1 Tag rename unfinished business
2 (things resulting from searching for `tag' where searching for
3 `debiantag' wouldn't find them)
5 - in dopush (?), we need to sort out synthetic fast forard
6 this should not depend entirely on quilt mode
7 but will always be needed if quiltmode_splitbrain
8 * On push, if ff fixup mode enabled, check debian/changelog,
9 and maybe make fake merge, maybe in dgit view, maybe
12 - there is some confusion between $split_brain (did we actually
13 generate any commits) and quiltmode_splitbrain (might we have done,
14 including in previous pushes); search for $split_brain and check
15 that use sites are correct
20 - start repo with old tag, push with new tag, run
21 "old" dgit to push again with new tag, push again with
24 - rejecting push of same version with one tag when then the other
27 Test that, where appropriate, maint view tags are pushed to dgit git repo
30 * Provide --gpbpq[=:] options
32 * Document --quilt=unapplied|gbp|dpm
34 * Mention new tag format in debian/changelog
36 * Mention new --quilt=* in debian/changelog
38 * Test all three new --quilt=unapplied|gbp|dpm
41 * IN ARCHIVE COPY Consider and implement fast-forward fixup
43 - On fetch, check if specified git hash is ff from
44 existing dgit branch, the same way we do for constructed
45 dscs. The .dsc tells us the git tree but not necessarily
50 complicated there are many branches
51 dgit-repos: refs/dgit/suite
53 local; remotes/dgit/dgit/suite
56 local: dgit/suite (not relevant to fetch)
58 archive dsc must always be regarded
59 as ff from (or same as) dgit-repos
60 err depending on version number ?
62 this code is what needs changing
64 fail "git head (".lrref()."=$lastpush_hash) is not a ".
65 "descendant of archive's .dsc hash ($dsc_hash)";
67 it just needs to generate the fake merge and set $hash
68 the archive skew stuff is all fine with this
70 xxx actually, no, we need to stop from squashing our
73 * Mention in-archive copy fix in debian/changelog
75 * Add in-archive-move test case