summaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAge
* dgit: baredebian: Provide $quilt_upstream_commitish_messageIan Jackson2019-06-29
| | | | | | No functional change. Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
* dgit: baredebian: Note upstream version and commitish sourceIan Jackson2019-06-29
| | | | Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
* test suite: baredebian-push: Actually test --quilt=baredebianIan Jackson2019-06-29
| | | | | | This is all quite awful really. I think that is inherent. Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
* test suite: baredebian setup, and a skeleton to run itIan Jackson2019-06-29
| | | | | | The baredebian-push test will be fleshed out in a moment. Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
* i18n: Add translator notes about ASCII art changes displayIan Jackson2019-06-29
| | | | Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
* dgit: baredebian: New quilt modeIan Jackson2019-06-29
| | | | | | | This is the bulk of the implementation including all essential changes. Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
* dgit: baredebian: Introduce --upstream-commitishIan Jackson2019-06-29
| | | | | | | | | Not documented yet. Nor, used, since $quilt_mode is not ever baredebian yet. So currently just an option you can pass to make dgit bomb out. Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
* test suite: Add DEBORIG pseudo-dependencyIan Jackson2019-06-29
| | | | Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
* Dgit.pm: resolve_upstream_version: Return a message tooIan Jackson2019-06-29
| | | | | | No functional change for existing callers. Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
* dgit: Introduce $uhead and $uhead_whatshortIan Jackson2019-06-29
| | | | | | | | No overall functional change other than (sadly) to complicate the life of translators. I will add some notes for translators after I've made this string actually vary. Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
* dgit: Make $format available in build_or_push_prep_modesIan Jackson2019-06-29
| | | | | | | | By returning it from determine_whether_split_brain. No functional change yet. Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
* git-debrebase: Replace one open-coded hash-object with hash_commitIan Jackson2019-06-29
| | | | | | No functional change. Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
* Dgit.pm: Move hash_commit from dgitIan Jackson2019-06-29
| | | | | | No functional change other than to no longer honour @git. Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
* Dgit.pm: Move make_commit (git-commit-tree wrapper) from gdrIan Jackson2019-06-29
| | | | | | No functional change other than to no longer honour @git. Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
* Rename various *make_commit* to *hash_commit*Ian Jackson2019-06-29
| | | | | | | | This avoids a subroutine name clash with make_commit in gdr. No functional change. Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
* Dgit.pm: Move read-tree wrappers from git-debrebaseIan Jackson2019-06-29
| | | | | | No functional change other than to no longer honour @git. Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
* Dgit.pm: Move resolve_upstream_version from git-debrebaseIan Jackson2019-06-29
| | | | | | No functional change. Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
* Dgit.pm: Move upstream_commitish_search from git-debrebaseIan Jackson2019-06-29
| | | | | | No functional change. Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
* docs: Document --split-view and change terminologyIan Jackson2019-06-29
| | | | | | | | | We now speak in the docs of * splitting quilt mode(s) * split view (being) in operation Closes: #926640 Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
* test suite: new test forcesplit-linearIan Jackson2019-06-28
| | | | | | Test split view with non-splitting quilt mode. Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
* test suite: forcesplit-overwrite: Test with bare --overwriteIan Jackson2019-06-28
| | | | Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
* test suite: forcesplit-overwrite: Test without --overwriteIan Jackson2019-06-28
| | | | Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
* test suite: forcesplit-overwrite: Break out "try"Ian Jackson2019-06-28
| | | | | | No functional change. Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
* test suite: forcesplit-overwrite: Check -pushed-goodIan Jackson2019-06-28
| | | | Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
* test suite; New test forcesplit-overwriteIan Jackson2019-06-28
| | | | | | | | Still needs: * checking that we didn't move HEAD * checking HEAD tree is equal Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
* dgit: Reject split view with correct messageIan Jackson2019-06-28
| | | | | | | | | | Now, it is not just a splitting quilt mode which might cause this, but also an explicit request for split view. This has some duplicated text, which I understand is easier for translators. Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
* dgit: Support split view in non-splitting quilt modesIan Jackson2019-06-28
| | | | | | | | | Provide --split-view|brain=auto|always|never and the corresponding access cfg option .split-view. This is not documented or tested yet. Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
* dgit: In code, rename splitbrain quilt modes to splittingIan Jackson2019-06-28
| | | | | | | | | | | | | A "splitting" quilt mode is going to be one which requires split brain mode. But split brain mode is going to be possible in other quilt modes too. The existing name of the splitbrain quilt cache is correct: it is used precisely for quilt fixup, in split brain modes (even for non splitting quilt modes, now). In split brain mode without quilt, it is not needed or used. Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
* dgit: Reject dgit pull in split brain modeIan Jackson2019-06-28
| | | | | | | | | | | In splitting quilt modes dgit pull would have to back-convert the changes and currently there is no code to do this, and these cases were correctly rejected before. But in other quilt modes it would still have to strip off pseudomerges or quilt fixup commits. Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
* dgit: Move determine_whether_split_brain further up the fileIan Jackson2019-06-28
| | | | | | No functional change. Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
* dgit: Break out determine_whether_split_brainIan Jackson2019-06-28
| | | | | | No functional change. Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
* dgit: Crash if we forgot to set $do_split_brainIan Jackson2019-06-28
| | | | Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
* dgit: Put !! inside do_split_brain()Ian Jackson2019-06-28
| | | | | | No functional change. Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
* dgit: Introduce do_split_brain()Ian Jackson2019-06-28
| | | | | | No functional change yet. Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
* dgit(1): Tidying up after tag format: Drop splitbrain conditionalIan Jackson2019-06-28
| | | | | | | | | Now, debiantags is always just debiantag_new plus debiantag_maintview. So we can drop this complexity. No functional change. Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
* dgit: Tidying up after tag format: Drop debiantagIan Jackson2019-06-28
| | | | | | | | | | | Remove the now-superfluous level of indirection. No functional change. At some future point we may rename debiantag_new to debiantag everywhere, but not right now. Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
* Dgit.pm: Tidying up after tag format: Drop debiantag_oldIan Jackson2019-06-28
| | | | | | | | | The one remaining call site was in debiantags, where debiantag_maintview is (also) correct. No functional change. Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
* dgit(1): Tidying up after tag format: Drop qualificationsIan Jackson2019-06-28
| | | | | | All servers are now assumed to cope with split view tags. Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
* dgit(1): Correct --always-dep14tagIan Jackson2019-06-28
| | | | | | | This was documented as --always-dep14tag but dgit only accepts --dep14tag-always. Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
* dgit: Tidying up after tag format: Collapse $dodep14tag testIan Jackson2019-06-28
| | | | | | No functional change. Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
* infra: Drop support for old tag name on server sideIan Jackson2019-06-28
| | | | Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
* dgit: Introduce $protovsn 5; part of tidying up after tag formatIan Jackson2019-06-28
| | | | | | | | | | | This drops the tagformat param; the semantics are as if `new' were specified. Eventually, when we can drop support for $protovsn 4, all traces of this parameter can then be gone. No overall functional change. No immediate compatibility implications. Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
* dgit: Drop $protovsn < 4; part of tidying up after tag formatIan Jackson2019-06-28
| | | | | | | | | | This drops support for rpush where our peer is older than dgit 2.0. (This does not include the dgit in any Debian release.) No nontrivial functional change, since we already rejected older dgits due to them implying the old tag format. Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
* dgit: Tidying up after tag format: drop ..._can_splitbrainIan Jackson2019-06-28
| | | | | | Substitute this fixed value into its call sites. No functional change. Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
* dgit: Drop support for old tag formatIan Jackson2019-06-28
| | | | | | | | | | access_cfg_tagformats_can_splitbrain always returns true now. We'll delete it in a moment. We now insist on protocol version 4 because older ones imply the old tag format. Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
* dgit: tag format: Add missing $protovsn comment re tagformat paramIan Jackson2019-06-28
| | | | | | | No fuctional change. We are going to get rid of this in a moment, but this change makes clearer what is going on. Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
* test suite: Drop all testing of old tag formats, and compatibilityIan Jackson2019-06-28
| | | | | | | | | | | This is quite old now. We are dropping support for it. The new tags came in in dgit 2.0 in October 2016. No version of dgit without them was in any Debian release. We don't expect there are any sites using old servers which do not support the new format, or anyone using a dgit client older than 2.x who cannot upgrade. Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
* test suite: break out t-splitbrain-rm-1-patchIan Jackson2019-06-28
| | | | | | | | We are going to call this from one of the force split tests. No functional change. Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
* test suite: t-{unapplied,dpm}-pushed-good: set $suiteIan Jackson2019-06-28
| | | | | | | | t-pushed-good-core wants suite to be set. Rather than having it implicit, make it an argument. Fix the two call sites where suite was not trivially `sid'. Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
* dgit: Do split brain cache and saving in build_maybe_quilt_fixupIan Jackson2019-06-28
| | | | | | | This wants to happy whenever we have a split brain, not only in splitting multipatch modes. Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
generated by cgit v1.2.3 (git 2.25.1) at 2024-06-02 05:19:26 +0000