Matt Mackall <mpm@selenic.com> [Thu, 03 May 2007 17:24:43 -0500] rev 4399
merge: expand and simplify the invalid handling for directory moves
Matt Mackall <mpm@selenic.com> [Thu, 03 May 2007 17:24:43 -0500] rev 4398
merge: fix a bug detecting directory moves
When all the files in a directory are moved, it may be incorrectly marked as moved even if it contains subdirectories.
Matt Mackall <mpm@selenic.com> [Thu, 03 May 2007 17:24:43 -0500] rev 4397
merge: clarify the findcopies code
Brendan Cully <brendan@kublai.com> [Tue, 01 May 2007 13:23:32 -0700] rev 4396
mq: account for readheaders munging in qrefresh
Matt Mackall <mpm@selenic.com> [Tue, 01 May 2007 14:42:54 -0500] rev 4395
verify: hold write lock
Verify will report spurious errors if commits or pulls happen during verify.
Brendan Cully <brendan@kublai.com> [Mon, 30 Apr 2007 09:32:44 -0700] rev 4394
Also restore old rm -A behavior for inexact patterns
Brendan Cully <brendan@kublai.com> [Mon, 30 Apr 2007 09:13:27 -0700] rev 4393
Fix rm -A with no patterns, which I just broke.
Brendan Cully <brendan@kublai.com> [Mon, 30 Apr 2007 08:51:59 -0700] rev 4392
Make rm --after simply mark files as removed, unless --force is also given
Michael Gebetsroither <michael.geb@gmx.at> [Sun, 29 Apr 2007 18:17:35 -0700] rev 4391
Add hgwebdir.fcgi to contrib
Brendan Cully <brendan@kublai.com> [Sun, 29 Apr 2007 17:35:03 -0700] rev 4390
Flesh out bisect help text
Matt Mackall <mpm@selenic.com> [Sun, 29 Apr 2007 12:33:24 -0500] rev 4389
Merge with crew
Matt Mackall <mpm@selenic.com> [Thu, 26 Apr 2007 18:41:18 -0500] rev 4388
Merge with stable
Matt Mackall <mpm@selenic.com> [Wed, 25 Apr 2007 13:35:18 -0500] rev 4387
hg serve: don't do DNS lookups
Alexis S. L. Carvalho <alexis@cecm.usp.br> [Sat, 28 Apr 2007 17:55:45 -0300] rev 4386
Fix minor mismerge.
The caching of umask ended up being duplicated.
Alexis S. L. Carvalho <alexis@cecm.usp.br> [Sat, 28 Apr 2007 17:55:45 -0300] rev 4385
run-tests.py: pass -L/--label to simplemerge
This should give us conflict markers that don't change from one run
to another.
Remove some sed's/egrep's from some tests.
Lee Cantey <lcantey@gmail.com> [Thu, 26 Apr 2007 23:58:18 -0700] rev 4384
Add notes about configuring a merge program
Lee Cantey <lcantey@gmail.com> [Fri, 27 Apr 2007 16:04:17 -0700] rev 4383
Update copyright year and change to range
Patrick Mezard <pmezard@gmail.com> [Sat, 28 Apr 2007 11:53:36 +0200] rev 4382
Use patch.patch() when checking for patch availability.
The test is closer to what we want to know. It also makes sense when you deal with patched versions of patch() as most people do under win32.
Brendan Cully <brendan@kublai.com> [Sat, 28 Apr 2007 09:03:47 -0700] rev 4381
Merge with crew
Patrick Mezard <pmezard@gmail.com> [Sat, 28 Apr 2007 11:43:31 +0200] rev 4380
Fix find_in_path not including some file extension logic under win32.
Windows shell resolves utility path by combining PATH, the utility name and a set of file extensions from PATHEXT.
Joel Rosdahl <joel@rosdahl.net> [Sat, 28 Apr 2007 16:28:54 +0200] rev 4379
Fix bad reference to RepoError in import command
Bryan O'Sullivan <bos@serpentine.com> [Fri, 27 Apr 2007 21:30:55 -0700] rev 4378
Refactor commands.serve to allow other commands to run as services.
This introduces a new function, cmdutil.service.
Brendan Cully <brendan@kublai.com> [Wed, 25 Apr 2007 13:14:01 -0700] rev 4377
Do not automatically rename an atomicfile if a write to it has generated an exception.
Thomas Arendsen Hein <thomas@intevation.de> [Wed, 25 Apr 2007 18:43:55 +0200] rev 4376
merge with stable
Thomas Arendsen Hein <thomas@intevation.de> [Sat, 27 Jan 2007 23:07:06 +0100] rev 4375
Fix segfaults when parsing bdiff hunks in mpatch.decode() and .patchedsize()
- fix off by 11 when checking if there are more hunks (found by Maris Fogels)
- bail out if start is greater than end
- check if new hunk starts after start/end/len block of current hunk as
the pointer can wrap around on very large values, reproducible with
import mpatch; mpatch.patchedsize(12, 'x'*12)
Brendan Cully <brendan@kublai.com> [Tue, 24 Apr 2007 18:43:18 -0700] rev 4374
Make copy --after work for files that have been hg added.
Bryan O'Sullivan <bos@serpentine.com> [Tue, 24 Apr 2007 12:02:51 -0700] rev 4373
When reloading the dirstate, recompute ignore information if needed.
Bryan O'Sullivan <bos@serpentine.com> [Tue, 24 Apr 2007 12:02:42 -0700] rev 4372
dirstate: speed up write by 50%.
Bryan O'Sullivan <bos@serpentine.com> [Tue, 24 Apr 2007 11:46:15 -0700] rev 4371
dirstate: make parents() faster.
This allows localrepository.status() to avoid reading all but the
first 40 bytes of the dirstate in the common case of running "hg
status".
Bryan O'Sullivan <bos@serpentine.com> [Tue, 24 Apr 2007 11:05:39 -0700] rev 4370
localrepository.status: only acquire wlock if actually needed.
This speeds up the common case of not needing to update the dirstate,
and avoids the need to reload and parse the dirstate "just in case".
Bryan O'Sullivan <bos@serpentine.com> [Tue, 24 Apr 2007 10:53:25 -0700] rev 4369
util._matcher: speed up regexp matching.
In 4babaa52badf, Benoit made a change that substantially slows matching
when a big .hgignore file is in play, because it calls into the regexp
matching engine potentially hundreds of times per file to be matched.
I've partly rolled back his change, so that we only call into the matcher
once per file, but preserved the ability to report a meaningful error
message if there's a syntax error in the regexp.
Brendan Cully <brendan@kublai.com> [Tue, 24 Apr 2007 10:47:41 -0700] rev 4368
Merge with crew-stable