Boris Feld <boris.feld@octobus.net> [Fri, 12 Jan 2018 11:09:04 +0000] rev 35608
visibility: fix a comment introduced before which is not up-to-date
265cd9e19d26 introduced a comment in _filterederror that was not updated with
the latest iterations of the patch, fix the comment.
Differential Revision: https://phab.mercurial-scm.org/D1851
Gregory Szorc <gregory.szorc@gmail.com> [Wed, 10 Jan 2018 19:24:58 -0800] rev 35607
rust: avoid redundant 'static lifetime
'static is apparently automatic for const variables.
Differential Revision: https://phab.mercurial-scm.org/D1846
Boris Feld <boris.feld@octobus.net> [Thu, 11 Jan 2018 11:57:59 +0000] rev 35606
pull: hold wlock for the full operation when --update is used
With now, the wlock is not held between the pull and the update. This can lead
to race condition and make logic checking to post pull results more complicated
(eg: with _afterlock).
Anton Shestakov <av6@dwimlabs.net> [Thu, 11 Jan 2018 18:20:08 +0800] rev 35605
hgweb: drop support of browsers that don't understand <canvas> (BC)
Internet Explorer 8 and below need excanvas aka ExplorerCanvas to support
canvas tag at all. We used to vendor the library in Mercurial, but IE8 was
discontinued and all support for it ceased in January 2016. We should do the
same and remove excanvas library.
Apart from just cleaning up code, this will also make downstream Debian
packages stop depending on libjs-excanvas, for example.
Gregory Szorc <gregory.szorc@gmail.com> [Wed, 10 Jan 2018 19:36:45 -0800] rev 35604
rust: add TODO about lifetime of program_name variable
Per review comment in D1581.
Differential Revision: https://phab.mercurial-scm.org/D1847
Gregory Szorc <gregory.szorc@gmail.com> [Wed, 10 Jan 2018 19:23:36 -0800] rev 35603
rust: move import of PathBuf
By moving it to the single function that uses it, we eliminate a
#cfg.
Differential Revision: https://phab.mercurial-scm.org/D1845
Jun Wu <quark@fb.com> [Wed, 10 Jan 2018 21:09:57 -0800] rev 35602
perf: do not import util.queue
This was added by
af25237be091. But util.queue is not a module and is not
importable. I guess it wasn't noticed because of demandimport.
Differential Revision: https://phab.mercurial-scm.org/D1848
Gregory Szorc <gregory.szorc@gmail.com> [Wed, 10 Jan 2018 19:08:51 -0800] rev 35601
rust: move Cargo.lock
If you do a `cargo build` in rust/, Cargo will create rust/Cargo.lock.
Furthermore, the previous rust/hgcli/Cargo.lock appears to not even be
consulted because the auto-generated rust/Cargo.lock varied from
rust/hgcli/Cargo.lock.
This commit moves Cargo.lock to where Cargo expects it to be.
While we're here, bump the version of libc to the latest available.
Differential Revision: https://phab.mercurial-scm.org/D1844
Gregory Szorc <gregory.szorc@gmail.com> [Wed, 10 Jan 2018 19:04:52 -0800] rev 35600
run-tests: fix regular expression for path test
The previous regexp would match the empty string. This fixes
a mistake added in
964212780daf.
Differential Revision: https://phab.mercurial-scm.org/D1843
Boris Feld <boris.feld@octobus.net> [Sun, 17 Dec 2017 04:31:27 +0100] rev 35599
perf: add threading capability to perfbdiff
Since we are releasing the GIL during diffing, it is interesting to see how a
thread pool would perform on diffing. We add a new `--threads` argument to
commands. Synchronizing the thread pool is a bit complex because we want to be
able to reuse it from one run to another.
On my computer (i7 with 4 cores + hyperthreading), I get the following data for
about 12000 revisions:
threads wall comb wall gain comb overhead
none 31.596715 31.59 0.00% 0.00%
1 31.621228 31.62 -0.08% 0.09%
2 16.406202 32.8 48.08% 3.83%
3 11.598334 34.76 63.29% 10.03%
4 9.205421 36.77 70.87% 16.40%
5 8.517604 42.51 73.04% 34.57%
6 7.94645 47.58 74.85% 50.62%
7 7.434972 51.92 76.47% 64.36%
8 7.070638 55.34 77.62% 75.18%
Compared to the feature disabled (threads=0), the overhead is negligible with
the threading code (threads=1), and the gain is already 48% with two threads.
Paul Morelle <paul.morelle@octobus.net> [Wed, 10 Jan 2018 17:24:25 +0100] rev 35598
upgraderepo: select correct deltareuse depending on actions
Only 'redeltafulladd' was taken into account because of a small typo.
Yuya Nishihara <yuya@tcha.org> [Tue, 02 Jan 2018 18:35:11 +0900] rev 35597
revsetlang: add %p specifier to format list of function arguments
This will be used to construct parameters passed to _matchfiles().
Yuya Nishihara <yuya@tcha.org> [Tue, 02 Jan 2018 21:05:40 +0900] rev 35596
revsetlang: unnest inner functions from formatspec()
Yuya Nishihara <yuya@tcha.org> [Sat, 01 Apr 2017 17:51:56 +0900] rev 35595
revsetlang: fix quoting of %ls string
Before, "'" wasn't escaped appropriately. This also changes the separator
'\0' to '\\0', but that's okay as a string token is unescaped.