Greg Ward <greg-hg@gerg.ca> [Thu, 25 Mar 2010 15:37:39 -0400] rev 10776
Fix default style so 'log --copies' has a start and an end.
Mads Kiilerich <mads@kiilerich.com> [Wed, 24 Mar 2010 01:43:24 +0100] rev 10775
Tests with spaces in paths
This allows most tests to succeed with
./run-tests.py --tmpdir='/tmp/hg test'
and introduces other tests for spaces and shell quoting
Matt Mackall <mpm@selenic.com> [Mon, 29 Mar 2010 12:56:16 -0500] rev 10774
tests: add helpers.sh
Sune Foldager <cryo@cyanite.org> [Fri, 26 Mar 2010 17:09:56 +0100] rev 10773
merge with stable
Sune Foldager <cryo@cyanite.org> [Fri, 26 Mar 2010 17:02:49 +0100] rev 10772
push: refactor and optimize prepush
Sune Foldager <cryo@cyanite.org> [Fri, 26 Mar 2010 17:02:36 +0100] rev 10771
push: fix bug in prepush logic and its tests
Sune Foldager <cryo@cyanite.org> [Fri, 26 Mar 2010 17:02:23 +0100] rev 10770
localrepo: change _updatebranchcache to use a context generator
Benoit Boissinot <benoit.boissinot@ens-lyon.org> [Wed, 24 Mar 2010 19:44:53 +0100] rev 10769
merge with stable
Benoit Boissinot <benoit.boissinot@ens-lyon.org> [Wed, 24 Mar 2010 19:44:30 +0100] rev 10768
setup.py: don't use tabs
Matt Mackall <mpm@selenic.com> [Tue, 23 Mar 2010 17:24:17 -0500] rev 10767
Merge with crew
Ronny Pfannschmidt <Ronny.Pfannschmidt@gmx.de> [Tue, 23 Mar 2010 15:23:04 +0100] rev 10766
run-tests: force to test pure on pypy as well
Ronny Pfannschmidt <Ronny.Pfannschmidt@gmx.de> [Tue, 23 Mar 2010 15:22:17 +0100] rev 10765
Fix run-tests.py -jX after
2ed667a9dfcb
Matt Mackall <mpm@selenic.com> [Tue, 23 Mar 2010 17:06:58 -0500] rev 10764
Merge with crew
Patrick Mezard <pmezard@gmail.com> [Tue, 23 Mar 2010 22:59:31 +0100] rev 10763
Merge with crew-stable
Patrick Mezard <pmezard@gmail.com> [Tue, 23 Mar 2010 20:45:08 +0100] rev 10762
rebase: fix --collapse with --keepbranches (
issue2100)
This was broken in
38fe86fb16e3.
Dirkjan Ochtman <dirkjan@ochtman.nl> [Tue, 23 Mar 2010 20:01:30 +0100] rev 10761
setup: fail if bz2 is not available
Pierre-Yves David <pierre-yves.david@ens-lyon.org> [Mon, 22 Mar 2010 22:19:40 +0100] rev 10760
test-convert-cvs: filter unstable cvs output.
The order choosen by cvs to display commit is not stable when the tests are run
under heavy CPU/IO load. This leads to irrelevant failures of the test suite.
This changeset filters four lines impacted by this instability. The result of
the conversion is still verified.
Dirkjan Ochtman <dirkjan@ochtman.nl> [Tue, 23 Mar 2010 11:48:14 +0100] rev 10759
help: point out need for stringification
Ronny Pfannschmidt <Ronny.Pfannschmidt@gmx.de> [Tue, 23 Mar 2010 11:37:31 +0100] rev 10758
tests: adapt the test runner to work with jython
Ronny Pfannschmidt <Ronny.Pfannschmidt@gmx.de> [Tue, 23 Mar 2010 11:37:01 +0100] rev 10757
posix: move a global fcntl import to keep it from breaking jython
Only needed on darwin, anyway.
Ronny Pfannschmidt <Ronny.Pfannschmidt@gmx.de> [Tue, 23 Mar 2010 11:36:19 +0100] rev 10756
util: fake the builtin buffer if it's missing (jython)
Patrick Mezard <pmezard@gmail.com> [Tue, 23 Mar 2010 10:18:46 +0100] rev 10755
mq: fix qcommit documentation wrt --mq option
Matt Mackall <mpm@selenic.com> [Mon, 22 Mar 2010 17:03:33 -0500] rev 10754
Merge with stable
Konstantin Zemlyak <zart@zartsoft.ru> [Mon, 22 Mar 2010 15:16:27 +0100] rev 10753
wsgicgi: call close() on iterable to avoid resource leaks
Quoting PEP 333 (WSGI):
"If the iterable returned by the application has a close() method, the server
or gateway must call that method upon completion of the current request,
whether the request was completed normally, or terminated early due to
an error. (This is to support resource release by the application.
This protocol is intended to complement PEP 325's generator support,
and other common iterables with close() methods."
Benoit Boissinot <benoit.boissinot@ens-lyon.org> [Mon, 22 Mar 2010 11:08:52 +0100] rev 10752
merge with mpm
Benoit Boissinot <benoit.boissinot@ens-lyon.org> [Mon, 22 Mar 2010 11:08:42 +0100] rev 10751
patch: warn when the internal patcher fails
hopefully we'll soon fix the remaining failures from the internal patcher, and
not use the external one anymore.
Brodie Rao <brodie@bitheap.org> [Mon, 22 Mar 2010 01:36:12 -0400] rev 10750
run-tests: make sure GREP_OPTIONS isn't set
This can interfere with test output, especially for options like --color.
Matt Mackall <mpm@selenic.com> [Sat, 20 Mar 2010 16:57:24 -0500] rev 10749
Merge with stable
Benoit Boissinot <benoit.boissinot@ens-lyon.org> [Sat, 20 Mar 2010 14:47:05 +0100] rev 10748
patch: more precise NoHunk, raised for every file (
issue2102)
Benoit Boissinot <benoit.boissinot@ens-lyon.org> [Sat, 20 Mar 2010 21:16:22 +0100] rev 10747
patch: differentiate start of file with diff --git vs '--- '
Matt Mackall <mpm@selenic.com> [Sat, 20 Mar 2010 10:01:56 -0500] rev 10746
Merge with stable
Benoit Boissinot <benoit.boissinot@ens-lyon.org> [Fri, 19 Mar 2010 22:52:38 +0100] rev 10745
patch: try harder to find the file to patch on file creation (
issue2041)
accept the following patch header:
first line: foo/a.orig
second line: foo/a
Matt Mackall <mpm@selenic.com> [Fri, 19 Mar 2010 16:05:16 -0500] rev 10744
Merge with stable
Matt Mackall <mpm@selenic.com> [Fri, 19 Mar 2010 16:04:49 -0500] rev 10743
Merge with i18n
Matt Mackall <mpm@selenic.com> [Fri, 19 Mar 2010 16:04:00 -0500] rev 10742
Merge with i18n
Wagner Bruna <wbruna@softwareexpress.com.br> [Tue, 16 Mar 2010 11:37:14 -0300] rev 10741
merge with i18n
Wagner Bruna <wbruna@softwareexpress.com.br> [Tue, 16 Mar 2010 11:36:44 -0300] rev 10740
merge with i18n stable
Wagner Bruna <wbruna@softwareexpress.com.br> [Tue, 16 Mar 2010 11:31:08 -0300] rev 10739
i18n-pt_BR: synchronized with
e930017f9e2a
Martin Geisler <mg@lazybytes.net> [Sat, 06 Mar 2010 10:02:45 +0100] rev 10738
Merge with stable
Martin Geisler <mg@lazybytes.net> [Sat, 06 Mar 2010 08:58:03 +0000] rev 10737
l10n: Updates to Danish (da) translation
Transmitted-via: Transifex (www.transifex.net)
Benoit Boissinot <benoit.boissinot@ens-lyon.org> [Fri, 19 Mar 2010 12:45:39 +0100] rev 10736
patch: second line of a context diff starts with '--- ', not '+++ '
Martin Geisler <mg@lazybytes.net> [Fri, 19 Mar 2010 11:32:55 +0100] rev 10735
Merge with stable
Cédric Duval <cedricduval@free.fr> [Mon, 15 Mar 2010 22:37:38 +0100] rev 10734
patchbomb: --desc implies --intro
For a single patch, an introductory text should not remain unsent
if --intro has been omitted, as the intent of --desc is clear.
Adrian Buehlmann <adrian@cadifra.com> [Fri, 19 Mar 2010 10:39:01 +0100] rev 10733
wix: don't show version in title of root feature
The version is already shown in the title bar of the feature selection dialog
and in the text of the welcome dialog. It's unusual for Windows programs to repeat
the version in the root feature.
Adrian Buehlmann <adrian@cadifra.com> [Fri, 19 Mar 2010 10:26:19 +0100] rev 10732
wix: use build variable for VC redist source dir
Brendan Cully <brendan@kublai.com> [Thu, 18 Mar 2010 14:36:24 -0700] rev 10731
Merge with stable
Benoit Boissinot <benoit.boissinot@ens-lyon.org> [Thu, 18 Mar 2010 19:26:56 +0100] rev 10730
mq: allow lines starting with '--- ' in patch messages
Benoit Boissinot <benoit.boissinot@ens-lyon.org> [Thu, 18 Mar 2010 18:22:34 +0100] rev 10729
patch: enhance diff detection regexp, allow '--- ' in patch message
Sune Foldager <cryo@cyanite.org> [Thu, 18 Mar 2010 16:38:07 +0100] rev 10728
fix bug in hg.clone when passing in a repository object
Benoit Boissinot <benoit.boissinot@ens-lyon.org> [Thu, 18 Mar 2010 16:32:03 +0100] rev 10727
check-code: more tests and more robust python filtering
Martin Geisler <mg@lazybytes.net> [Thu, 18 Mar 2010 01:59:53 +0100] rev 10726
Merge with stable
Martin Geisler <mg@lazybytes.net> [Thu, 18 Mar 2010 01:58:26 +0100] rev 10725
localrepo: more specific warning on adding large files
With better language as suggested by timeless.
Matt Mackall <mpm@selenic.com> [Wed, 17 Mar 2010 18:15:16 -0500] rev 10724
progress: drop extra args for pos=None calls (
issue2087)
Matt Mackall <mpm@selenic.com> [Wed, 17 Mar 2010 15:36:48 -0500] rev 10723
check-code: fix check-code complaint
Benoit Boissinot <benoit.boissinot@ens-lyon.org> [Wed, 17 Mar 2010 14:15:33 +0100] rev 10722
check-code: improve quote detection regexp, add tests
Martin Geisler <mg@lazybytes.net> [Wed, 17 Mar 2010 10:55:28 +0100] rev 10721
Merge with stable
Pierre-Yves David <pierre-yves.david@logilab.fr> [Wed, 17 Mar 2010 10:51:26 +0100] rev 10720
check-code: add a return value to checkfile function
The checkfile function returns True if the file is correct, False
otherwise.
Pierre-Yves David <pierre-yves.david@logilab.fr> [Tue, 16 Mar 2010 19:53:00 +0100] rev 10719
code-code: Add a logfunc argument to checkfile
This helps external tool customisation. The logfunc argument must be a
function called to report errors. A new logger class is also added to
keep the old behaviour by default.
Pierre-Yves David <pierre-yves.david@logilab.fr> [Tue, 16 Mar 2010 19:52:58 +0100] rev 10718
check-code: Add a ``maxerr`` argument to the ``checkfile`` function
check-code.py used to halt after 15 errors. This changeset adds a new
argument to the checkfile function to control this limit.
Pierre-Yves David <pierre-yves.david@logilab.fr> [Tue, 16 Mar 2010 19:52:57 +0100] rev 10717
check-code: Add a ``checkfile`` function
The part of the code actually checking each file is moved in the
``checkfile`` function to allow external reuses.
Pierre-Yves David <pierre-yves.david@logilab.fr> [Tue, 16 Mar 2010 19:52:56 +0100] rev 10716
check-code: Only call check-code if __name__ = "__main__".
This changeset moves the code that actually executes something to the
"if __name__ = '__main__'" section to allow the file to be imported as
a module.
Pierre-Yves David <pierre-yves.david@logilab.fr> [Tue, 16 Mar 2010 19:52:55 +0100] rev 10715
add a small test for contrib/check-code.py
This changeset adds a test file for contrib/check-code.py. This test runs
check-code.py against two small files. The first file should not raise any
check-code.py error while the second should. A line contains multiple errors to
verify the check-code.py behaviour: In the case of multiple errors at the same
line, the context line should only be displayed once.
Christian Ebert <blacktrash@gmx.net> [Tue, 16 Mar 2010 22:29:16 +0100] rev 10714
keyword: do not bother about detecting extension path in demo
Detecting and showing the path to a keyword extension in a
non-standard place only made sense while keyword.py was not
shipped with Mercurial.
The test output has changed because we do not have a spurious
space at eol anymore.
Christian Ebert <blacktrash@gmx.net> [Tue, 16 Mar 2010 22:28:58 +0100] rev 10713
keyword: make kwdemo less verbose
1) Set the branchname always silently with
dirstate.setbranch().
We create a branch so that testing the {branches} template
does not come up empty. But kwdemo is hardly the place to
inform the user by inference why {branches} is empty on the
default branch.
"demobranch" is ascii and cannot be changed, so using the
internal command instead of commands.branch() is safe.
2) Do not show full path to temporary directory
(distracting long lines on Mac OS X).
3) No special debug output. Output only related to keyword,
no internals like unsetting of commit hooks etc.
Christian Ebert <blacktrash@gmx.net> [Tue, 16 Mar 2010 22:28:23 +0100] rev 10712
keyword: 1 variable name for overwriting candidates
Christian Ebert <blacktrash@gmx.net> [Tue, 16 Mar 2010 22:27:22 +0100] rev 10711
keyword: do not customize ui.note() when overwriting
Behaviour change:
hg commit --verbose now prints out the names of the files
which are overwritten.
Cédric Duval <cedricduval@free.fr> [Mon, 15 Mar 2010 20:52:22 +0100] rev 10710
mq: test for init --mq with a destination argument
Adrian Buehlmann <adrian@cadifra.com> [Tue, 16 Mar 2010 17:39:12 +0100] rev 10709
contrib/wix: fix product name to include version
Without this, Mercurial is registered under the installed programs
on Windows as just 'Mercurial', which is unusual for a program to do
on Windows.
Including the version in the registered product name registers the
software in the Windows control panel of installed programs as, for
example, 'Mercurial 1.5' (for 1.5), or as
'Mercurial 1.5.1032' for a stable 1.5+32-
35893dcfd40c
'Mercurial 1.5.5080' for a unstable 1.5+80-
1ee60e82333c
when using http://bitbucket.org/tortoisehg/thg-winbuild to create
daily builds.
Furthermore, the install UI texts in the built installer database (MSI
file) are more clear on what the user is about to install, as, for example,
the text in the initial installer dialog now reads
"The Setup Wizard will install Mercurial 1.5 on your computer"
instead of just
"The Setup Wizard will install Mercurial on your computer"
Benoit Boissinot <benoit.boissinot@ens-lyon.org> [Tue, 16 Mar 2010 16:41:22 +0100] rev 10708
help: it's not necessary to escape quotes
Benoit Boissinot <benoit.boissinot@ens-lyon.org> [Tue, 16 Mar 2010 16:39:17 +0100] rev 10707
check-code: remove simple quotes first
Benoit Boissinot <benoit.boissinot@ens-lyon.org> [Tue, 16 Mar 2010 01:16:19 +0100] rev 10706
merge with stable
Benoit Boissinot <benoit.boissinot@ens-lyon.org> [Tue, 16 Mar 2010 01:16:04 +0100] rev 10705
filelog: no need to optimize an uncommon case, assume meta = {}
Benoit Boissinot <benoit.boissinot@ens-lyon.org> [Tue, 16 Mar 2010 01:12:46 +0100] rev 10704
filelog: text is stored modified when it starts with '\1\n'
Matt Mackall <mpm@selenic.com> [Mon, 15 Mar 2010 16:36:32 -0500] rev 10703
Merge with stable
Matt Mackall <mpm@selenic.com> [Fri, 12 Mar 2010 16:16:47 -0600] rev 10702
check-code: add check for any/all
Benoit Boissinot <benoit.boissinot@ens-lyon.org> [Mon, 15 Mar 2010 19:55:52 +0100] rev 10701
cvsps: fix traceback involving 'synthetic'
https://bugzilla.redhat.com/show_bug.cgi?id=573666
Reported by Sjoerd Mullender
Martin Geisler <mg@lazybytes.net> [Mon, 15 Mar 2010 18:40:00 +0100] rev 10700
progress: use a verb in present participle
Martin Geisler <mg@lazybytes.net> [Mon, 15 Mar 2010 18:36:11 +0100] rev 10699
Merge with stable
Martin Geisler <mg@lazybytes.net> [Mon, 15 Mar 2010 18:33:39 +0100] rev 10698
progress: use a verb (+noun) in present participle
Edouard Gomez <ed.gomez@free.fr> [Sun, 14 Mar 2010 00:28:10 +0100] rev 10697
subrepo: fix hgrc paths section during subrepo pulling
There are two bugs in current subrepo ctor:
- hgrc paths were using parent repo (ctx._repo) instead of self._repo
- it was joined with local extract path, not its source repo path
Fixed both bugs using _abssource on self once it is fully initialized.
It gets the job done perfectly for pull and push urls.
Patrick Mezard <pmezard@gmail.com> [Sun, 14 Mar 2010 23:18:50 +0100] rev 10696
Merge with crew-stable
Mathieu Clabaut <mathieu.clabaut@systerel.fr> [Tue, 09 Mar 2010 12:09:57 +0100] rev 10695
convert: teach cvsps to handle . repository (
issue1649)
For a CVS repository checked out with "cvs co .", the prefix used to strip of
what we get from CVS was previously erroneously set to "repopath/.".
We now prevent the dot to be added.
Test folded in test-convert-cvs and simplified by Patrick Mézard
<pmezard@gmail.com>.
Martin Geisler <mg@lazybytes.net> [Sun, 14 Mar 2010 23:10:52 +0100] rev 10694
record: separate each hunk with a blank line
This makes it easier for the eye to find the beginning of a hunk when
scrolling up in the terminal.
Benoit Boissinot <benoit.boissinot@ens-lyon.org> [Sun, 14 Mar 2010 22:17:10 +0100] rev 10693
mq: do not try to delete non-existent files (regression from
e8e56d8377ab)
Martin Geisler <mg@lazybytes.net> [Sun, 14 Mar 2010 22:21:48 +0100] rev 10692
Merge with stable
Cédric Duval <cedricduval@free.fr> [Sun, 14 Mar 2010 18:46:38 +0100] rev 10691
mq: fix init with nonexistent or non-local repository
Cédric Duval <cedricduval@free.fr> [Sun, 14 Mar 2010 19:21:24 +0100] rev 10690
mq: no longer refer to deprecated qinit -c command
Cédric Duval <cedricduval@free.fr> [Sun, 14 Mar 2010 19:17:06 +0100] rev 10689
mq: don't mention qinit as a "common task"
Without -c, it is a no-op (the queue is implicitly initialized by all
relevant commands if needed), and queue repositories with -c are an
advanced usage of mq, and not otherwise required.
Benoit Boissinot <benoit.boissinot@ens-lyon.org> [Sun, 14 Mar 2010 19:09:55 +0100] rev 10688
mq: don't use regexp when not necessary
Benoit Boissinot <benoit.boissinot@ens-lyon.org> [Sun, 14 Mar 2010 14:53:20 +0100] rev 10687
mq: use xrange/enumerate instead of += 1
Benoit Boissinot <benoit.boissinot@ens-lyon.org> [Sun, 14 Mar 2010 14:45:01 +0100] rev 10686
mq: don't use len(list) unless necessary
Benoit Boissinot <benoit.boissinot@ens-lyon.org> [Sun, 14 Mar 2010 12:08:14 +0100] rev 10685
mq: find_series() simplify and don't use regexps
Benoit Boissinot <benoit.boissinot@ens-lyon.org> [Sun, 14 Mar 2010 11:12:51 +0100] rev 10684
mq: fix coding style (missing space)
Benoit Boissinot <benoit.boissinot@ens-lyon.org> [Sun, 14 Mar 2010 11:12:07 +0100] rev 10683
mq: qsave creates entries with the left part empty (':patchname')
Benoit Boissinot <benoit.boissinot@ens-lyon.org> [Sun, 14 Mar 2010 01:35:54 +0100] rev 10682
mq: simplify statusentry(), fix restore broken by
ee48e5ef8753
Benoit Boissinot <benoit.boissinot@ens-lyon.org> [Sun, 14 Mar 2010 01:06:58 +0100] rev 10681
mq: use context API