Sushil khanchi <sushilkhanchi97@gmail.com> [Sat, 28 Dec 2019 23:20:05 +0530] rev 5051
evolve: remove dead code
Sushil khanchi <sushilkhanchi97@gmail.com> [Sat, 07 Dec 2019 23:13:40 +0530] rev 5050
evolve: use utility._singlesuccessor() in _solveunstable()
Pierre-Yves David <pierre-yves.david@octobus.net> [Wed, 18 Dec 2019 16:21:37 +0100] rev 5049
evolve: preliminary indentation
The whole code will end up in a try/except clause. We pre-indent the code to
clarify the next patch.
Anton Shestakov <av6@dwimlabs.net> [Fri, 03 Jan 2020 23:12:19 +0700] rev 5048
evolve: provide cache argument to obsutil.successorssets() correctly
After 68f3e819d41d (4.3) the 3rd argument to that function is `closest`.
Anton Shestakov <av6@dwimlabs.net> [Fri, 20 Dec 2019 14:08:37 +0700] rev 5047
pullbundle: random.sample() chokes on a spanset, turn it into a list instead
On Python 3.
Anton Shestakov <av6@dwimlabs.net> [Fri, 20 Dec 2019 13:50:31 +0700] rev 5046
pullbundle: use % instead of .format() because % exists for bytes
Anton Shestakov <av6@dwimlabs.net> [Thu, 19 Dec 2019 18:16:18 +0700] rev 5045
tests: test that debugpullbundlecacheoverlap command works
Anton Shestakov <av6@dwimlabs.net> [Thu, 19 Dec 2019 18:02:31 +0700] rev 5044
pullbundle: compatibility for progress in hg <= 4.6
Anton Shestakov <av6@dwimlabs.net> [Thu, 19 Dec 2019 18:01:15 +0700] rev 5043
compat: add hg version to progress compatibility code
Anton Shestakov <av6@dwimlabs.net> [Mon, 09 Dec 2019 19:59:51 +0700] rev 5042
serverminitopic: ask pytype to disable import-error on a block handling hg 4.5
Anton Shestakov <av6@dwimlabs.net> [Mon, 09 Dec 2019 19:58:58 +0700] rev 5041
topic: ask pytype to disable import-error on a block handling hg 4.5
Anton Shestakov <av6@dwimlabs.net> [Mon, 09 Dec 2019 16:40:53 +0700] rev 5040
evolve: remove unused struct import
Anton Shestakov <av6@dwimlabs.net> [Mon, 09 Dec 2019 19:44:10 +0700] rev 5039
compat: ask pytype to ignore some warnings
It's expected to have a bunch of import and attribute errors in this module
because it's for compatibility with different versions of hg and Python.
Although it doesn't explain why a function wouldn't have __code__ attribute? So
far I think it's a quirk of pytype.
Sushil khanchi <sushilkhanchi97@gmail.com> [Wed, 06 Nov 2019 18:15:00 +0530] rev 5038
evolve: extract logic of pre-checking divergence risk to its own function
Extracting the logic (which is used to check if rewriting an obsolete
revision will create divergence) to a function. This function will be used
in upcoming patches of pre-checking.
This is how we check if rewriting an obsolete revision will create
divergence. We need to check two cases:
If there is any
1) non-obsolete successor.
-> easily done by `obsutil.successorssets()`
2) precursor with non-obsolete successor.
-> using `evolvecmd.divergentsets()`
Anton Shestakov <av6@dwimlabs.net> [Tue, 24 Dec 2019 23:31:42 +0700] rev 5037
heptapod-ci: skip test-evolve.t in tests-py2-pure job
It seems pure mode on Mercurial 4.5 was broken in case of this test. The error
was:
Traceback (most recent call last):
File "/tmp/hgtests.YoK65c/install/bin/hg", line 41, in <module>
dispatch.run()
File "/tmp/hgtests.YoK65c/install/lib/python/mercurial/dispatch.py", line 88, in run
status = (dispatch(req) or 0) & 255
File "/tmp/hgtests.YoK65c/install/lib/python/mercurial/dispatch.py", line 183, in dispatch
ret = _runcatch(req)
File "/tmp/hgtests.YoK65c/install/lib/python/mercurial/dispatch.py", line 324, in _runcatch
return _callcatch(ui, _runcatchfunc)
File "/tmp/hgtests.YoK65c/install/lib/python/mercurial/dispatch.py", line 332, in _callcatch
return scmutil.callcatch(ui, func)
File "/tmp/hgtests.YoK65c/install/lib/python/mercurial/scmutil.py", line 154, in callcatch
return func()
File "/tmp/hgtests.YoK65c/install/lib/python/mercurial/dispatch.py", line 314, in _runcatchfunc
return _dispatch(req)
File "/tmp/hgtests.YoK65c/install/lib/python/mercurial/dispatch.py", line 918, in _dispatch
cmdpats, cmdoptions)
File "/tmp/hgtests.YoK65c/install/lib/python/mercurial/dispatch.py", line 673, in runcommand
ret = _runcommand(ui, options, cmd, d)
File "/tmp/hgtests.YoK65c/install/lib/python/mercurial/dispatch.py", line 926, in _runcommand
return cmdfunc()
File "/tmp/hgtests.YoK65c/install/lib/python/mercurial/dispatch.py", line 915, in <lambda>
d = lambda: util.checksignature(func)(ui, *args, **strcmdopt)
File "/tmp/hgtests.YoK65c/install/lib/python/mercurial/util.py", line 1195, in check
return func(*args, **kwargs)
File "/builds/mercurial/evolve/hgext3rd/evolve/evolvecmd.py", line 1603, in evolve
return _performevolve(ui, repo, **opts)
File "/builds/mercurial/evolve/hgext3rd/evolve/evolvecmd.py", line 1726, in _performevolve
targetcat, lastsolved)
File "/builds/mercurial/evolve/hgext3rd/evolve/evolvecmd.py", line 1752, in _solveonerev
lastsolved=lastsolved, stacktmplt=stacktmplt)
File "/builds/mercurial/evolve/hgext3rd/evolve/evolvecmd.py", line 80, in _solveone
lastsolved=lastsolved)
File "/builds/mercurial/evolve/hgext3rd/evolve/evolvecmd.py", line 183, in _solveunstable
keepbranch, b'orphan')
File "/builds/mercurial/evolve/hgext3rd/evolve/evolvecmd.py", line 952, in relocate
fullnode = unfi.changelog.index.partialmatch(sha1)
AttributeError: 'InlinedIndexObject' object has no attribute 'partialmatch'
Pierre-Yves David <pierre-yves.david@octobus.net> [Tue, 10 Dec 2019 00:05:23 +0100] rev 5036
heptapod-ci: test python 4.5 with the pure version of Mercurial
the evolve extension is currently broken on 4.5-pure. So we get it in its own
commit.
Pierre-Yves David <pierre-yves.david@octobus.net> [Thu, 26 Dec 2019 21:23:30 +0100] rev 5035
branching: merge with stable
Anton Shestakov <av6@dwimlabs.net> [Thu, 26 Dec 2019 12:22:49 +0700] rev 5034
pullbundle: kwarg to changegroup.makestream() is called matcher
Caught by pytype.
Sushil khanchi <sushilkhanchi97@gmail.com> [Sun, 01 Dec 2019 12:41:20 +0530] rev 5033
evolve: extract pre merging content-div csets logic to its own function
Pierre-Yves David <pierre-yves.david@octobus.net> [Thu, 26 Dec 2019 20:28:21 +0100] rev 5032
evolve: further simplify a conditional
The case is simple enough to be handled on its own.
Pierre-Yves David <pierre-yves.david@octobus.net> [Thu, 26 Dec 2019 20:23:27 +0100] rev 5031
evolve: simplify some conditional
extracted from a changeset of Sushil khanchi <sushilkhanchi97@gmail.com>
Pierre-Yves David <pierre-yves.david@octobus.net> [Thu, 26 Dec 2019 20:17:01 +0100] rev 5030
evolve: rename some variable ahead of time
A refactoring of the content divergence resolution in on its way. use a
distinctive name for some variable early to make the change easier to read
later.
Sushil khanchi <sushilkhanchi97@gmail.com> [Thu, 28 Nov 2019 19:01:28 +0530] rev 5029
evolve: add logic to cover --stop when "divergent" relocation in play
Changes in test file reflect the added behaviour. The intermediate changeset has
been properly cleaned up by the `hg evolve --stop` call.
Sushil khanchi <sushilkhanchi97@gmail.com> [Thu, 28 Nov 2019 17:58:45 +0530] rev 5028
evolve: test to show that --stop is not working when "divergent" relocated
Right now, --stop leave the intermediate changeset (relocating the divergent
changeset to its new location behind). This this is an incomplete step it should
be removed by `hg --stop`.
Sushil khanchi <sushilkhanchi97@gmail.com> [Wed, 27 Nov 2019 22:52:08 +0530] rev 5027
evolve: add test to show that --abort is working fine
When resolving content-divergence where relocation of "divergent"
changeset is also required, this test file make sure that --abort
option is working fine.
Sushil khanchi <sushilkhanchi97@gmail.com> [Fri, 20 Dec 2019 17:39:44 +0530] rev 5026
evolve: fix a bug in logic to choose destination when no active topic
Sushil khanchi <sushilkhanchi97@gmail.com> [Fri, 20 Dec 2019 17:32:07 +0530] rev 5025
evolve: add test to demonstrate a bug when no active topic
Added test shows that the logic to find the destination for
`hg up` in the case when "working directory parent has a topic
but wdir has no active topic" is buggy.
This will be fixed in next patch.
Sushil khanchi <sushilkhanchi97@gmail.com> [Fri, 20 Dec 2019 17:31:57 +0530] rev 5024
evolve: fix a bug in logic of choosing destination for hg update
Sushil khanchi <sushilkhanchi97@gmail.com> [Fri, 20 Dec 2019 17:17:30 +0530] rev 5023
evolve: add test to demonstrate a bug when active topic is empty
Added test shows that when current active topic is empty and we try
to run `hg up`, it fails with an error. It's because this case is not
correctly handled when we find a destination to update (when topics
are in play)
Next patch will fix this.
Martin von Zweigbergk <martinvonz@google.com> [Thu, 19 Dec 2019 21:36:59 -0800] rev 5022
evolvecmd: fix a typo node->None
Martin von Zweigbergk <martinvonz@google.com> [Thu, 19 Dec 2019 21:15:31 -0800] rev 5021
evolvecmd: avoid a ctx->node->ctx conversion
Martin von Zweigbergk <martinvonz@google.com> [Thu, 19 Dec 2019 13:06:29 -0800] rev 5020
evolvecmd: mark relocate() private
Martin von Zweigbergk <martinvonz@google.com> [Thu, 19 Dec 2019 22:31:13 -0800] rev 5019
cleanup: replace .parents()[0] by .p1()
Sushil khanchi <sushilkhanchi97@gmail.com> [Wed, 27 Nov 2019 20:54:48 +0530] rev 5018
evolve: add more tests for --continue case when relocating "divergent"
Sushil khanchi <sushilkhanchi97@gmail.com> [Wed, 27 Nov 2019 19:50:33 +0530] rev 5017
evolve: use meaningful variable names for a return value
Sushil khanchi <sushilkhanchi97@gmail.com> [Wed, 27 Nov 2019 12:34:19 +0530] rev 5016
evolve: add logic to cover --continue case when relocating "divergent"
Changes in test file demonstrate the added behaviour.
Sushil khanchi <sushilkhanchi97@gmail.com> [Mon, 25 Nov 2019 21:28:52 +0530] rev 5015
evolve: add test for `--continue` case when relocating "divergent"
This test shows that for now we don't handle the continue case when
relocating "divergent" cset hit merge conflict while resolving
content-divergence.
Next patch will be adding the support for --continue case.
Sushil khanchi <sushilkhanchi97@gmail.com> [Mon, 25 Nov 2019 19:12:51 +0530] rev 5014
evolve: use more specific key name to store in evolvestate
Now, as we also relocate divergent cset if required, so let's use
more specific key to track the relocation of "divergent" and "other".
Sushil khanchi <sushilkhanchi97@gmail.com> [Fri, 13 Dec 2019 01:29:27 +0530] rev 5013
evolve: move a test where it makes more sense
Sushil khanchi <sushilkhanchi97@gmail.com> [Thu, 12 Dec 2019 17:22:18 +0530] rev 5012
evolve: remove the unnecessary condition to check empty successors set
I annotated the history of this "newer == [()]" condition and found
that it was added in the initial stages of evolution project (more than
8 years ago) and there is no test where we get [()] as a successor set.
So looks like "if not newer" is enough to check if it is empty.
I also looked into obsutil.successorssets() implementation and
I don't think it would give us [()] in any case.
Martin von Zweigbergk <martinvonz@google.com> [Tue, 19 Nov 2019 14:54:13 -0800] rev 5011
evolve: delete redundant and incorrect code for updating dirstate
In relocate() when we notice conflicts, we will try to fix up the
dirstate by calling copies.duplicatecopies() and
dirstatedance(). However, we had called _evolvemerge() just before
that point and that has already fixed up the dirstate, so there's no
need to do it again. Also, we did it incorrectly as the test case
shows.
Martin von Zweigbergk <martinvonz@google.com> [Tue, 19 Nov 2019 14:24:31 -0800] rev 5010
tests: demonstrate how evolve loses transitive copies when interrupted
Let's say `b` has been renamed to `c` in commit A and then renamed
to `d` in commit B. We now modify A in a way that will cause a
conflict when we evolve B. `hg evolve` will now stop and let the user
resolve the conflicts, but it will have lost the copy information in
the working copy (should still be c->d).
Martin von Zweigbergk <martinvonz@google.com> [Mon, 09 Dec 2019 10:36:11 -0800] rev 5009
evolve: use ui.configoverride() instead of ui.backupconfig()
ui.configoverride() was added in hg 4.1. util.nullcontextmanager() was
added in hg 4.4.
Martin von Zweigbergk <martinvonz@google.com> [Mon, 09 Dec 2019 10:31:34 -0800] rev 5008
topic: use ui.configoverride() instead of ui.backupconfig()
ui.configoverride() was added in hg 4.1. util.nullcontextmanager() was
added in hg 4.4.
Martin von Zweigbergk <martinvonz@google.com> [Mon, 09 Dec 2019 10:28:58 -0800] rev 5007
topic: use ui.configoverride() instead of ui.backupconfig()
ui.configoverride() was added in hg 4.1. util.nullcontextmanager() was
added in hg 4.4.
Martin von Zweigbergk <martinvonz@google.com> [Mon, 09 Dec 2019 10:33:48 -0800] rev 5006
evolve: use ui.configoverride() instead of ui.backupconfig()
ui.configoverride() was added in hg 4.1.
Martin von Zweigbergk <martinvonz@google.com> [Mon, 09 Dec 2019 10:43:04 -0800] rev 5005
evolve: remove obsolete backup of allowemptycommit config
We stopped setting the config in 88601e1cd5d8 (evolve: create new
commit while resolving content-divergence, 2018-04-23), so there's no
need to back it up.
Martin von Zweigbergk <martinvonz@google.com> [Mon, 09 Dec 2019 10:44:42 -0800] rev 5004
evolve: remove dead initilization of "newnode" variable
It's been dead since f9dad99a90d5 (evolve: create a new commit instead
of amending one of the divergents, 2018-06-13).
Sushil khanchi <sushilkhanchi97@gmail.com> [Thu, 12 Dec 2019 17:16:02 +0530] rev 5003
evolve: fix a typo
Pierre-Yves David <pierre-yves.david@octobus.net> [Tue, 10 Dec 2019 21:03:51 +0100] rev 5002
changelog: mention the python 3 fix
Pierre-Yves David <pierre-yves.david@octobus.net> [Tue, 10 Dec 2019 21:03:23 +0100] rev 5001
changelog: mention the date fix for content divergence
Pierre-Yves David <pierre-yves.david@octobus.net> [Tue, 10 Dec 2019 20:47:13 +0100] rev 5000
branching: merge with stable
Denis Laxalde <denis.laxalde@logilab.fr> [Tue, 10 Dec 2019 20:35:56 +0100] rev 4999
py3: fix several TypeError formatting bytes with an exception object
Manuel Jacob <me@manueljacob.de> [Fri, 11 Oct 2019 21:32:27 +0200] rev 4998
evolve: preserve date when resolving content-divergence (issue6202)
This patch changes the implementation to do a 3 way merge. In case of actual
conflict, it take the newest date from both sides of the content-divergence.
Pierre-Yves David <pierre-yves.david@octobus.net> [Tue, 10 Dec 2019 19:58:51 +0100] rev 4997
test: add test for content divergence involving a date update
In this case, each side update the date.
Pierre-Yves David <pierre-yves.david@octobus.net> [Tue, 10 Dec 2019 19:55:37 +0100] rev 4996
test: add test for content divergence involving a date update
In this case, one side update to an earlier date.
Pierre-Yves David <pierre-yves.david@octobus.net> [Tue, 10 Dec 2019 19:27:55 +0100] rev 4995
test: add test for content divergence involving a date update
In this case, one side update to an older date.
Pierre-Yves David <pierre-yves.david@octobus.net> [Tue, 10 Dec 2019 19:13:06 +0100] rev 4994
test: highlight data handling when solving divergence
The current handling of date is lacking, we add test to highlight this.
Sushil khanchi <sushilkhanchi97@gmail.com> [Sat, 07 Dec 2019 03:27:50 +0530] rev 4993
evolve: add tests for the case when resolution parent is ambiguous
Sushil khanchi <sushilkhanchi97@gmail.com> [Fri, 06 Dec 2019 23:43:00 +0530] rev 4992
evolve: abort if deciding resolution parent is ambiguous
While solving content-divergence, it could be hard to decide
which cset should be treated as a resolution parent if there
are multiple successors of divergent cset. i.e split case.
But we are planning to make some update in handling this split
case:
1) consider highest one if all splitted csets are in a single
topological branch
2) still need to decide if 1) doesn't apply