chiark / gitweb /
dgit.git
14 months agotest suite: Drop dependency arrangement for tests/tmp
Ian Jackson [Thu, 26 Jul 2018 08:56:37 +0000 (09:56 +0100)]
test suite: Drop dependency arrangement for tests/tmp

run-all makes the directory now adays so this is not needed.  It's
going to be harmful for --progressive, since it would make make want
to rerun things every time tests/tmp/ is updated.

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
14 months agotest suite: run-all: Minor refactoring etc.
Ian Jackson [Thu, 26 Jul 2018 05:35:31 +0000 (06:35 +0100)]
test suite: run-all: Minor refactoring etc.

* Makefile: Move the DGIT_TEST_RETRY_COUNT if/else into the rule.
  This removes a small amount of duplication.

* run-all: Document the option to pass explicitly list of scripts.

* run-all: Add an option parser, which currently supports no options.

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
14 months agotest suite: Honour DGIT_SCHROOT_CHROOT to set the schroot to use for the sbuild tests.
Ian Jackson [Thu, 26 Jul 2018 04:00:51 +0000 (05:00 +0100)]
test suite: Honour DGIT_SCHROOT_CHROOT to set the schroot to use for the sbuild tests.

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
14 months agodgit: make $need_split_build_invocation always true
Sean Whitton [Wed, 25 Jul 2018 13:05:28 +0000 (21:05 +0800)]
dgit: make $need_split_build_invocation always true

dgit will always build the source package.

One minor consequence is that dgit will usualliy generate
*_multi.changes rather than *_$arch.changes, so we need to update the
one test that makes the contrary assumption.

Bump the dgit major version number as this is quite a significant
change in implementation and also a behavioural change.

This change makes a lot of code dead.  Removing that is left as a task
for the future.

Signed-off-by: Sean Whitton <spwhitton@spwhitton.name>
Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
14 months agodgit: Move WANTSRC_* definitions up the file
Ian Jackson [Wed, 25 Jul 2018 14:11:04 +0000 (15:11 +0100)]
dgit: Move WANTSRC_* definitions up the file

We are going to want these a bit earlier.  NFC.

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
14 months agodgit: Disregard *_multi.changes for .changes ambiguity purposes
Ian Jackson [Thu, 26 Jul 2018 02:15:24 +0000 (03:15 +0100)]
dgit: Disregard *_multi.changes for .changes ambiguity purposes

The changes file ambiguity problem arises because dgit does not know
what architecture changes file the build is going to generate.  (To
know that it would have to delve even more into the command line
options the user is passing through dgit to the builder.)

dgit --always-split-build generally makes a _multi.changes file,
because it merges source changes with binaries from the build.  We are
going to make --always-split-build the only way things are done.  This
would result in lots more situations where --rm-old-changes is needed.

However, actually, we can assume that the builder does not generate a
*_multi.changes.  That will allow us to spot the builder-generated
changes file even if there is already a dgit-generated *_multi.changes
file from a previous build.

So: disregard _multi.changes, both when pre-checking for confusing
files, and when actually figuring out what the build produced.

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
14 months agodgit: push-source dies if user tried to include uncommitted changes
Sean Whitton [Wed, 25 Jul 2018 11:34:56 +0000 (19:34 +0800)]
dgit: push-source dies if user tried to include uncommitted changes

This does not make sense because you cannot dgit push uncommitted
changes.

Signed-off-by: Sean Whitton <spwhitton@spwhitton.name>
14 months agodgit: don't let the user pass --include-dirty in split brain mode
Sean Whitton [Wed, 25 Jul 2018 13:04:00 +0000 (21:04 +0800)]
dgit: don't let the user pass --include-dirty in split brain mode

It is not supported.

Signed-off-by: Sean Whitton <spwhitton@spwhitton.name>
Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
14 months agodgit: Rename --ignore-dirty to --include-dirty
Ian Jackson [Thu, 26 Jul 2018 03:22:42 +0000 (04:22 +0100)]
dgit: Rename --ignore-dirty to --include-dirty

Leaving the old name supported as a deprecated alias.

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
14 months agodgit: linkorigs: Fix error handling when bpd missing
Ian Jackson [Thu, 26 Jul 2018 03:11:10 +0000 (04:11 +0100)]
dgit: linkorigs: Fix error handling when bpd missing

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
14 months agodgit: bpd_abs(): Fix return value when bpd is absolute.
Ian Jackson [Thu, 26 Jul 2018 03:10:47 +0000 (04:10 +0100)]
dgit: bpd_abs(): Fix return value when bpd is absolute.

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
14 months agodgit: massage_dbp_args: Make names for $wantsrc values
Ian Jackson [Wed, 25 Jul 2018 13:59:27 +0000 (14:59 +0100)]
dgit: massage_dbp_args: Make names for $wantsrc values

   0 - source will NOT need to be built separately by caller
  +1 - source will need to be built separately by caller
  +2 - source will need to be built separately by caller AND

So:

   0 => WANTSRC_BUILDER
  +1 => WANTSRC_SOURCE | WANTSRC_BUILDER
  +2 => WANTSRC_SOURCE

No functional change except to debugging output.

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
14 months agochangelog: Document some recent changes
Ian Jackson [Wed, 25 Jul 2018 09:46:44 +0000 (10:46 +0100)]
changelog: Document some recent changes

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
14 months agodgit: rename $ignoredirty -> $includedirty
Sean Whitton [Wed, 25 Jul 2018 09:17:08 +0000 (17:17 +0800)]
dgit: rename $ignoredirty -> $includedirty

This better reflects the effects that the option has.

Signed-off-by: Sean Whitton <spwhitton@spwhitton.name>
14 months agodgit: rename two quilt_fixup_ functions
Sean Whitton [Thu, 28 Jun 2018 10:25:56 +0000 (11:25 +0100)]
dgit: rename two quilt_fixup_ functions

They are useful for doing things other than quilt fixup.

No functional change.

Signed-off-by: Sean Whitton <spwhitton@spwhitton.name>
14 months agodgit: actually respect $buildproductsdir in many places
Sean Whitton [Wed, 25 Jul 2018 09:02:16 +0000 (10:02 +0100)]
dgit: actually respect $buildproductsdir in many places

Signed-off-by: Sean Whitton <spwhitton@spwhitton.name>
Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
14 months agodgit: Provide $bpd_glob
Ian Jackson [Wed, 25 Jul 2018 09:01:28 +0000 (10:01 +0100)]
dgit: Provide $bpd_glob

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
14 months agodgit: add dgit.default.build-products-dir git configuration key
Sean Whitton [Mon, 23 Jul 2018 04:10:02 +0000 (12:10 +0800)]
dgit: add dgit.default.build-products-dir git configuration key

Signed-off-by: Sean Whitton <spwhitton@spwhitton.name>
Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
14 months agomessages: Be a lot more friendly about NEW in particular
Ian Jackson [Wed, 25 Jul 2018 06:23:18 +0000 (07:23 +0100)]
messages: Be a lot more friendly about NEW in particular

And also add a couple of `please's.  Closes:#904448.

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
14 months agoapt-get method: when apt does not update release files, print hint
Ian Jackson [Mon, 23 Jul 2018 10:00:09 +0000 (11:00 +0100)]
apt-get method: when apt does not update release files, print hint

Unconditionally print hint about noatime.  This problem seems to occur
quite rarely (noatime is out of fashion nowadays, and I haven't seen
any other causes) so it doesn't seem worth fiddling about writing code
to determine the mount options.  Instead, print a hint and let the
user investigate.

Closes:#851873.

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
14 months agotest suite: unset VISUAL, which interferes.
Ian Jackson [Mon, 23 Jul 2018 09:38:30 +0000 (10:38 +0100)]
test suite: unset VISUAL, which interferes.

Closes:#904308.

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
15 months agochangelog: start 5.11
Ian Jackson [Sat, 14 Jul 2018 17:13:49 +0000 (18:13 +0100)]
changelog: start 5.11

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
15 months agochangelog: finalise 5.10 archive/debian/5.10 debian/5.10
Ian Jackson [Sat, 14 Jul 2018 17:07:16 +0000 (18:07 +0100)]
changelog: finalise 5.10

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
15 months agoinfrastructure: Run git gc --auto before mirroring. Closes:#841414.
Ian Jackson [Sat, 14 Jul 2018 16:23:37 +0000 (17:23 +0100)]
infrastructure: Run git gc --auto before mirroring.  Closes:#841414.

This would not have avoided the trouble reported here
  https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=841414#74

Empirically, I managed to work around the problem by running `git gc'
(rather than `git gc --auto') and then rsyncing the result to the
mirror.  I will request that DSA increase the RAM (or swap) size on
the relevant host.

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
15 months agotest suite: Drop a couple of useless test log output lines.
Ian Jackson [Sat, 14 Jul 2018 15:58:04 +0000 (16:58 +0100)]
test suite: Drop a couple of useless test log output lines.

We don't need these (((((((((( )))))))))) now that it's just one line
inside them.

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
15 months agochangelog: rework for merging experimental tests into master
Ian Jackson [Sat, 14 Jul 2018 15:57:17 +0000 (16:57 +0100)]
changelog: rework for merging experimental tests into master

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
15 months agochangelog: finalise 5.9+exp4 archive/debian/5.9+exp4 debian/5.9+exp4
Ian Jackson [Fri, 13 Jul 2018 11:10:37 +0000 (12:10 +0100)]
changelog: finalise 5.9+exp4

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
15 months agotest suite: tar up the artifacts, and name the tarball after the test
Ian Jackson [Fri, 13 Jul 2018 10:38:36 +0000 (11:38 +0100)]
test suite: tar up the artifacts, and name the tarball after the test

Naming the output after the test works around #903696.

tarring the whole lot up turns out to make it a lot smaller.
Before:
    $ du -sk ; du -sk --apparent-size
    1640    .
    853     .
After:
    236     .
    235     .

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
15 months agotest suite: chmod a+r $AUTOPKGTEST explicitly. archive/debian/5.9+exp3 debian/5.9+exp3
Ian Jackson [Thu, 12 Jul 2018 23:54:27 +0000 (00:54 +0100)]
test suite: chmod a+r $AUTOPKGTEST explicitly.

15 months agotest suite: Do not save non-world-readable artifacts archive/debian/5.9+exp2 debian/5.9+exp2
Ian Jackson [Thu, 12 Jul 2018 23:42:26 +0000 (00:42 +0100)]
test suite: Do not save non-world-readable artifacts

Save things to $AUTOPKGTEST_ARTIFACTS without preserving access
control information.

Works around #903680.

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
15 months agochangelog: finalise 5.9+exp1 archive/debian/5.9+exp1 debian/5.9+exp1
Ian Jackson [Thu, 12 Jul 2018 12:45:17 +0000 (13:45 +0100)]
changelog: finalise 5.9+exp1

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
15 months agotest suite: Save much stuff to $AUTOPKGTEST_ARTIFACTS.
Ian Jackson [Thu, 12 Jul 2018 11:57:32 +0000 (12:57 +0100)]
test suite: Save much stuff to $AUTOPKGTEST_ARTIFACTS.

We take some care to make this of reasonable size.  In particular,
we delete raw tarballs (typically, IMPORT) and .git's sample hooks.

And, we git gc the git trees.

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
15 months agotest suite: Introduce playtree-save-refs
Ian Jackson [Thu, 12 Jul 2018 12:32:11 +0000 (13:32 +0100)]
test suite: Introduce playtree-save-refs

Arranges for all the refs in each playtree to be saved in the parent's
ref namespace.

This is going to be used when running git-gc when saving artifacts.

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
15 months agotest suite: Drop t-ok from setup/*
Ian Jackson [Thu, 12 Jul 2018 12:27:19 +0000 (13:27 +0100)]
test suite: Drop t-ok from setup/*

t-ok is going to save artifacts, which we don't want these to do.

It is not actually needed.  All it does that is useful is the fsck,
which the actual tests will do.

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
15 months agotest suite: Use dch -r -D sid '' not dch -r sid.
Ian Jackson [Thu, 12 Jul 2018 11:48:32 +0000 (12:48 +0100)]
test suite: Use  dch -r -D sid ''  not  dch -r sid.

Bizarrely,
   dch -r [-D suite]
runs $EDITOR but
   dch -r [-D suite] weevils
ignores `weevils' but does not run $EDITOR.  This is #903598.

Apparently, this is well known, and convention is to pass ''.
See https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=903598#8.

So use that.

Closes:#903441.

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
15 months agotest suite: Factor out t-dch-commit-r
Ian Jackson [Thu, 12 Jul 2018 11:37:45 +0000 (12:37 +0100)]
test suite: Factor out t-dch-commit-r

The pattern `dch -r sid' is actually wrong and we will want to change
it.  See #903441 and #903598.  For future-proofing, factor this out
before we fix it.

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
15 months agochangelog: start 5.9+exp1
Ian Jackson [Thu, 12 Jul 2018 12:43:07 +0000 (13:43 +0100)]
changelog: start 5.9+exp1

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
15 months agochangelog: finalise 5.9 archive/debian/5.9 debian/5.9
Ian Jackson [Thu, 12 Jul 2018 12:37:29 +0000 (13:37 +0100)]
changelog: finalise 5.9

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
15 months agogit-debrebase(5): Suggest `gitk -- :.' too
Ian Jackson [Mon, 9 Jul 2018 00:21:11 +0000 (01:21 +0100)]
git-debrebase(5): Suggest `gitk -- :.' too

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
15 months agogit-debrebase(5): Document best gitk options. Closes:#901927.
Ian Jackson [Sun, 8 Jul 2018 23:53:54 +0000 (00:53 +0100)]
git-debrebase(5): Document best gitk options.  Closes:#901927.

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
15 months agogit-debrebase(5): Add new subsection DISCUSSION
Ian Jackson [Sun, 8 Jul 2018 23:51:19 +0000 (00:51 +0100)]
git-debrebase(5): Add new subsection DISCUSSION

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
15 months agoNOTES.dgit-downstream-dsc.7.pod: WIP for #810829
Ian Jackson [Sun, 8 Jul 2018 23:47:02 +0000 (00:47 +0100)]
NOTES.dgit-downstream-dsc.7.pod: WIP for #810829

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
15 months agodgit-downstream-dsc(7): New manpage.
Ian Jackson [Sun, 8 Jul 2018 23:44:09 +0000 (00:44 +0100)]
dgit-downstream-dsc(7): New manpage.

Closes:#842643,#851194.

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
15 months agochangelog: Document reorganisations
Ian Jackson [Sun, 8 Jul 2018 23:30:23 +0000 (00:30 +0100)]
changelog: Document reorganisations

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
15 months agotest suite: gdr-convert-gbp: drop --overwrite
Ian Jackson [Sun, 8 Jul 2018 17:37:56 +0000 (18:37 +0100)]
test suite: gdr-convert-gbp: drop --overwrite

With #903132 properly fixed, this is no longer needed.

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
15 months agogit-debrebase convert-from-gbp: Add ffq check
Ian Jackson [Sun, 8 Jul 2018 19:05:26 +0000 (20:05 +0100)]
git-debrebase convert-from-gbp: Add ffq check

Check that the result will not count as having diverged.  This will
usually turn failures to make the ff pseudomerge into -fdiverged.

Related to #903132.

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
15 months agogit-debrebase convert-from-gbp: Make it ff of dgit view, if we can.
Ian Jackson [Sun, 8 Jul 2018 17:37:34 +0000 (18:37 +0100)]
git-debrebase convert-from-gbp: Make it ff of dgit view, if we can.

Look for dgit-generated tags so we can usually make the new branch ff
of the dgit view.

Specifically, we fish the most recent released version out of
debian/changelog and look for a DISTRO/VSN tag and a corresponding
archive/DISTRO/VSN tag.  If this all looks good, we declare ff of
archive/.

This should avoid the need for dgit push --overwrite in most cases.

We do not check the DISTRO name.  In principle dgit could find the
distro name from the suite name, but that makes the software layering
quite complicated.  (For example, we would have to provide a way to
tell gdr what program to run for dgit, for the benefit of the test
suite.)  I think the current approach is good enough; there are not
likely to be any tag pairs that look like the right kind of thing but
are in fact something else entirely, so we're not likely to
accidentally trash some wanted changes.

Closes:#903132.

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
15 months agogit-debrebase: Set $|
Ian Jackson [Sun, 8 Jul 2018 22:57:01 +0000 (23:57 +0100)]
git-debrebase: Set $|

This will avoid pepeering everything with calls to flush.  Without it
some messages from ffq_check, say, can arrive after the stderr output
from a subsequent snags_maybe_bail, which is quite confusing.

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
15 months agogit-debrebase: Provide snags_maybe_bail_early
Ian Jackson [Sun, 8 Jul 2018 18:43:06 +0000 (19:43 +0100)]
git-debrebase: Provide snags_maybe_bail_early

No callers yet.

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
15 months agogit-debrebase: Replace `die badusage' with `badusage'.
Ian Jackson [Sun, 8 Jul 2018 18:42:08 +0000 (19:42 +0100)]
git-debrebase: Replace `die badusage' with `badusage'.

badusage contains a call to finish.  So the die is redundant and never
gets called.

No functional change.

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
15 months agogit-debrebase: ffq_check: Move defaults for $ff and $notff into sub
Ian Jackson [Sun, 8 Jul 2018 18:40:35 +0000 (19:40 +0100)]
git-debrebase: ffq_check: Move defaults for $ff and $notff into sub

No functional change.

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
15 months agoDgit.pm: Move debinatag_maintview from dgit
Ian Jackson [Sun, 8 Jul 2018 17:32:09 +0000 (18:32 +0100)]
Dgit.pm: Move debinatag_maintview from dgit

git-debrebase is about to want this for convert-from-gbp dgit split
brain fixup.

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
15 months agoDgit.pm: Break out parsechangelog_loop from dgit
Ian Jackson [Sun, 8 Jul 2018 19:01:50 +0000 (20:01 +0100)]
Dgit.pm: Break out parsechangelog_loop from dgit

git-debrebase is going to want this.

No functional change except to some debugging messages.

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
15 months agoDgit.pm: Move parsecontrol etc. out of dgit
Ian Jackson [Sun, 8 Jul 2018 18:51:39 +0000 (19:51 +0100)]
Dgit.pm: Move parsecontrol etc. out of dgit

No functional change.

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
15 months agodgit: Fix a typo in a comment
Ian Jackson [Sun, 8 Jul 2018 16:54:42 +0000 (17:54 +0100)]
dgit: Fix a typo in a comment

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
15 months agogit-debrebase: Break out ff_check (for reuse by convert-from-gbp)
Ian Jackson [Sat, 7 Jul 2018 21:44:09 +0000 (22:44 +0100)]
git-debrebase: Break out ff_check (for reuse by convert-from-gbp)

No functional change.

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
15 months agogit-debrebase: record_ffq_prev_deferred: Move get_head();
Ian Jackson [Sat, 7 Jul 2018 21:27:42 +0000 (22:27 +0100)]
git-debrebase: record_ffq_prev_deferred: Move get_head();

No functional change.  This will make ff check easier to split out.

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
15 months agochangelog: start 5.9
Ian Jackson [Sun, 8 Jul 2018 10:44:27 +0000 (11:44 +0100)]
changelog: start 5.9

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
15 months agochangelog: finalise 5.8 archive/debian/5.8 debian/5.8
Ian Jackson [Sun, 8 Jul 2018 10:42:53 +0000 (11:42 +0100)]
changelog: finalise 5.8

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
15 months agogit-debrebase: Improve grammar if one blocking snag.
Ian Jackson [Fri, 6 Jul 2018 23:50:17 +0000 (00:50 +0100)]
git-debrebase: Improve grammar if one blocking snag.

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
15 months agotest suite: gdr-newupstream: Test adding revision
Ian Jackson [Fri, 6 Jul 2018 23:31:09 +0000 (00:31 +0100)]
test suite: gdr-newupstream: Test adding revision

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
15 months agogit-debrebase new-upstream: Add a -1 revision if the user didn't supply one.
Ian Jackson [Fri, 6 Jul 2018 23:41:56 +0000 (00:41 +0100)]
git-debrebase new-upstream: Add a -1 revision if the user didn't supply one.

Closes:#903127.

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
15 months agogit-debrebase: status: Reorganise commit reporting.
Ian Jackson [Fri, 6 Jul 2018 23:15:46 +0000 (00:15 +0100)]
git-debrebase: status: Reorganise commit reporting.

Replace the references to @_ with a hash.  Using a hash is much less
confusing than all this numerical indexing.  Also, \@_ is wrong
because there is only one @_, so by the time it is dereference, the
commit being printed is the anchor.

This is why in #903131 we see

      branch is unlaundered
      found mixed upstream/packaging commit (7badba627162337c7057002f37e9a1a593d07d08)
        abcc6260 Update to upstream 2.9.7

which appears to sugggest that the anchor is a mixed commit.  Actually
the mixed commit is 7badba627162 but the subsequent message is based
on @_ which has different information in it by then.

And, set $kcmsg from the main message, not including the commit
information.  We are the more sophisticated caller mentioned in the
previous commmit.

Closes: #903131.

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
15 months agogit-debrebase: keycommits: Callbacks get separate $mainwhy
Ian Jackson [Fri, 6 Jul 2018 23:13:02 +0000 (00:13 +0100)]
git-debrebase: keycommits: Callbacks get separate $mainwhy

Separate out the extra contextual info, naming the hash of the commit
being complained about, into a separate argument to $x->().  This
allows us to pass the message without that as an additional argument
to the callbacks.

This way we can still pass \&snag as a callback but other callers can
do something more sophisticated involving $cl.

No functional change for any existing callers.

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
15 months agotest suite: makepatches7: Add a nasty .gitignore
Ian Jackson [Fri, 6 Jul 2018 20:14:05 +0000 (21:14 +0100)]
test suite: makepatches7: Add a nasty .gitignore

Prior to the fix for #903130 this prevents both dgit and gdr from
making patches properly.

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
15 months agodgit, git-debrebase: Properly make patches with nasty .gitignores.
Ian Jackson [Fri, 6 Jul 2018 20:17:18 +0000 (21:17 +0100)]
dgit, git-debrebase: Properly make patches with nasty .gitignores.

Specifically:

 * When running git status --porcelain, pass --ignored so it
   lists ignored files, and handle the ! output that it then
   produces.

 * Add missing -f to a few git add invocations.

I have done some greps ('git.*add', 'git.*status', 'porcelain', etc.)
to try to find other missed cases and none turned up.

git diff is OK because normally we pass two treeish arguments, in
which case the ignores are ignored by git diff.  When we are asking it
to look at the working tree, we are expecting it to ignore untracked
files (whether ignored or not), and diff tracked ones, which is what
it does.

Closes:#903130.

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
15 months agochangelog: Document dgit(1) build unscrambling
Ian Jackson [Fri, 6 Jul 2018 17:44:18 +0000 (18:44 +0100)]
changelog: Document dgit(1) build unscrambling

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
squash! changelog: Document manpage fix

15 months agodgit(1): Unscramble push and push-source descriptions
Sean Whitton [Fri, 6 Jul 2018 10:59:59 +0000 (11:59 +0100)]
dgit(1): Unscramble push and push-source descriptions

These got mixed up in a recent commit.

Signed-off-by: Sean Whitton <spwhitton@spwhitton.name>
15 months agochangelog: start 5.8
Ian Jackson [Thu, 5 Jul 2018 14:04:21 +0000 (15:04 +0100)]
changelog: start 5.8

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
15 months agochangelog: finalise 5.7 archive/debian/5.7 debian/5.7
Ian Jackson [Thu, 5 Jul 2018 14:02:31 +0000 (15:02 +0100)]
changelog: finalise 5.7

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
15 months agotest suite: sourceonlypolicy: New test
Ian Jackson [Thu, 5 Jul 2018 00:27:09 +0000 (01:27 +0100)]
test suite: sourceonlypolicy: New test

Test the fix for #801435.

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
15 months agodgit: Check that entirely-new uploads to Debian are not source-only-uploads.
Ian Jackson [Sun, 1 Jul 2018 10:08:21 +0000 (11:08 +0100)]
dgit: Check that entirely-new uploads to Debian are not source-only-uploads.

Avoids the user pushing things that will inevitably be REJECTed.

We end up introducing some new config and force options to support
this.

We do not reuse test_source_only_changes, as it doesn't have quite the
right shape (in particular, it sometimes blithers on stdout).  Also
arguably it is better to look, specifically, for .debs, for the
purpose of this test.

This requires a new archive protocol method.  We implement it for
ftpmasterapi (where it is actualliy needed) and dummycatapi (where we
need it for tests.)  Implementing it for madisonish methods would be
easy.  For aptget, it would probably involve iterating over suites,
and not be adviseable.

Closes:#801435.

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
15 months agotest suite: Test dput-ng compatibility.
Ian Jackson [Thu, 5 Jul 2018 00:08:03 +0000 (01:08 +0100)]
test suite: Test dput-ng compatibility.

Add dput-ng as a dependency to trustingpolicy-replay.

This is the test which went wrong with dput-ng before.  It also has a
number of pushes, so it will do.

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
15 months agotest suite: Fix trustingpolicy-replay & dput-ng.
Ian Jackson [Thu, 5 Jul 2018 00:06:26 +0000 (01:06 +0100)]
test suite: Fix trustingpolicy-replay & dput-ng.

Specifically:

 * Only dput-ng actually complains about the existence of .upload
   files.  dput itself just says there is nothing to do and calls that
   success.

 * The call to t-rm-dput-dropping was in the wrong place.  t-commit
   bumps the version to 1.1 but it's 1.0 that needs removing.

Closes:#903007.

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
15 months agotest suite: checkout: Check we end up on the right branch each time
Ian Jackson [Wed, 4 Jul 2018 23:56:36 +0000 (00:56 +0100)]
test suite: checkout: Check we end up on the right branch each time

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
15 months agodgit: checkout: Do not leave detached HEAD!
Ian Jackson [Wed, 4 Jul 2018 23:56:04 +0000 (00:56 +0100)]
dgit: checkout: Do not leave detached HEAD!

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
15 months agotest suite: Always pass LC_COLLATE=C to sort(1).
Ian Jackson [Wed, 4 Jul 2018 23:15:44 +0000 (00:15 +0100)]
test suite: Always pass LC_COLLATE=C to sort(1).

Specifically, by replacing every invocation of sort with the new
function t-sort.  The one case where we already took care of this, we
change LC_ALL to LC_COLLATE, which should be sufficient.

Closes:#903006.

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
15 months agodgit(7): Mention git-debrebase and gbp pq alongside git-dpm
Ian Jackson [Wed, 4 Jul 2018 01:19:05 +0000 (02:19 +0100)]
dgit(7): Mention git-debrebase and gbp pq alongside git-dpm

in the comment about handling patch stacks.

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
15 months agotest suite: checkout: new test
Ian Jackson [Wed, 4 Jul 2018 00:54:13 +0000 (01:54 +0100)]
test suite: checkout: new test

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
15 months agodgit(1): checkout: Document when this acceses the network
Ian Jackson [Wed, 4 Jul 2018 00:27:42 +0000 (01:27 +0100)]
dgit(1): checkout: Document when this acceses the network

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
15 months agodgit: checkout: New command.
Ian Jackson [Sun, 1 Jul 2018 11:41:40 +0000 (12:41 +0100)]
dgit: checkout: New command.

Closes:#878443.

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
15 months agodgit: fetch: Introduce new dofetch()
Ian Jackson [Wed, 4 Jul 2018 00:21:43 +0000 (01:21 +0100)]
dgit: fetch: Introduce new dofetch()

This actually does all that is needed.  It replaces two now-identical
calling patterns.

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
15 months agodgit: fetch: Reorganise cmd_fetch
Ian Jackson [Wed, 4 Jul 2018 00:18:25 +0000 (01:18 +0100)]
dgit: fetch: Reorganise cmd_fetch

This slightly odd construction is actually how the return value from
fork_for_multisuite should be handled.  It now looks much more like
the calling pattern in pull().

No functional change: if $multi_fetched is falseish, we call
fetch_one, and then, no matter what, we finish 0.

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
15 months agodgit: fetch: Slightly refactor pull()
Ian Jackson [Wed, 4 Jul 2018 00:16:19 +0000 (01:16 +0100)]
dgit: fetch: Slightly refactor pull()

When for_for_multisuite returns '0', meaning we are a child and should
do one suite, use finish 0 rather than returning, after doing the
fetch.

This has no functional change because the one call site is at the end
of cmd_pull, after which we finish 0 anyway.

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
15 months agodgit: fetch: Rename fetch_one from fetch (nfc)
Ian Jackson [Wed, 4 Jul 2018 00:12:52 +0000 (01:12 +0100)]
dgit: fetch: Rename fetch_one from fetch (nfc)

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
15 months agodgit(1): Document as-yet-unwritten `dgit checkout'
Ian Jackson [Sun, 1 Jul 2018 11:00:07 +0000 (12:00 +0100)]
dgit(1): Document as-yet-unwritten `dgit checkout'

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
15 months agodgit: update-vcs-git: Honour --package properly.
Ian Jackson [Sun, 1 Jul 2018 11:20:50 +0000 (12:20 +0100)]
dgit: update-vcs-git: Honour --package properly.

Specifically, use package_from_d_control rather than open-coding
a substandard implementation.

The consequent abolition of $sourcep does mean that if the user
specifies the suite `.', and does not specify -p, we needlessly parse
debian/control twice.  This doesn't really matter.

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
15 months agodgit: Break out package_from_d_control (nfc)
Ian Jackson [Sun, 1 Jul 2018 11:18:22 +0000 (12:18 +0100)]
dgit: Break out package_from_d_control (nfc)

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
15 months agochangelog: start 5.7
Ian Jackson [Sat, 30 Jun 2018 23:43:27 +0000 (00:43 +0100)]
changelog: start 5.7

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
15 months agochangelog: finalise 5.6 archive/debian/5.6 debian/5.6
Ian Jackson [Sat, 30 Jun 2018 23:41:46 +0000 (00:41 +0100)]
changelog: finalise 5.6

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
15 months agochangelog: Document empirical effectiveness of new gnupg workaround
Ian Jackson [Sat, 30 Jun 2018 22:58:56 +0000 (23:58 +0100)]
changelog: Document empirical effectiveness of new gnupg workaround

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
15 months agotest suite: manpages-format: Test -Tps too
Ian Jackson [Thu, 28 Jun 2018 14:53:42 +0000 (15:53 +0100)]
test suite: manpages-format: Test -Tps too

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
15 months agotest suite: manpages-format: work properly when no expected warnings
Ian Jackson [Thu, 28 Jun 2018 14:53:28 +0000 (15:53 +0100)]
test suite: manpages-format: work properly when no expected warnings

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
15 months agotest suite: manpages-format: Remove some pointless redirects
Ian Jackson [Thu, 28 Jun 2018 14:52:42 +0000 (15:52 +0100)]
test suite: manpages-format: Remove some pointless redirects

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
15 months agotest suite: manpages-format: Prep for other formats (nfc)
Ian Jackson [Thu, 28 Jun 2018 14:46:06 +0000 (15:46 +0100)]
test suite: manpages-format: Prep for other formats (nfc)

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
15 months agotest suite: manpages-format: New test
Ian Jackson [Thu, 28 Jun 2018 14:31:56 +0000 (15:31 +0100)]
test suite: manpages-format: New test

Test that manpages format with only expected roff warnings.

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
15 months agoMerge remote-tracking branch 'dgit/dgit/experimental'
Ian Jackson [Sat, 30 Jun 2018 21:40:02 +0000 (22:40 +0100)]
Merge remote-tracking branch 'dgit/dgit/experimental'

Conflicts:
debian/changelog

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
15 months agotest suite: gpg agent workaround: Fix bugs in log neatening. archive/debian/5.5+exp9 debian/5.5+exp9
Ian Jackson [Sat, 30 Jun 2018 18:03:10 +0000 (19:03 +0100)]
test suite: gpg agent workaround: Fix bugs in log neatening.

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
15 months agotest suite: gpg agent workaround: Neaten the logging. archive/debian/5.5+exp8 debian/5.5+exp8
Ian Jackson [Sat, 30 Jun 2018 09:21:32 +0000 (10:21 +0100)]
test suite: gpg agent workaround: Neaten the logging.

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
15 months agotest suite: Provide DGIT_TEST_BYHAND if appropriate
Ian Jackson [Sat, 30 Jun 2018 09:20:57 +0000 (10:20 +0100)]
test suite: Provide DGIT_TEST_BYHAND if appropriate

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>