Toshi MARUYAMA <marutosijp2@gmail.com> [Mon, 12 Jun 2017 03:22:45 +0900] rev 32795
rpms: remove '%if "%{?pythonver}" != "2.4"'
Mercurial requires python >= 2.7.
Jun Wu <quark@fb.com> [Fri, 02 Jun 2017 20:49:42 -0700] rev 32794
obsstore: do not load all markers to detect duplication
This will make duplication detection something like O(newmarkers) instead of
O(obsstore).
Gregory Szorc <gregory.szorc@gmail.com> [Sat, 10 Jun 2017 11:33:57 -0700] rev 32793
clonebundles: reference correct config option
This option is no longer experimental.
Siddharth Agarwal <sid0@fb.com> [Sat, 10 Jun 2017 14:09:54 -0700] rev 32792
filestat: move __init__ to frompath constructor
We're going to add a `fromfp` constructor soon, and this also allows a filestat
object for a non-existent file to be created.
Siddharth Agarwal <sid0@fb.com> [Sat, 10 Jun 2017 14:07:31 -0700] rev 32791
test-dirstate-race: ensure that a isn't in the lookup set at the end
We're going to rely on this in upcoming patches.
Siddharth Agarwal <sid0@fb.com> [Sat, 10 Jun 2017 14:07:31 -0700] rev 32790
hghave: add test for whether fsmonitor is enabled
This uses the HGFSMONITOR_TESTS environment variable that
fsmonitor-run-tests.py adds.
Siddharth Agarwal <sid0@fb.com> [Sat, 10 Jun 2017 14:07:30 -0700] rev 32789
tests: add a wrapper to run fsmonitor tests
This script does a bunch of non-trivial configuration work: in particular, it
sets up an isolated instance of Watchman which isn't affected by global state
and can be torn down on completion.
This script also sets the HGFSMONITOR_TESTS environment variable, which hghave
will use in the next patch to allow gating on whether fsmonitor is enabled.
With fsmonitor enabled, there appear to be a number of failures in the test
suite. It's not yet clear to me why they're happening, but if someone would
like to jump in and fix some of them I hope this will be helpful for that.
Pierre-Yves David <pierre-yves.david@octobus.net> [Sat, 10 Jun 2017 18:47:09 +0100] rev 32788
setdiscovery: improves logged message
The 'srvheads' list contains all server heads including the common ones. We
adjust 'ui.log' message to provide more useful information about server heads
locally unknown. The performance impact of turning the list to set is
negligible (about 1e-4s) compared to the rest of the discovery cost, so I'm
taking the easy path.
Augie Fackler <raf@durin42.com> [Sat, 10 Jun 2017 10:46:06 -0400] rev 32787
tests: handle variation between pure and normal output in annotate --skip
I'm pretty sure that both results are valid, depending on how you
slice the edits.
Brandon McCaig <bamccaig@gmail.com> [Fri, 09 Jun 2017 20:12:39 -0400] rev 32786
bisect: improve option validation message