Mads Kiilerich <mads@kiilerich.com> [Wed, 04 Mar 2009 23:23:59 +0100] rev 7835
Optimization of pure.base85.b85encode
This makes pure python base85 encoding 3x faster than before. Now it is only
40x slower than the C version.
Matt Mackall <mpm@selenic.com> [Mon, 09 Mar 2009 11:41:38 -0500] rev 7834
Merge with crew-stable
Benoit Boissinot <benoit.boissinot@ens-lyon.org> [Sat, 07 Mar 2009 20:16:03 +0100] rev 7833
verify: find correct first corrupted cset for missing/corrupted revlogs
Benoit Boissinot <benoit.boissinot@ens-lyon.org> [Sat, 07 Mar 2009 20:04:56 +0100] rev 7832
verify: do not abort on fully corrupted revlog
Benoit Boissinot <benoit.boissinot@ens-lyon.org> [Sat, 07 Mar 2009 00:59:41 +0100] rev 7831
hgweb: fix allow_read permissions when a user is specified
Thanks to Juan Pablo Aroztegi for reporting and debugging it.
Mads Kiilerich <mads@kiilerich.com> [Thu, 05 Mar 2009 14:53:59 +0100] rev 7830
test-identify: require no-outer-repo
The test expects "There is no Mercurial repository here (.hg not found)"
Benoit Allard <benoit@aeteurope.nl> [Thu, 05 Mar 2009 10:38:39 +0100] rev 7829
update copyright date in version information
Dongsheng Song <dongsheng.song@gmail.com> [Thu, 05 Mar 2009 14:44:13 +0800] rev 7828
Typo fix
Steve Borho <steve@borho.org> [Wed, 04 Mar 2009 22:56:17 -0600] rev 7827
update copyright in ReadMe.html
Matt Mackall <mpm@selenic.com> [Fri, 06 Mar 2009 09:15:47 -0600] rev 7826
setup.py: fix version detection for tarball
- don't do version detection if there's no .hg directory
- shrink try: clause
- don't write __version__.py if version is unknown
(we might overwrite the real version)
Matt Mackall <mpm@selenic.com> [Wed, 04 Mar 2009 19:04:21 -0600] rev 7825
Added signature for changeset
11efa41037e2
Matt Mackall <mpm@selenic.com> [Wed, 04 Mar 2009 19:04:20 -0600] rev 7824
Added tag 1.2 for changeset
11efa41037e2
Frank Kingswood <frank@kingswood-consulting.co.uk> [Tue, 03 Mar 2009 21:32:23 +0000] rev 7823
convert: Perforce source for conversion to Mercurial
Matt Mackall <mpm@selenic.com> [Wed, 04 Mar 2009 18:42:24 -0600] rev 7822
remove tests for long-gone imerge extension
Augie Fackler <durin42@gmail.com> [Wed, 04 Mar 2009 13:06:16 -0600] rev 7821
merge: better error messages to lead users to hg update --clean to abandon merges.
Fixes
issue1533.
Matt Mackall <mpm@selenic.com> [Wed, 04 Mar 2009 17:02:16 -0600] rev 7820
audit: be even pickier (
issue1450)
Bill Barry <after.fallout@gmail.com> [Thu, 12 Feb 2009 09:36:15 -0700] rev 7819
dispatch: extract command execution block into method
This pulls the pre-command hook/command/post-command hook workflow out of
the method it is in and puts it into its own method so that it potentially
could be exposed for extensions to wrap.
Benoit Allard <benoit@aeteurope.nl> [Thu, 05 Feb 2009 23:47:31 +0100] rev 7818
bookmarks: change NAME to REV
Alex Unden <alu@zpuppet.org> [Fri, 27 Feb 2009 02:01:45 -0800] rev 7817
bookmarks: fixes bug where a deleted bookmark may still be treated as current when track.current option is set
David Soria Parra <dsp@php.net> [Sun, 01 Mar 2009 00:30:01 +0100] rev 7816
bookmarks: Set current bookmark if we create a new one on the tip
If track.current is enabled we set the newly created bookmark as the
current tracked bookmark. We do not do this if a revision is specified.
Matt Mackall <mpm@selenic.com> [Mon, 02 Mar 2009 19:19:12 -0600] rev 7815
convert: change hg.saverev default to False
This needlessly breaks changeset identifiers for the common case (cleaning
history), which is sad.
Matt Mackall <mpm@selenic.com> [Mon, 02 Mar 2009 19:19:09 -0600] rev 7814
tag: force load of tag cache
Mads Kiilerich <mads@kiilerich.com> [Fri, 27 Feb 2009 19:10:38 +0100] rev 7813
run-tests.py: avoid using popen2.Popen4 - use subprocess instead
Use subprocess to emulate Popen4 if available - similar to how it is done in
util.py.
Using popen2 under python 2.6 gives
DeprecationWarning: The popen2 module is deprecated. Use the subprocess module.
Mads Kiilerich <mads@kiilerich.com> [Fri, 27 Feb 2009 17:52:31 +0100] rev 7812
test-convert-cvs*: mute output from "cvs up"
Apparently some versions of cvs outputs more than other versions. That is
however not relevant for these tests.