evolve merge requestshttps://foss.heptapod.net/mercurial/evolve/-/merge_requests2020-07-18T21:04:13Zhttps://foss.heptapod.net/mercurial/evolve/-/merge_requests/196topic: provide compatibility for the API changes of D86542020-07-18T21:04:13ZJörg Sonnenbergertopic: provide compatibility for the API changes of D8654Also affects serverminitopic, reaching over to topic.compat.Also affects serverminitopic, reaching over to topic.compat.https://foss.heptapod.net/mercurial/evolve/-/merge_requests/201evolve: add first version of extensive census of troubles and resolution2020-08-26T10:59:07ZSushil Khanchievolve: add first version of extensive census of troubles and resolutionPierre-Yves DavidPierre-Yves Davidhttps://foss.heptapod.net/mercurial/evolve/-/merge_requests/212evolve: make sure we always pick min revision to solve content-div2021-05-20T07:48:55ZSushil Khanchievolve: make sure we always pick min revision to solve content-divSushil KhanchiSushil Khanchihttps://foss.heptapod.net/mercurial/evolve/-/merge_requests/220evolve: fix bug in public content divergence merging logic in continue case2020-08-31T08:07:04ZSushil Khanchievolve: fix bug in public content divergence merging logic in continue caseSushil KhanchiSushil Khanchihttps://foss.heptapod.net/mercurial/evolve/-/merge_requests/221refactoring for public content divergence2020-09-19T18:44:11ZSushil Khanchirefactoring for public content divergencePierre-Yves DavidPierre-Yves Davidhttps://foss.heptapod.net/mercurial/evolve/-/merge_requests/222compatibility with cmdutil.revert() without parents2020-08-25T08:05:21ZAnton Shestakovcompatibility with cmdutil.revert() without parentsNot sure about the last commit (split command), but the tests are passing.Not sure about the last commit (split command), but the tests are passing.Pierre-Yves DavidPierre-Yves Davidhttps://foss.heptapod.net/mercurial/evolve/-/merge_requests/223WIP: split discard action weirdness2020-08-25T08:26:02ZAnton ShestakovWIP: split discard action weirdnesshttps://foss.heptapod.net/mercurial/evolve/-/merge_requests/227Topic/default/movedirstate2020-09-04T12:46:22ZMartin von ZweigbergkTopic/default/movedirstatePierre-Yves DavidPierre-Yves Davidhttps://foss.heptapod.net/mercurial/evolve/-/merge_requests/231Topic/default/fold phases2020-09-10T17:09:24ZMartin von ZweigbergkTopic/default/fold phasesAnton ShestakovAnton Shestakovhttps://foss.heptapod.net/mercurial/evolve/-/merge_requests/234topic: fix the `topic(...)` revset to not select commits without a topic2020-09-03T12:56:32ZMatt Harbisontopic: fix the `topic(...)` revset to not select commits without a topicThe problem seems to be that the now removed __contains__ check would select any
revision indicated by a non-string argument. This led to `topic(head())`
returning a bunch of non-topic heads, though not as many as simply `head()`.
I'm ...The problem seems to be that the now removed __contains__ check would select any
revision indicated by a non-string argument. This led to `topic(head())`
returning a bunch of non-topic heads, though not as many as simply `head()`.
I'm likely misunderstanding something, because there appears to be a similar
__contains__ check for `branch()` in core. But maybe that's just harder to see
the issue, because branch names doesn't disappear like topics.Pierre-Yves DavidPierre-Yves Davidhttps://foss.heptapod.net/mercurial/evolve/-/merge_requests/236small fixes2020-09-10T01:05:40ZMatt Harbisonsmall fixesmoved to !239
moved to !239
Pierre-Yves DavidPierre-Yves Davidhttps://foss.heptapod.net/mercurial/evolve/-/merge_requests/238WIP: rewind improvements WIP2020-12-22T11:08:38ZAnton ShestakovWIP: rewind improvements WIPWIPWIPhttps://foss.heptapod.net/mercurial/evolve/-/merge_requests/240Topic/default/prune already obsolete2020-09-10T16:03:26ZMartin von ZweigbergkTopic/default/prune already obsoleteAnton ShestakovAnton Shestakovhttps://foss.heptapod.net/mercurial/evolve/-/merge_requests/244evolvecmd: don't set `branchmerge=True` while resolving phase divergence2020-09-19T16:25:31ZPulkit Goyalevolvecmd: don't set `branchmerge=True` while resolving phase divergenceWe are not going to create a merge commit, neither we are using two parents in
any way.
I am unable to see any reason why `branchmerge` should be passed as `True` here.
Passing branchmerge as True leads to side effect of having merge r...We are not going to create a merge commit, neither we are using two parents in
any way.
I am unable to see any reason why `branchmerge` should be passed as `True` here.
Passing branchmerge as True leads to side effect of having merge related commit
information stored in mergestate which leads us to having an active mergestate.Pierre-Yves DavidPierre-Yves Davidhttps://foss.heptapod.net/mercurial/evolve/-/merge_requests/259tests: {successorssets} now always return a list, affecting json output2020-10-12T15:09:21ZMartin von Zweigbergktests: {successorssets} now always return a list, affecting json outputThe upstream commit is not public yet.The upstream commit is not public yet.https://foss.heptapod.net/mercurial/evolve/-/merge_requests/260khanchi-gca refactoring2021-03-01T12:43:36ZSushil Khanchikhanchi-gca refactoringSushil KhanchiSushil Khanchihttps://foss.heptapod.net/mercurial/evolve/-/merge_requests/261Fix TypeError caused by the `ngtip` revset returning a node instead of a rev2020-10-16T12:42:36ZSietse BrouwerFix TypeError caused by the `ngtip` revset returning a node instead of a revMR moved to !266
See https://bz.mercurial-scm.org/show_bug.cgi?id=6405
## How to trigger the error
The error was triggered by running
hg log -r 'ngtip("default")'
in a repository with more than one changeset. When the flow of c...MR moved to !266
See https://bz.mercurial-scm.org/show_bug.cgi?id=6405
## How to trigger the error
The error was triggered by running
hg log -r 'ngtip("default")'
in a repository with more than one changeset. When the flow of control reached
`reachableroots2`, it raised
TypeError: an integer is required (got type bytes)
See the Bugzilla link above for a full traceback.
## Error cause investigation
That node hash (of type `bytes`) got added to the revlog earlier, by the topic extension's `revset.py:ngtipset` function (in evolve source `/hgext3rd/topic/revset.py`). The call stack from there proceeds thus:
Mercurial calls (in an example repo with two commits)
```
hgext_topic.revset.ngtipset(
repo,
subset=<fullreposet+ 0:2>,
x=('symbol', 'default')
)
```
which calls `hgext_topic.destination.ngtip`,
which calls `hgext_topic.topicmap.branchtip`,
which calls `mercurial.branchmap.branchtip`
which returns a node hash.
## Open questions for this Merge Request
Question is: At what point in the call stack above should the node hash be turned into a rev number? The current patch chooses `revset.py:ngtipset`.
- I'm certain `revset.ngtipset` should add a rev number to the changeset, not a node hash
- I'm not certain what callers of `destination.ngtip` expect it to return.Sietse BrouwerSietse Brouwerhttps://foss.heptapod.net/mercurial/evolve/-/merge_requests/270changelog: update for 10.1.02020-10-31T10:57:39ZAnton Shestakovchangelog: update for 10.1.0Pierre-Yves DavidPierre-Yves Davidhttps://foss.heptapod.net/mercurial/evolve/-/merge_requests/271changelog: update for 10.0.32020-10-31T10:57:36ZAnton Shestakovchangelog: update for 10.0.3--HG--
branch : stable--HG--
branch : stablePierre-Yves DavidPierre-Yves Davidhttps://foss.heptapod.net/mercurial/evolve/-/merge_requests/274debian changelog and merge back to default2020-11-03T14:02:14ZAnton Shestakovdebian changelog and merge back to defaultPierre-Yves DavidPierre-Yves David