enhance Makefile for language translators To update a po-file just do 'make i18n/xx.po'. No other po-file will be touched.

Fixing issue1542, adding a relevant test inotify is smart enough to notify you about any changes in a directory, even if you only watch the directory, and none if its contents: the recursive add_watch I added was unnecessary. The only thing that matters here is the recursive status update on directory deletion. And scan() has to be called _before_ the deferred call is registered. (race condition: depending on the times, the previous patch could apparently fail on the provided test. It's not the case anymore.)

rebase: remove unused statement

util: split out posix, windows, and win32 modules

patchbomb: option to set the name of bundle attachment (issue1452) specifying --bundlename=NAME will create a "NAME.hg" attachment

Merge with -stable

Typo fix in help.

i18n: add German translation * Most basic commands are translated * hgext/alias.py * hgext/acl.py * hgext/fetch.py * help topics * urls * dates * patterns * diffs 433 translated messages, 1350 untranslated messages. Translators: * Tobias Bell * Fabian Kreutz (fabian DOT kreutz AT qvantel.com) * Lutz Horn (lutz DOT horn AT fastmail DOT fm) Please report issues to http://bitbucket.org/tobidope/i18n-german-translation/issues/ or to the mailing list.

i18n: updated Chinese translation

i18n: update italian translation

revlog: faster hash computation when one of the parent node is null Because we often compute sha1(nullid), it's interesting to copy a precomputed hash of nullid instead of computing everytime the same hash. Similarly, when one of the parents is null, we can avoid a < comparison (sort). Overall, this change adds a string equality comparison on each hash() call, but when p2 is null, we drop one string < comparison, and copy a hash instead of computing it. Since it is common to have revisions with only one parent, this change makes hash() 25% faster when cloning a big repository.

ancestor: caching the parent list to improve performance When computing the DAG depth, we walk through all ancestors: this commit adds memoization during that first step. Then, the memorized parents are fetched from a dict instead of calling parents() on each vertex. This tweak, according to Kcachegrind, improves ancestor() performance by 16% when cloning a big repository.

kill another trailing space

add debugcommands command: an easy to parse command + option index

templater: ability to display diffstat for log-like commands

templater: use contexts consistently throughout changeset_templater

cleanup: whitespace cleanup

cleanup: drop enumerate() when index is not used

cleanup: drop unused assignments

cleanup: drop variables for unused return values They are unnecessary. I did leave them in localrepo.py where there is something like: _junk = foo() _junk = None to free memory early. I don't know if just `foo()` will free the return value as early.

cleanup: drop unused imports

contrib: add perflog and perftemplating commands to perf extension

merge with crew-stable

Returns lines changed for paths specified as arguments correctly. This fixes issue 1569. hg churn <path> now returns only the number of lines changed in the path, if the path is specified by filtering files through a match filter at the changeset level. test-churn has been updated to take care of this issue.

cleanup: remove all trailing whitespace

keepalive: fix reference to IncompleteRead

merge tools: fix typo in Beyond Compare's 3 command line Fix a typo error preventing 'base' to be displayed properly.

Merge with -stable

Fix new CVS test output

Added signature for changeset 02981000012e

Added tag 1.2.1 for changeset 02981000012e

cvsps: recognize and eliminate CVS' synthetic "file added" revisions.

demandimport: blacklist pythoncom win32com.shell would segfault at import time if pythoncom was demand loaded.

diffstat: use width 80 by default and avoid division by zero

Fix for merge changes introduced in rev 9fe7e6acf525

merge with crew-stable

convert: honor 2.3 compatibility (rsplit)

Merge with stable

regression test for issue1552 Ensures that fetch correctly infers what to merge when pulling a repository with inactive branches.

fetch: do not count inactive branches when inferring a merge The fetch extension would erroneously consider inactive branches when inferring which branches to merge. It now considers only active branches.

bundlerepo: reintroduce dirstate

exporting patch: Fixed behavior of revsplit for branch names including the @-symbol.

imported patch issue1546-hg-bisect-command-doesn-t-work-with-shell-scripts.patch

bisect: fix --command for me.

atom: fix URLs used in entry GUIDs

resolve: move reset to localrepo.commit This way rebase doesn't leave a stale resolve state

resolve: keep .orig files

test-merge-tools: test ui.merge functionality and selection of internal tools