Steve Borho <steve@borho.org> [Fri, 07 Dec 2007 18:18:11 -0600] rev 5624
win32: it is no longer necessary to bundle patch.exe
Mercurial can use it's built-in mpatch library
Steve Borho <steve@borho.org> [Fri, 07 Dec 2007 18:15:24 -0600] rev 5623
win32: do not install bash hgmerge on Windows
Steve Borho <steve@borho.org> [Fri, 07 Dec 2007 18:14:16 -0600] rev 5622
win32: ignore all installers generated in Output/
Patrick Mezard <pmezard@gmail.com> [Sun, 09 Dec 2007 15:25:36 +0100] rev 5621
convert: move commands definition to ease demandload job (issue 860)
Matt Mackall <mpm@selenic.com> [Fri, 07 Dec 2007 14:59:33 -0600] rev 5620
Merge with crew
Steve Borho <steve@borho.org> [Wed, 05 Dec 2007 22:57:35 -0600] rev 5619
win32: favor local mercurial.ini over registry key
Hopefully this is the best of both worlds. Multiple py2exe based
installers can coexist on one machine, while source and lib based
Mercurial users can still use a registry key in lieu of resorting to
HGRCPATH.
Bryan O'Sullivan <bos@serpentine.com> [Thu, 06 Dec 2007 13:11:36 -0800] rev 5618
Merge with stable
David J. Mellor <dmellor@whistlingcat.com> [Thu, 06 Dec 2007 13:10:25 -0800] rev 5617
convert: fix SVN date parser dropping the final whole second digit
Christian Ebert <blacktrash@gmx.net> [Wed, 05 Dec 2007 12:48:33 +0100] rev 5616
highlight: adapt to hgweb_mode refactoring
Bryan O'Sullivan <bos@serpentine.com> [Thu, 06 Dec 2007 09:55:56 -0800] rev 5615
Merge with -stable
Jim Hague <jim.hague@acm.org> [Tue, 23 Oct 2007 10:39:24 +0000] rev 5614
Fix bdiff test failures on AIX.
The test fails when, in the call to calloc(), the number of elements (bn)
is 0. In that case, calloc() on AIX will return NULL, while the code expects
a valid heap pointer. Both results are permissible under C99, Unix98 etc.
Work around by ensuring that at least 1 element is requested.
Steve Borho <steve@borho.org> [Mon, 03 Dec 2007 17:28:26 -0600] rev 5613
workaround for raw_input() on Windows
Since change a3fe91b4f6eb, Mercurial has set_binary() on stdin, stdout, and stderr.
On Windows, this had the side effect of causing raw_input() to emit trailing carriage
returns on it's returned strings. This was causing web authentication problems.
Patrick Mezard <pmezard@gmail.com> [Tue, 04 Dec 2007 23:28:10 +0100] rev 5612
Merge with crew-stable
Patrick Mezard <pmezard@gmail.com> [Tue, 04 Dec 2007 23:10:13 +0100] rev 5611
utils_win32: calling SHGetPathFromIDList under NT is useless and broken
The call was done under NT when running as a service (CGI script for instance) and always crash, even in interactive processes.
Matt Mackall <mpm@selenic.com> [Fri, 07 Dec 2007 02:29:55 -0600] rev 5610
copy: handle rename internally
- add rename parameter
- don't collect copied list
- report "moving" rather than "copying"
- remove renamed files
- don't return copied list
- simplify callers
Matt Mackall <mpm@selenic.com> [Fri, 07 Dec 2007 02:03:42 -0600] rev 5609
copy: move rename logic
Matt Mackall <mpm@selenic.com> [Fri, 07 Dec 2007 02:01:13 -0600] rev 5608
copy: simplify inner copy
- save dirstate lookup and exists check
- do all fs work inside a single dryrun clause
- move unlinking into inner copy section
- move target dir creation
- eliminate undelete mess
- fix a bug on mv a -> b -> a after merge
Matt Mackall <mpm@selenic.com> [Fri, 07 Dec 2007 02:01:10 -0600] rev 5607
copy: minor cleanups
- add after and dryrun variables
- add some comments
Matt Mackall <mpm@selenic.com> [Thu, 06 Dec 2007 15:15:47 -0600] rev 5606
copy: propagate errors properly
Matt Mackall <mpm@selenic.com> [Thu, 06 Dec 2007 14:58:57 -0600] rev 5605
copy: refactor okaytocopy into walkpat
- rename core copy function to copyfile
- move origsrc details into copyfile
- turn okaytocopy loop into walkpat
Matt Mackall <mpm@selenic.com> [Sun, 02 Dec 2007 18:41:22 -0600] rev 5604
copy: fix copying back with -A (issue836)
Dirkjan Ochtman <dirkjan@ochtman.nl> [Mon, 03 Dec 2007 19:19:12 +0100] rev 5603
hgwebdir: refactor inner loop
Dirkjan Ochtman <dirkjan@ochtman.nl> [Mon, 03 Dec 2007 18:58:18 +0100] rev 5602
hgwebdir: split out templater creation
Dirkjan Ochtman <dirkjan@ochtman.nl> [Mon, 03 Dec 2007 18:40:37 +0100] rev 5601
hgwebdir: split out makeindex function, facilitate test failure diagnosis
Dirkjan Ochtman <dirkjan@ochtman.nl> [Mon, 03 Dec 2007 13:30:08 +0100] rev 5600
hgweb: explicitly pass around the templater
Dirkjan Ochtman <dirkjan@ochtman.nl> [Mon, 03 Dec 2007 12:40:29 +0100] rev 5599
hgweb: split out templater definition
Dirkjan Ochtman <dirkjan@ochtman.nl> [Mon, 03 Dec 2007 12:27:11 +0100] rev 5598
separate the wire protocol commands from the user interface commands
Dirkjan Ochtman <dirkjan@ochtman.nl> [Mon, 03 Dec 2007 12:06:21 +0100] rev 5597
hgweb: extract constant to global level
Dirkjan Ochtman <dirkjan@ochtman.nl> [Mon, 03 Dec 2007 11:58:52 +0100] rev 5596
hgweb: get rid of some nested functions
Dirkjan Ochtman <dirkjan@ochtman.nl> [Mon, 03 Dec 2007 11:58:29 +0100] rev 5595
fix some imports; pushing bz2 bundles over http is currently untested
Bryan O'Sullivan <bos@serpentine.com> [Sun, 02 Dec 2007 21:54:56 -0800] rev 5594
coverage: return to not assuming that python is in /usr/bin
Bryan O'Sullivan <bos@serpentine.com> [Sun, 02 Dec 2007 21:52:40 -0800] rev 5593
Fix breaking test from 05451f6b5f07
Dirkjan Ochtman <dirkjan@ochtman.nl> [Sun, 02 Dec 2007 23:26:40 +0100] rev 5592
import latest coverage.py version
Dirkjan Ochtman <dirkjan@ochtman.nl> [Sun, 02 Dec 2007 23:26:40 +0100] rev 5591
split out hgweb commands into a separate file, move some code around
Dirkjan Ochtman <dirkjan@ochtman.nl> [Sun, 02 Dec 2007 23:26:40 +0100] rev 5590
add more tests for hgweb commands
Matt Mackall <mpm@selenic.com> [Sun, 02 Dec 2007 18:11:59 -0600] rev 5589
move commands.docopy to cmdutil.copy
Matt Mackall <mpm@selenic.com> [Sun, 02 Dec 2007 17:04:16 -0600] rev 5588
churn: avoid division by zero
Matt Mackall <mpm@selenic.com> [Sun, 02 Dec 2007 16:26:56 -0600] rev 5587
update test output
Patrick Mezard <pmezard@gmail.com> [Sun, 02 Dec 2007 19:39:29 +0100] rev 5586
hgwebdir: test virtual paths leading or trailing separators are stripped
Patrick Mezard <pmezard@gmail.com> [Sun, 02 Dec 2007 19:45:38 +0100] rev 5585
Merge with crew-stable
Patrick Mezard <pmezard@gmail.com> [Sun, 02 Dec 2007 19:39:27 +0100] rev 5584
hgwebdir: normalize virtual paths before stripping the separator
It is not even clear that virtual paths should be normalized as normal paths at all, we could expect slash to be the natural separator.
Steve Borho <steve@borho.org> [Sun, 02 Dec 2007 17:15:03 +0100] rev 5583
win32: read system rcpath from registry
Using the module name was not always helpful. It breaks down
when Mercurial is installed as source and when the Mercurial
libs are used by external applications.
This patch allows Mercurial installers to store the system wide
rcpath in the registry, where it can always be found. HGRCPATH
is a poor option for storing the system wide rcpath, since it
overrides both the system and user rcpaths.
Patrick Mezard <pmezard@gmail.com> [Sun, 02 Dec 2007 14:06:10 +0100] rev 5582
Merge with crew-stable
Patrick Mezard <pmezard@gmail.com> [Sun, 02 Dec 2007 13:53:29 +0100] rev 5581
mq: missing target files do not make qpush to fail immediately (issue 835)
Reported and explained by Peter Arrenbrecht <peter.arrenbrecht@gmail.com>.
Following file additions were skipped but empty files were still created. This situation could lead to qrefresh losing patch information.
Dirkjan Ochtman <dirkjan@ochtman.nl> [Sun, 02 Dec 2007 10:12:26 +0100] rev 5580
Fix style nit and add some comments to tests.
Dirkjan Ochtman <dirkjan@ochtman.nl> [Sat, 01 Dec 2007 19:19:08 +0100] rev 5579
Use SCRIPT_NAME and PATH_INFO instead of REQUEST_URI. This is required by WSGI (fixes issue846).
Dirkjan Ochtman <dirkjan@ochtman.nl> [Sat, 01 Dec 2007 19:10:00 +0100] rev 5578
Add tests for hgwebdir repository names with slashes in them.
Dirkjan Ochtman <dirkjan@ochtman.nl> [Sat, 01 Dec 2007 18:26:27 +0100] rev 5577
CGI compatibility fix for d74fc8dec2b4.