Matt Mackall <mpm@selenic.com> [Sat, 24 Nov 2007 12:35:02 -0600] rev 5546
Merge with crew
Patrick Mezard <pmezard@gmail.com> [Sat, 24 Nov 2007 17:56:15 +0100] rev 5545
cmdutil: always expose "files_add", "files_del" and "manifest" templater properties
Matt Mackall <mpm@selenic.com> [Sat, 24 Nov 2007 12:13:32 -0600] rev 5544
revlog: make revlogv0 loading more robust against corruption
Matt Mackall <mpm@selenic.com> [Wed, 21 Nov 2007 13:26:18 -0600] rev 5543
push/pull: abort if we try to visit a missing or empty revlog
Matt Mackall <mpm@selenic.com> [Tue, 20 Nov 2007 15:54:25 -0600] rev 5542
i18n: fix complaints from pygettext
Matt Mackall <mpm@selenic.com> [Tue, 20 Nov 2007 15:46:20 -0600] rev 5541
verify: improve handling of empty or missing files
- explicitly notice empty or missing changeset, manifest or file
revlogs
- shortcircuit warnings relating to missing changeset or manifest
revlogs
- fix changeset/manifest crosscheck swapped variables
- properly sort crosscheck messages
Patrick Mezard <pmezard@gmail.com> [Tue, 20 Nov 2007 14:41:47 +0100] rev 5540
Merge with crew-stable
Patrick Mezard <pmezard@gmail.com> [Sun, 18 Nov 2007 17:25:28 +0100] rev 5539
convert: read CVS files in chunks (issue 800)
socket.makefile() fails on large read requests (more than 10MB) with MemoryError.
Patrick Mezard <pmezard@gmail.com> [Sat, 17 Nov 2007 18:41:31 +0100] rev 5538
test-convert-svn-sink: test directory additions
Patrick Mezard <pmezard@gmail.com> [Sat, 17 Nov 2007 18:41:31 +0100] rev 5537
convert: add directories to subversion from root to leaves
Patrick Mezard <pmezard@gmail.com> [Sun, 11 Nov 2007 19:34:00 +0100] rev 5536
convert: force svn:executable when execute-bit is not supported
The util.is_exec fallback would be to ask subversion for the current value of svn:executable. Setting it directly is just simpler.
Patrick Mezard <pmezard@gmail.com> [Sun, 11 Nov 2007 14:17:30 +0100] rev 5535
convert: fix svn file:// URL generation under Windows
Patrick Mezard <pmezard@gmail.com> [Tue, 20 Nov 2007 14:31:58 +0100] rev 5534
mq: avoid qnew committing everything on invalid paths (issue 803)
If paths are supplied but resolve to nothing, localrepo.commit() is called with an empty set and commits the whole dirstate. Avoid this by passing the match function to commit.
Bryan O'Sullivan <bos@serpentine.com> [Mon, 12 Nov 2007 10:49:26 -0800] rev 5533
highlight: clean up coding style a little
Adam Hupp <adam@hupp.org> [Sat, 10 Nov 2007 17:54:57 -0500] rev 5532
extension for synax highlighting in the hgweb file revision view
Depends on the pygments syntax highlighting library:
http://pygments.org/
Patrick Mezard <pmezard@gmail.com> [Sat, 10 Nov 2007 21:30:59 +0100] rev 5531
Change standard streams mode to binary at hg startup
Standard streams are expected to operate in binary mode everywhere, especially with archive, cat, diff and export commands. Rewriting these to separate informational output from binary content is complicated to do and to maintain, nonwithstanding mode switching reliability. Changing all output mode to binary should not have much impact on Windows were stream processing tools are barely used and usually cope with unix style endings.
Streams mode being process wide, the switch is performed in the startup script to avoid polluting existing API users who may have solved this issue already or ignored it at least for the mercurial part.
Patrick Mezard <pmezard@gmail.com> [Sat, 10 Nov 2007 21:30:43 +0100] rev 5530
test-import: read email payload in binary mode
Patrick Mezard <pmezard@gmail.com> [Sat, 10 Nov 2007 17:09:56 +0100] rev 5529
convert: fix util.popen regression in darcs converter
Alexis S. L. Carvalho <alexis@cecm.usp.br> [Fri, 09 Nov 2007 21:24:25 -0200] rev 5528
merge with crew-stable
Alexis S. L. Carvalho <alexis@cecm.usp.br> [Fri, 09 Nov 2007 20:21:35 -0200] rev 5527
mq: really remove undo after a qpush (and after a strip)
For qpush, removeundo has to be called after the transaction has finished.
strip may use unbundle, which also leaves an undo file.
Fixes
issue780.
Alexis S. L. Carvalho <alexis@cecm.usp.br> [Fri, 09 Nov 2007 20:21:35 -0200] rev 5526
httprepo: give self._url and the netloc to the password manager
This should hide some differences between different python versions.
Manuel Holtgrewe <purestorm@ggnore.net> [Mon, 05 Nov 2007 20:29:32 +0100] rev 5525
Do not display passwords with pull/push/incoming/outgoing
Passwords specified in the repository URL are now displayed as '***'
when accessing the remote repository.
Alexis S. L. Carvalho <alexis@cecm.usp.br> [Fri, 09 Nov 2007 20:21:35 -0200] rev 5524
run-tests.py: add a default --date "0 0" argument to commit et al
Part of test-tags was modified just to be sure this works.
The change in test-archive-symlinks is necessary to avoid a "helpful"
warning from GNU tar ("implausibly old time stamp 1970-01-01 00:00:00").
Alexis S. L. Carvalho <alexis@cecm.usp.br> [Fri, 09 Nov 2007 20:21:35 -0200] rev 5523
tests/*: avoid losing the original settings from $HGRCPATH
Alexis S. L. Carvalho <alexis@cecm.usp.br> [Fri, 09 Nov 2007 20:21:35 -0200] rev 5522
convert: make sure mercurial_source has a local hg repo
Alexis S. L. Carvalho <alexis@cecm.usp.br> [Fri, 09 Nov 2007 20:21:35 -0200] rev 5521
convert: display all errors if we couldn't open the source repo
This should give the user a better hint of what's going wrong.
Improve some error messages. In particular, mention "CVS checkout" instead
of "CVS repo".
Fixes
issue822 and
issue826.
Alexis S. L. Carvalho <alexis@cecm.usp.br> [Fri, 09 Nov 2007 20:21:35 -0200] rev 5520
test-convert-darcs: skip if we can't find the elementtree module
This patch is a bit dirty to avoid having to repeat the dance required
to import elementtree.
Alexis S. L. Carvalho <alexis@cecm.usp.br> [Fri, 09 Nov 2007 20:21:35 -0200] rev 5519
test-permissions: add a chmod +w to allow the directory to be removed
Alexis S. L. Carvalho <alexis@cecm.usp.br> [Fri, 09 Nov 2007 20:21:35 -0200] rev 5518
run-tests.py: don't print a dot for skipped tests
Alexis S. L. Carvalho <alexis@cecm.usp.br> [Fri, 09 Nov 2007 20:21:35 -0200] rev 5517
purge: simplify safety net for case mangling filesystems
Relying on the exact return of statwalk would cause us to abort
when there was at least one tracked file inside an ignored directory.
This patch forces an extra walk of the whole working directory even
on sane filesystems, where it wouldn't be needed.
Fixes
issue621.