://www.biblibre.com Expert en Logiciels Libres pour l'info-doc Tel : (33) 4 91 81 35 08