Augie Fackler <raf@durin42.com> [Mon, 28 Jul 2014 20:54:14 -0400] rev 21995
test-run-tests.t: add extra data to tests for keyword tests
This adds a fair amount of overall instability in the enclosing .t. As
such, this is performed in its own commit, and a test for --keyword on
run-tests.t will be added in a followup change.
Matt Mackall <mpm@selenic.com> [Mon, 04 Aug 2014 15:24:57 -0500] rev 21994
purge: drop stat import
Matt Mackall <mpm@selenic.com> [Mon, 04 Aug 2014 14:42:24 -0500] rev 21993
run-tests: add iolock to failure output
Matt Mackall <mpm@selenic.com> [Mon, 04 Aug 2014 14:32:34 -0500] rev 21992
merge with stable
Sean Farley <sean.michael.farley@gmail.com> [Fri, 01 Aug 2014 13:09:06 -0500] rev 21991
color: pass on key error for win32 (issue4298)
This is a quick fix for some consoles on windows (consoles that are not mingw
based) so that the debugcolor command doesn't throw a KeyError when effects
aren't supported (e.g. italic).
Siddharth Agarwal <sid0@fb.com> [Fri, 01 Aug 2014 18:30:18 -0700] rev 21990
context: call normal on the right object
dirstate.normal is the method that marks files as unchanged/normal.
Rev 20a30cd41d21 started caching dirstate.normal in order to improve
performance. However, there was an error in the patch: taking the wlock, under
some conditions depending on platform, can cause a new dirstate object to be
created. Caching dirstate.normal before calling wlock would then cause the
fixup calls below to be on the old dirstate object, effectively disappearing
into the ether.
On Unix and Unix-like OSes, the condition under which we create a new dirstate
object is 'the dirstate file has been modified since the last time we opened
it'. This happens pretty rarely, so the object is usually the same -- there's
little impact.
On Windows, the condition is 'always'. This means files in the lookup state are
never marked normal, so the bug has a serious performance impact since all the
files in the lookup state are re-read every time hg status is run.