Pulkit Goyal <7895pulkit@gmail.com> [Tue, 26 Jun 2018 01:08:47 +0530] rev 38474
py3: add b'' prefixes in tests/test-bundle2-pushback.t
This makes the test run on Python 3.
# skip-blame because just b'' prefixes.
Differential Revision: https://phab.mercurial-scm.org/D3834
Pulkit Goyal <7895pulkit@gmail.com> [Tue, 26 Jun 2018 00:37:02 +0530] rev 38473
py3: make tests/test-diff-antipatience.t work with python 3
On python3, if we do list(b'abcd'), we get [97, 98, 99, 100]. So this patch does
the list() operation on unicodes and then convert things to bytes.
This also adds `and None` to suppress return values of .write() calls.
Differential Revision: https://phab.mercurial-scm.org/D3833
Martin von Zweigbergk <martinvonz@google.com> [Mon, 25 Jun 2018 11:04:17 -0700] rev 38472
terse: pass "clean" and "unknown" booleans by name for clarity
Differential Revision: https://phab.mercurial-scm.org/D3838
Martin von Zweigbergk <martinvonz@google.com> [Mon, 25 Jun 2018 11:01:11 -0700] rev 38471
terse: add tests of running from subdirectory
I also included comments explaining what I think the output should be
(discussion started on D3628).
Differential Revision: https://phab.mercurial-scm.org/D3837
Yuya Nishihara <yuya@tcha.org> [Sat, 23 Jun 2018 19:23:53 +0900] rev 38470
convert: don't pass --no-files to "darcs show repo" command
The test fails with darcs 2.14.0 because --no-files is no longer supported.
It was removed at the patch 97973a52bf496657558412562d6fad2ee651b1e0, which
says "show repo: removed --files option, removed manual flags parsing."
As far as I can tell, --no-files was specified just for optimization, so old
darcs should work without it.
Boris Feld <boris.feld@octobus.net> [Tue, 29 May 2018 00:15:44 +0200] rev 38469
shelve: stop testing missing rebase state file
We no longer use the file. Neither during rebase not during continue.
Differential Revision: https://phab.mercurial-scm.org/D3691
Boris Feld <boris.feld@octobus.net> [Tue, 29 May 2018 00:12:35 +0200] rev 38468
shelve: actually test corrupted shelve state
The test was previous moving the corrupted shelve state file as an histedit
state file.
Differential Revision: https://phab.mercurial-scm.org/D3690
Boris Feld <boris.feld@octobus.net> [Tue, 29 May 2018 00:13:48 +0200] rev 38467
shelve: directly handle `--continue`
Shelve is currently sub-contracting some of its work to the rebase extension.
In order to make shelve more independent and flexible we would like shelve to
handle the parent alignment directly.
This changeset takes on the next step, handling the abort process. Same as for
--abort. It turns out we have all the necessary bits in the `shelvestate`
file. So we do not need anything from the interrupted rebase.
Differential Revision: https://phab.mercurial-scm.org/D3689
Boris Feld <boris.feld@octobus.net> [Mon, 28 May 2018 18:15:21 +0200] rev 38466
shelve: directly handle the abort process
Shelve is currently sub-contracting some of its work to the rebase extension.
In order to make shelve more independent and flexible we would like shelve to
handle the parent alignment directly.
This change starts with the simplest bits, handling the abort process. It
turns out we have all the necessary bits in the `shelvestate` file. So we do
not need anything from the interrupted rebase.
As a nice side effect, a test about missing `shelverebasestate` state file now
behave better.
Differential Revision: https://phab.mercurial-scm.org/D3688
Boris Feld <boris.feld@octobus.net> [Mon, 28 May 2018 20:51:20 +0200] rev 38465
shelve: check the actual shelvestate in morestatus
The rebasestate is a lower level implementation details that we are trying to
remove.
Differential Revision: https://phab.mercurial-scm.org/D3687