Ian Jackson writes ("Re: Triggers status?"): > If I rebase this branch, will this cause trouble for my flex-based > parser branch, which I took off from the triggers code ?
Following somewhat inconclusive discussion on IRC about these and related questions, I said I would go away and try this. Firstly, I had to merge my triggers branch with the latest head. I've done this now and the result is in the same place as before: http://www.chiark.greenend.org.uk/~ian/git/dpkg/dpkg.triggers/ While doing the merge I found that the triggers patch conflicted with the change to the semantics of successful postinst completion, which was made pursuant to bug #443334. As readers may know, I disagree with that change. In any case, as the triggers branch consolidates some of the cleanup postinst functionality in a single place, it will be more expedient to revert the patch from #4433334 first and redo it again if it is still felt to be correct. So for now I have reverted that change in my triggers branch. Obviously I think it should be merged as-is. I then tried to use rebase, like this: cp -a dpkg.triggers dpkg.triggers.rebase cd dpkg.triggers.rebase/ git-fetch ../dpkg.debian master:debian-head # where ../dpkg.debian is a straightforward pull of the head git-rebase debian-head This produced a great deal of output including complaints, for example: error: patch failed: debian/control:6 error: debian/control: patch does not apply and: CONFLICT (content): Merge conflict in src/cleanup.c Just to check that it's not anything wrong with my triggers branch I did this too: cp -a dpkg.debian dpkg.debian.merge cd dpkg.debian.merge/ git pull ../dpkg.triggers which seemed to work perfectly. Ian. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]