http server: support persistent connections. only "hg serve" affected yet. http server running cgi script will not use persistent connections. support for fastcgi will help that. clients that support keepalive can use one tcp connection for all commands during clone and pull. this makes latency of binary search during pull much lower over wan. if server does not know content-length, it will force connection to close at end. right fix is to use chunked transfer-encoding but this is easier and does not hurt performance. only command that is affected is "changegroup" which is always last command during a pull.

merge with crew.

move hgk.py into hgext. now to enable "hg view" is one less step. edit ~/.hgrc: [extensions] hgext.hgk = cp contrib/hgk ~/bin

replace old-http:// syntax by static-http:// and deprecate the redundant hg://

Add web.contact to the hgrc documentation.

dirstate: fix call to os.lstat when st is None

honor signals during load of extensions.

dirstate.read: make 15% faster. time to read dirstate for my kernel repo goes from 0.22s to 0.18s.

merge with crew.

dirstate: speed up inner loop of read.

Report on push/pull if heads are merged, too, like for new heads. Use case: If a remote repo has two heads and I _want_ to merge them, I merge and push. Meanwhile someone else pushed on top of one of the heads. He won't get a warning, because he doesn't create a new head, I won't notice that I don't close a head, because I don't get a message telling me.

merge with upstream

add --lsprof option. 3x faster than --profile, more useful output. results include calls to c code and nested calls. requires python 2.5 or lsprof installed from svn at http://codespeak.net/svn/user/arigo/hack/misc/lsprof/

ssh: add capability detection at startup Because older servers don't return any output for unknown commands, it's tricky to add new commands. The approach is this: we add a "hello" command that reports any interesting capabilities (and other things that might be of interest in the future). To detect whether this new command is supported, we issue both it and our startup detection command ("between") at the beginning of a connection.

ssh: gather initial output so we can do capability detection

Initial implementation of hello command for ssh

update test-help output after new options.

copy: fix introduced bug.

make description of -n option better.

revert: add -n/--dry-run option

remove: add -n/--dry-run option

copy: add -n/--dry-run option

addremove: add -n/--dry-run option

add: add -n/--dry-run option

debugconfig: allow to print by section and exact item name

Fix diff header (line endings) for failed test output in run-tests.py.

Don't run tests with dots in their name (e.g. test-foo.orig)

revert: better fix for not printing 'reverting' message

revert: check for changes against target rev

pick a non-conflicting module name when loading an extension prefix the module name by 'hgext_' so that it doesn't conflict with any core python module (for example bisect -> hgext_bisect) fix issue271

Fixed 'hg serve --webdir-conf foo' which broke due to split of hgweb.

Further cleanup of ui.py (changeset 74d569332f8b used one-char variable names).

packagescan can't scan nested packages

move purge extension out of hgext. it is not as clean as other modules in hgext. can move back after rewrite.

import: allow to import stdin with file name "-"

fix comment.

merge with crew.

merge change to ssh protocol.

refactor ssh server.

remove appendfile for the manifest when adding a changegroup Since the changelog is using appendfile, the manifest entries cannot be referenced, so we don't need to use append file for the manifest.

hgweb: fix errors and warnings found by pychecker - fix missing import - use type_ instead of type - remove unused variable

dirstate: refactor the dirstate binary format, remove magic numbers

Adjusting hgweb splitup to be a little cleaner.

Fixing up comment headers for split up code.

merge with crew.

merge with crew.

Cleanup: unifiy the coding style in the ui.py configitems forwarders. No functional changes.

Expand '~' in path to extensions.

merge with crew.

The line used to ignore the setup file for win32 is a regexp but it was after a "syntax: glob"

Automatically add the path of hg to %PATH% using add_path

Ignore the .exe setup file for Windows

Make indentation of purge's cmdtable match to mercurial/commands.py

Rewritten install instructions for hg-purge to match new situation, fixed typos.

Merged hg-purge

Reorganized files in hg-purge repo to push them into Mercurial. README -> hgext/purge/README purge.py -> hgext/purge/__init__.py removed COPYING.GPL and .hgignore (already in the Mercurial repo)

Removed --nothing, added --print and --print0

The meaning of the directories on the command line is now explained correctly

Marked strings for translation

Added some comments

Even on Windows self._repo.dirstate.state() wants '/' in paths

Some functions were called as globals (i.e. I forgot some "self.")

Changset de893ad6bd17 wrongly reversed the meaning of --nothing

Use self._ui.note(...) instead of if self._ui.verbose: self._ui.status(...)

Command line options are read in a saner way

Purge.from_command is now a function called purge

Added a link to the home page of hg purge

Added a link to the page for CVS utilities

Added some lines on the requirements for purge

Added an explanation on how to enable the extension in "~/.hgrc"

Initial commit

Give a response to unknown SSH commands

Renamed localrepo.undo() to rollback() and talk about "rollback information".

Fix automatic decompression of tarballs with Firefox. The encoding was set to gzip/bzip2, so Firefox (correctly) assumed, that this has to be "decoded" (i.e. decompressed).

hgweb: fix tracebacks on both index and repo pages

Allow comma to separate types in allow_archive, too. Use longer variable name.

hgweb: add allow_archive support to [web] section of hgrc

merge with crew.

Final stage of the hgweb split up. hgweb and hgwebdir now have their own modules.

Splitting up hgweb so it's easier to change.

Manifest groups may be empty, so don't abort in this case (fixes issue210). Only abort on empty changelog and file revlogs.

Pass filename to decoder for 'hg archive' (fixes issue267) Despite writing to a file descriptor, the check in the [decode] section needs a filename. localrepo.wwrite() still writes to the file descriptor.

various fixes to darcs conversion script 1. removed files were missed 2. hidden files in top directory were not added/removed correctly 3. suppressed spurious warning on adding new files. 4. hg changeset now checked in with original author and date 5. __file__ not supported by pdb.py

contrib: move hg-menu.vim into contrib/vim directory

contrib: patch review plugin for vim 7.0 The plugin takes an 'hg export'ed patch (in fact any single or multi file patch) and opens multiple tabs containing vim diff/merge windows for each affected file in the patch allowing full visual code reviews.

darcs2hg.py: import darcs project into mercurial For now, the tags are not taken into account (so you have to reenter the tags by hand)

better ui for the bisect extension (and update to i18n)

use a more reliable way to find what are the new changesets on pull/unbundle The number of csets and the hooks where wrong (negative number of csets) when we unbundled a bundle which contains csets we already had. Remove unused variables.

merge with crew

n is always 'True', we can only stop the loop with the break statement

add acl extension, to limit who can push to subdirs of central repo.

add ui.has_config method.

merge with crew.

contrib: add restricted shell.

merge with crew

document and fix findincoming - add documentation about what the function does, notably the fact that it updates 'base' - transform the workflow to a more simple 'if elif elif else' - do not call remote.branches if not necessary - some nodes where missing in 'base' (from what I understand, if the root of a branch is missing but one parent is present, the parent should be in 'base') - add a testcase for an incorrect outgoing that is fixed by this cset - add a testcase for an empty group bug, it needs fixing

Minor typo change to reflect actual values used.

http: fix many problems with url parsing and auth. added proxy test. problems fixed: - https scheme handled properly for real and proxy urls. - url of form "http://user:password@host:port/path" now ok. - no-proxy check uses proper host names.

http: print better error if exception happens.

add ui.print_exc(), make all traceback printing central.

merge with crew.

new extension: extdiff. allows to use external diff program.

merge with crew.

diff: allow to use -rA:B syntax as well as -rA -rB

add 'uisetup' function to extension module protocol. if uisetup functin exists in extension, is called before cmdtable examined. called with ui object as parameter. lets module modify cmdtable before commands.py sees it.

notify: add debug output. do not fail if no config file. use --debug to see debug output.

hgweb.manifest: revno of manifest and changelog aren't always the same In the v4l-dvb repo, the manifest revno and the changelog revno are not in sync. This happened because the same patch was applied to the same revision in two different branches, resulting in the same manifest text, with the same parents and so the first revision was reused. Since hgweb.manifest was assuming the revnos of the manifest and of the changelog were always the same, clicking on manifest -> bz2 in the v4l-dvb site would download the wrong revision. Use the linkrev to go from manifest revision to changelog revision. This still won't be perfect since the page will still talk about "manifest for changeset XYZ", where XYZ was the first changeset to have this manifest, which is not necessarily the same changeset that the user clicked to get to this page - but at least the contents will be the same.

merge with crew.

notify: fix off by one error.

merge with crew.

merge with crew.

setup.py: install packagescan before any mercurial modules is imported Further the installation of packagescan over demandload is moved to the packagescan module. I added as well few more comments in the packagescan module to avoid the wrong use of package scan in the future. Reason: mercurial.packagescan acts as fake mercurial.demandload during a py2exe run. Unfortunatly the import of mercurial.version in setup.py is done before mercurial.packagescan is installed. This results in few imports without mercurial.packagescan in charge and therefore not all dependend modules are detected when running mercurial.packagescan.getmodules later e.g. winerror is missed.

packagescan: handle demandload module naming changes.

Update configuration note with mercurial.ini change from Thomas

fix parsing of tags. make parse errors useful. add new tag tests. old code read every head of .hgtags. delete and recreate of .hgtags gave new head, but if error in deleted rev, .hgtags had error messages every time it was parsed. this was very hard to fix, because deleted revs hard to get back and update, needed merges too. new code reads .hgtags on every head. advantage is if parse error happens with new code, is possible to fix them by editing .hgtags on a head and committing. NOTE: new code uses binary search of manifest of each head to be fast, but still much slower than old code. best thing would be to have delete record stored in filelog so we never touch manifest. could find live heads directly from filelog. this is more work than i want now. new tests check for parse of tags on different heads, and inaccessible heads created by delete and recreate of .hgtags.

test-globalopts: make sed expression better.

If HOME is defined on Windows, Mercurial uses it. Document this.

merge emacs patches.

Emacs: separate limitations of revision range and shown changesets. For revision range limitation, use "-r" opt with range notation at "hg log" invocation, and use hg-rev-completion-limit as default value. For shown changesets limitation, use "-l" opt at "hg log" invocation, and use hg-log-limit as default value.

Emacs: adapt read-file-name invocation for (non-X)Emacs 21.4. this patch deters hg-read-file-name from passing hg-file-history to read-file-name as 6th parameter in non-X Emacs environment.