On Fri, 2008-03-07 at 17:51 +0000, Kevin Kofler wrote: > That sounds like an interesting idea, and if the choice is between that or > making foo_1-1_fc9 directly immutable, I definitely prefer your idea. > (Rationale: That way one doesn't have to bump Release for trivial fixes after a > failed build.) > > I'd suggest, however, that the cronjob, when creating foo_1-1_fc9, also deletes > all the foo_1-1_fc9-build* tags. (Rationale: They're no longer useful once we > know which tag actually went into the successful build, and thus only clutter > up the history and slow things down (CVS is slower when there are more tags).) Actually, there is on reason really to write back the specially named tag, we could just rely on whatever tag was used to do the completed/shipped build. Koji can tell us what it is, based on the build task, that's all we need. Before we didn't have koji, and we didn't have a good way to discover what the CVS tag was used to do a build. Now we do. -- Jesse Keating Fedora -- All my bits are free, are yours?
Attachment:
signature.asc
Description: This is a digitally signed message part
-- fedora-devel-list mailing list fedora-devel-list@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/fedora-devel-list
