convert-cvs: correctly handle Removed replies, fixes bug 1427

Fix test-acl failure on AIX due to read only LOGNAME. On AIX /etc/profile sets LOGNAME read only. This causes test-acl to fail when it comes to set LOGNAME in do_push(). Work around this by using env to set LOGNAME and run the command.

Merge with crew

Update zsh completion to use debugcomplete -v instead of parsing help

Make debugcomplete -v print aliases as well as commands

Merge with crew-stable

patch: catch only IOError from makedirs()

test-mq-missing: add test for directory creation on git copy

When applying a git diff, ensure that the target dir exists for new files

Revise Bugzilla module comments into extension help. Revise the comments detailing module usage at the start of the Bugzilla extension into extension help text. Expand the detail given on extension settings, in an attempt to permit the user to configure successfully without having to study the extension code. Show a sample configuration as well.

test-convert-cvs-builtincvsps: test debugcvsps command

Add debugcvsps command, replacing cvsps script

merge stable changes, via crew-stable

merge changes from stable

hgrc 5: fix email.charsets documentation

notify: no charset conversion when testing Same as for patchbomb. Make "test" a notifier attribute so we don't have to check the boolean over and over.

Merge with crew-stable

issue1419: catch strange readline import error on windows

Merge with crew-stable

util: disable walkrepo() recursive behaviour Revert b1aea76f700 and 268c582bd42a for performances reasons. Traversing checkout working directories may be too expensive.

Fix Bugzilla integration to work with new Bugzilla 3.2. Bugzilla 3.2 changes the type of all MySQL tables it uses from MyISAM to InnoDB. MyISAM does not support transactions, and performs an implied commit after each update. InnoDB does support transactions, and so exposes a bug where changes to the Bugzilla database were not committed, and so with 3.2 are lost when the database connection closes.

tests: test-walkrepo shouldn't throw SystemExit Write to stdout instead of throwing Exceptions

bookmarks: this is a comment, not a string

Merge bookmarks

Merge backout

Backed out changeset 4c3e0ad58c5b In discussion with mpm and tonfa we decided that we need a more generic solution and not a bookmark specific.

Merge backout

Backed out changeset 187a13bd14c6 In discussion with mpm and tonfa we decided that we need a better rev syntax instead of a bookmark specific solution.

Merge bookmarks

bookmarks: do not overwrite bookmarks.current if not necessary If we update to the ref that the bookmarks.current points to, we do not want to update bookmarks.current

bookmarks: add --only-bookmark option to log

bookmarks: Fix tests As we merge tags now with bookmarks we have add those tags to the tests.

bookmarks: more git-like branches Bookmarks are not really git branches, but with the track.current option we try to focus more on a git like behavior as this was requested.

bookmarks: Include bookmarks in tags.

bookmarks: Fix indention

bookmarks: Adds support for a --non-bookmarked option to push

Merge with crew-stable

convert: backout a7492fb2107b This change is brain damaged, there is no reason the copyfrom revision of the project items may have any relevance when deciding the revision parent. It is meaningful only when fetching files content. Incorrect converted graph was spotted in pyglet svn repository at: ------------------------------------------------------------------------ r274 | r1chardj0n3s | 2006-12-21 02:02:14 +0100 (Jeu, 21 Dec 2006) | 2 lines Changed paths: A /branches/richard-glx-version (from /trunk:269) M /branches/richard-glx-version/pyglet/window/xlib/__init__.py R /branches/richard-glx-version/tests/test.py (from /trunk/tests/test.py:270) R /branches/richard-glx-version/tools/info.py (from /trunk/tools/info.py:272) R /branches/richard-glx-version/website/get_involved.php (from /trunk/website/get_involved.php:273) Branching to horribly mangle GLX

Rewrite svn tests using svndump They are: - faster to run (12s vs 30s in test-convert-svn-branches case) - easier to craft - easier to debug - more portable

use inst.args instead of using __getitem__ directly DeprecationWarning: __getitem__ not supported for exception classes in 3.x; use args attribute

exceptions should inherit the Exception class

convert/cvs: do not compare None with int

use repr() instead of backticks

use repr() instead of backticks

Merge with crew-stable

rebase: add support to keep branch names Callback on extra fields added by Patrick Mezard <pmezard@gmail.com>.

Merge with crew

Update zsh completion for new help format

tests: allow cvs import to reorder its filename list Also fix script to generate non-zero length test file b.

re-add +x bit for test-convert-darcs (was missing from 14ed6662af48)

Added signature for changeset 2a67430f92f1

Added tag 1.1 for changeset 2a67430f92f1

encoding: normalize some silly encoding names

push: use the fast changegroup() path on push The race doesn't happen on push (because the discovery is done in the same hg process), so use the fast path instead.

color: replace effect-specific reset control codes with general purpose one

color: Add tests for colorized diff and status output.

diff colorization: finish highlighting trailing whitespace

color: diff colorization This colorizes diff, qdiff, log -p, outgoing -p, incoming -p, and tip -p.

color: add --color switch --color works like GNU grep --color. --color=auto only colorizes output for non-dumb terminals that are TTYs.

mq: filter out empty commit messages in qfold

mq: drop copy sources removed during qfold (see issue1405)

inotify: do not complain that inotify is not running if autostart is False

inotify: close most file descriptors when autostarting Otherwise, operations that autostart while talking to an SSH repository prevent SSH stderr from closing normally. This causes hangs at the end of hg clone or hg pull -u.

hgweb: extend [paths] syntax to match repositories recursively (issue852) This feature somehow duplicates [collections] but it is simpler to use and has less issues under Windows where using absolute path as configuration file key is not supported. Suggested by Dirkjan Ochtman <dirkjan@ochtman.nl>

help: try topics before commands to avoid "diffs" showing help for diffstat

Fix non-empty $CDPATH causing failed tests. If CDPATH contains '.', every cd command prints the target path, which causes additional output in tests and makes them fail.

convert: improve reporting of invalid svn bindings

convert: work around svn.ra.get_files() not releasing input buffer

gitweb: fix graph again after previous faulty commit (1e0677756f60)

Improvement to 14ce129cfcd: Use try/except and pass filename on errors Without the second part, the error message would be abort: Is a directory instead of abort: Is a directory: /home/user/.cvspass

gitweb: fix graph display in IE

convert: check existence of ~/.cvspass before reading it

convert cvs: Fix branch name parsing cvsps version 2.2b1 as found in Fedora 10 outputs the following format: --------------------- PatchSet 1 Date: 2008/11/26 00:59:46 Author: mk Branch: HEAD Tag: (none) Branches: INITIAL Log: Initial revision Members: a:INITIAL->1.1 b/c:INITIAL->1.1 --------------------- The parser overwrote the Branch value with noise from the misparsed Branches value.

Fix test breakage introduced in 07faba78cf5a

Ignore eclipse droppings

Add hgcia to hgext

tests: fix test results for 07faba78cf5a

diff: fix obscure off-by-one error in diff -p

bundlerepo doesn't really have a dirstate, throw AttributeError if requested

kill some trailing spaces

merge with mpm

rpm spec: include python egg in rpm Better complicance with packaging guidelines http://fedoraproject.org/wiki/Packaging/Python http://fedoraproject.org/wiki/Packaging/Python/Eggs

buildrpm: complain when hg command isn't available Before it complained "You are not inside a Mercurial repository!" which wasn't true.

util: use existing never() instead of custom lambda

tests: Skip tests if they will fail because of outer repo For different reasons these tests will fail if run in a tmpdir which is in a hg repo. The following three tests assumes no .hg in path dirs - I don't know how to work around that: * test-dispatch explicitly tests for no repo and expects "abort: There is no Mercurial repository here (.hg not found)!" * test-extension expects parentui to be None when not cd'ed to a repo dir * test-globalopts tests that implicit -R works correctly - that could perhaps be done from another repo instead of assuming no repo The following two might be worth investigating further: * test-convert-svn-sink fails for unknown reasons, starting with "abort: unresolved merge conflicts (see hg resolve)" * test-glog gets strange failures when testing "from outer space"

tests: use killdaemons in hgweb tests hgweb tests often failed on my system because the serve port wasn't free when a new hgweb was started; the killed hg wasn't completely dead yet. Now we use killdaemons which waits for the process to die.

util_win32: fix Windows version checking (issue1358) shell.SHGetPathFromIDList() segfaults when called in Apache.

Graphlog extension adds a --graph option to log/in/out The --graph option shows the ascii revision graph when used in conjunction with the incoming, outgoing or log commands. It also makes sure that incompatible options (e.g. --newest-first) are not used.

tests: disable test-convert-darcs properly for darcs 2 The x bit was disabled without notice in 8046f0a070a6

Fix rpmlint warnings contrib/mercurial.spec:9: W: hardcoded-path-in-buildroot-tag /tmp/build.%{name}-%{version}-%{release} A path is hardcoded in your Buildroot tag. It should be replaced by something like %{_tmppath}/%name-root. contrib/mercurial.spec:31: W: rpm-buildroot-usage %prep rm -rf $RPM_BUILD_ROOT $RPM_BUILD_ROOT should not be touched during %build or %prep stage, as it will break short circuiting.

Make rpm spec less explicit Fixes: * current buildrpm failing on Fedora because of unexpected zeroconf extensions * unowned directories hgext/*/

Merge with crew

Catch both IOError and OSError, fix regression introduced by 8046f0a070a6

inotify: remove unused imports (thanks pyflakes)

color: don't run status twice for -n Just cut off the status characters.

color: fix status -n

keyword: do not shadow builtin format (detected by pychecker)

hooks: restore io correctly on exception

protocol: use changegroupsubset() if possible (issue1389) Due to the fix to the pull race, to avoid sending unnecessary changesets, use changegroupsubset if possible. This will increase the load on the server.

Merge with crew

email: add References field in the header This hopefully helps clients get the threading right.

coal/paper: link to shortlog instead of log from error pages

coal/paper: move actual templates from coal to paper

coal/paper: show branch name in changeset view

coal/paper: display tags and branch in filelog page

coal/paper/monoblue: show branch names in hgweb graph

coal/paper: display branch names in log

hgweb: support broken IIS 5 behavior with .cgi in PATH_INFO

manifest: fix _search() corner-case It failed when searching the empty string in the null revision.

Improved error message for log --follow Mads Kiilerich notes that the existing message was unclear when a nonexistent file was supplied.

Improve the monoblue theme for hgwebdir

import: add similarity option (issue295)

mq: recover more gracefully from interrupted qrefresh (issue1216)

mq: use atomictempfiles during patch refresh

mq: create patch header class to abstract header manipulation

mq: gracefully abort qpush/qgoto to guarded patch (issue1186)

More verbose logging when filemerge searches for merge-tool Previously it was very hard to find out what was going on when the expected merge tool wasn't used. This patch tries to improve that. hg merge -v now shows which tools were searched for but not found.

templater: return data in increasing chunk sizes Currently hgweb is not streaming its output -- it accumulates the entire response before sending it. This patch restores streaming behaviour. To avoid having to synchronously write many tiny fragments, this patch also adds buffering to the template generator. Local testing of a fetch of a 100,000 line file with wget produces a slight slowdown overall (up from 6.5 seconds to 7.2 seconds), but instead of waiting 6 seconds for headers to arrive, output begins immediately.

tests: fix tests broken by 03b60f2f90bf

hgweb (coal/paper): make file, diff and annotate easier to render This patch removes tables altogether for file and diff, and removes one column from annotate, making firefox's incremental renderer much faster for large files.