pushkey: add debugpushkey command for testing

pushkey: add http support pushkey requires the same permissions as push listitems requires the same permissions as pull

pushkey: add ssh support

pushkey: add localrepo support

pushkey: add pushkey core

sshrepo: sort arguments All current commands have zero or one args. Future multi-arg commands will want args to appear in a deterministic order.

mq: make 'qdelete <patchidx>' work again. This just backs out 1abd9442727c (a minor code cleanup that accidentally broke qdelete) and adds a test.

bdiff.c: Added support for py3k. This patch adds support for py3k in bdiff.c. This is accomplished by including a header file responsible for abstracting the API differences between python 2 and python 3.

diffhelpers.c: Added support for py3k. This patch adds support for py3k in diffhelpers.c. This is accomplished by including a header file responsible for abstracting the API differences between python 2 and python 3.

base85.c: Added support for py3k. This patch adds support for py3k in base85.c. This is accomplished by including a header file responsible for abstracting the API differences between python 2 and python 3.

parsers.c: Added support for py3k. This patch adds support for py3k in parsers.c. This is accomplished by including a header file responsible for abstracting the API differences between python 2 and python 3.

mpatch.c: Added preliminary support for py3k. This is done by including the util.h header file, that defines appropriate macros according to the current python version.

osutil.c: Support for py3k added. This patch adds support for py3k in osutil.c. This is accomplished by including a header file responsible for abstracting the API differences between python 2 and python 3. listdir_stat_type is also changed in the following way: A previous call to PyObject_HEAD_INIT is substituted to a call to PyVarObject_HEAD_INIT, which makes the object buildable in both python 2.x and 3.x without weird warnings. After testing on windows, some modifications were also made in the posixfile function, as it calls PyFile_FromFile and PyFile_SetBufSize, which are gone in py3k. In py3k the PyFile_* API is, actually a wrapper over the io module, and code has been adapted accordingly to fit py3k.

util.h: Utility macros for handling different Python APIs. If we are in py3k, a IS_PY3K symbol is defined. Apart from that, byte strings use the API defined in Python 2.6+ (_?PyBytes_.*). For Python < 2.6, the bytes API is defined accordingly for mercurial usage (shameless copy from bytesobject.h from Python's code). Some macros were backported from 2.6, as inspired by rPath's pycompat.h.

relink: ensure deterministic directory walk in collect

help: add "glossary" topic Joint work with Vishakh Harikumar, Pradeepkumar Gayam, David Champion, Mark Booth, timeless, Matt Mackall, Heinrik Stuart, Greg Ward, and Martin Geisler.

relink/progress: Adding progress for collecting stage

relink/progress: Adding progress for pruning stage

merge: when current branch has 1 or > 2 heads, actually abort. Currently merge just prints abort-like messages to stderr and then exits with a misleading status 0 (cleverly disguised as "False"). With this change it raises Abort, just like every other fatal error.

tags: remove inactive debugging code. It was left in just in case things went wrong with the tag cache in Mercurial 1.4, so we could easily crank up the verbosity with a one-line change. There have been no problems in 1.4 or 1.5, so it should be safe to drop this now.

tags: remove the old non-caching implementation of findglobaltags(). It was left behind just in case the tag cache blew up in Mercurial 1.4, so we could easily revert to the non-caching code. There have been no problems with the tag cache in 1.4 or 1.5, so it should be safe to get rid of this code now.

keyword: postpone manifest calculation in kwtemplater.overwrite We can check for file existence in the working directory (needed in case of recording) by simply using the given context and calculate the manifest only when there are in fact candidates for expansion/shrinking.

revset: fix call to ctx.extra() in closed()

p4: fix long line and bad spacing around %

Fix shell quoting on p4 conversion

tests: rename test-merge3 to test-add-remove. It tests that 'hg commit' does not crash if the user removes a newly added file, which has nothing to do with merging.

check-code: reformat long lines

dagparser: replace format with str

check-code: catch format(), introduced in Python 2.6

commands: get rid of generic exception in debugbuilddag

check-code: fix check for any/all function The old check would only detect any/all at the beginning of a line. The regexp was probably just modeled after the preceding regexp which (correctly) finds the 'with' keyword at the beginning of a line. We now complain about 'any(' and 'all(' anywhere in a line, unless it is preceded by 'def'. This allows us to define our own compatibility wrapper in util and use 'util.any(' in the code.

hgext/zeroconf/__init__.py: Separate relative and absolute imports. 2to3 complains when relative and absolute imports are mixed, this fix just separates them on the zeroconf extension. According to 2to3, the other modules are fine.

revset: improve filter argument handling

commands: format is a Python2.6 function

debugbuilddag: build a changelog dag from a concise description Useful in tests to quickly build a complex DAG in an empty repo. Handles local tags and named branches. Options to, at each rev, - create a new file, - overwrite the same file, - append to the same file, - write to a specific line in a mergeable file. Can run shell commands during DAG buildup.

debugdag and debugindexdag: emit changelog/revlog DAGs as concise text Mainly useful for reusing DAGs somewhere else, for example for attaching them to a bug report, or for importing them into other environments (like my test environment for incoming/outgoing discovery).

dagparser: parses and formats DAGs as concise text As discussed during the sprint. See the doc comment and doctests for specification and examples. This is used in subsequent patches to export revlog and changelog DAGs, and to generate a repo with a given changelog DAG.

tests: fix up webraw encoding

strip: backup bundles should use the .hg extension

hgweb: specify a charset when sending raw text files Gets the charset from encoding.encoding parameter.

summary: report copies and renames

provide pre- and post- hooks with parsed command line arguments. python hooks are passed two new keyword arguments: - opts: a dict of options; unsepcified options are set to their default - pats: a list of arguments shell hooks receive two new variables containing string representations of the above data: - $HG_OPTS - $HG_PATS for example, the opts and pats for 'hg -f v1.1' would be: {'force': True, 'message': '', 'rev': '', 'user': '', 'date': '', 'local': None, 'remove': None, 'mq': None} ['v1.1']

tests: fix some branch head fallout

pager: set ui.formatted() prior to redirecting stdout.

mq: use ui.formatted() instead of ui.plain().

color: use ui.formatted().

ui: document the formatted(), interactive() & plain() functions.

ui: add ui.formatted configuration variable and accessor function. This new configuration variable is similar in nature `ui.interactive', but applying to output instead of input. This allows as to support non-interactive sessions where formatted should be enabled, such as when using the pager extension. The variable itself is left undocumented; it is not intended for use outside Mercurial and its extensions.

revlog: fix inconsistent comment formatting

improve --branch processing (and differentiate from # syntax) Previously #foo and --branch foo were handled identically. The behavior of #foo hasn't changed, but --branch now works like this: 1) If branchmap is not supported on the remote, the operation fails. 2) If branch is '.', substitute with branch of the working dir parent. 3) If branch exists remotely, its heads are expanded. 4) Otherwise, the operation fails. Tests have been added for the new cases.

help: show value requirement and multiple occurrence of options this helps users to know what kind of option is: - no value is required(flag option) - value is required - value is required, and multiple occurrences are allowed each kinds are shown as below: -f --force force push -e --ssh CMD specify ssh command to use -b --branch BRANCH [+] a specific branch you would like to push if one or more 3rd type options are shown, explanation for '[+]' mark is also shown as footnote.

keyword: pass context to kwtemplater.overwrite Now that we have retrieved the context in every calling function except commit, pass it as argument to kwtemplater.overwrite to avoid looking it up twice. Reorder arguments to kwtemplater.overwrite to reflect their importance. Turn node argument into a simple boolean and rename it to iswctx.

parser: fix missing param in _match

commit: add tests for the 'created new head' message

commit: fix display of 'created new head' message (issue2186) To avoid recurrent fixes for the display of this message, a summary table with all case combinations has been added to the code. Basically, there is two condition for the message to be printed: * this is not an initial (named branch or topo) root * none of the parents are in the current branch heads

rebase: --abort doesn't strip away the target changeset (issue2220) When a changeset is skipped, rebase keeps the previous target as next target and if the skipped cset is the first one, the recorded target is actually the original target. --abort did not detect this situation but simply stripped away the cset.

test-branch-option: Updated output to match rev 3f1409082720

Merge with mpm

discovery: fix description line

Merge with stable

Merge backout

Backed out changeset: e1dde7363601

ui: handle leading newlines/spaces/commas in configlist Thanks to Greg Ward for spotting the problem introduced in 0a548640e012 and for supplying the test case.

mergetools: add configuration for Ultra Compare (issue2226) Initial config provided by Tim Pietzcker

mq: use hg reST role some more

fix encoding bug in 05ac42e56452

cleanups: undefined variables

cleanups: unused variables

move working dir/dirstate methods from localrepo to workingctx

color: labeled text should be passed to ui.write() as ui.labeled Some implementations of ui.label() (HTML versions in particular) must escape the provided text and then markup the text with their tags. When this marked up text is then passed to ui.write(), we must label the text as 'ui.labeled' so the implementation knows not to escape it a second time (exposing the initial markup). This required the addition of a 'ui.plain' label for text that is purposefully not marked up. I was a little pedantic here, passing even ' ' strings to ui.label() when it would be included with other labeled text in a ui.write() call. But it seemed appropriate to lean to the side of caution.

move discovery methods from localrepo into new discovery module

merge with stable

push/pull: fix bug in "--branch ." handling The branch wasn't expanded, so only the tip was operated on.

shrink-repo: wrong variable name

replace Python standard textwrap by MBCS sensitive one for i18n text Mercurial has problem around text wrapping/filling in MBCS encoding environment, because standard 'textwrap' module of Python can not treat it correctly. It splits byte sequence for one character into two lines. According to unicode specification, "east asian width" classifies characters into: W(ide), N(arrow), F(ull-width), H(alf-width), A(mbiguous) W/N/F/H can be always recognized as 2/1/2/1 bytes in byte sequence, but 'A' can not. Size of 'A' depends on language in which it is used. Unicode specification says: If the context(= language) cannot be established reliably they should be treated as narrow characters by default but many of class 'A' characters are full-width, at least, in Japanese environment. So, this patch treats class 'A' characters as full-width always for safety wrapping. This patch focuses only on MBCS safe-ness, not on writing/printing rule strict wrapping for each languages MBCS sensitive textwrap class is originally implemented by ITO Nobuaki <daydream.trippers@gmail.com>.

keyword: use context to detect uncommitted merge

tests: use --debug on test-clone

shrink-revlog: use util.mktempcopy() to preserve mode of index file. (There's still a chmod() call to manually preserve the mode of the data file.)

revset: delay import of hg to avoid start-up import loops

error: fix up test-hgrc

error: update test-trusted.py

revset: fix test failure with qfinish

revset: raise ParseError exceptions

error: add new ParseError for various parsing errors

error: change ParseError to CommandError

hgrc: improve docs for the trusted section

rebase: use usual util.abort rather than error.ParseError

revset: sort the predicate list

revset: fix - handling in the optimizer

revset: fix up tests

walkchangerevs: allow empty query sets

revset: add tagged predicate

revset: optimize the parse tree directly Rather than dynamically optimize in methods, we pre-optimize the parse tree directly. This also lets us do some substitution on some of the symbols like - and ::.

revset: add support for prefix and suffix versions of : and ::

revset: hook into revrange

revset: add a debugrevspec command

revset: introduce revset core

revset: introduce basic parser

remoteui: move from cmdutil to hg

mq: better qqueue test for troublesome names

mq: prevent the creation of a queue whose name is already taken Each check is moved under the code handling the relevant option, and a new one is added for --create. This fixes duplicated entries being added to the queues list.

mq: fix naming issues for qqueue directories

win32text: mark this extension as deprecated

Merge with stable

shrink-revlog: preserve mode of the shrunken index and data file. Otherwise, the shrunken index file always has mode 0600 thanks to mkstemp(). This is annoying on a server, where multiple users may need to read/write the manifest. chmod()ing the data file is not strictly necessary, but it's nice for consistency.

Fixed a bashism with the use of $RANDOM in hgeditor. The variable $RANDOM is not POSIX so a portable /bin/sh may not define it. When creating a directory with a random name it's better to use mktemp, which, even though is not POSIX, exists in common Unixes including Linux, OpenBSD, FreeBSD and MacOS X.

churn: add possibility to include/exclude paths

churn: support spaces in aliases (issue2222)

Merge with stable

Merge with i18n

Added signature for changeset 24fe2629c6fd

Added tag 1.5.4 for changeset 24fe2629c6fd

Merge with i18n

merge with i18n stable

i18n-pt_BR: synchronized with 0bb67503ad4b

Merge with stable

clone: save hardlink state of util.copyfiles() When trying to do hardlink-cloning, the os_link() call of the first file tried already fails on Windows, if the source is on a UNC path. This change avoids calling os_link() for the rest of files, leaving us with a *single* failed os_link() call per clone operation, if the source can't do hardlinks.

util.copyfiles: don't try os_link() again if it failed before If the os_link() call on the first file in the directory fails [1], we switch mode to using shutil.copy() for all remaining files. [1] happens for example on Windows for every file when cloning from a UNC path without specifying --pull.

keyword: retrieve added and modified files from commitctx

keyword: force dirstate normal when all changes in a file are recorded Before this bugfix a file whose changes were entirely recorded was still considered modified by "hg status". Note: the test must use hg record -l/--logfile, because this is not reproducible with hg record -m/--message.