mq: qimport: explain insertion point in doc string

test-convert-mtn: test descriptions with quotes

convert.monotone: fix quotes and backslashes in change descriptions.

Active branches fix (issue1104)

hgweb: correctly validate permissions with streamclone pulling

Fix bad span tag in Readme

test-mq-merge: test mq merge and explicit patch queue

mq: qpush -n fix

tests: don't run test-convert-cvs if there's no cvs server

improve help text for clone -U

Added signature for changeset d5cbbe2c49ce

Added tag 1.0.1 for changeset d5cbbe2c49ce

merge from crew-stable

mq: lose the trailing whitespace

improve help text for push (explanation of -r)

merge from crew-stable

merge from crew-stable

Merge with crew-stable

hgk: difftree must match all files when supplied list is empty

status: fix "added" attributes renamed in bf2bf986ff87

doc/hgrc.5.txt: add anchors for sections Enables links into hgrc.5.html of the form: hgrc.5.html#section

merge from crew-stable

gendoc: add anchors for commands Enables links into hg.1.html of the form: hg.1.html#command

mq: add correct documentation for qpop

Backed out changeset ff5ef3c0fe10 (fixes wrong documentation)

Slight improvement to qpop arguments help

test-mq: update qdiff summary output

mq: make qselect fail properly on an empty guard

mq: expand help text for qdiff Thanks to Waldo for thinking about this and formulating some text.

status: refactor status command - directly calculate a list of states to show - simplify repo.status call - use zip to build file list - only add relevant copies to copy dict

match: cleanup match classes a bit

match: remove files arg from repo.status and friends

match: remove files argument from patch.diff

match: use match.files() for patch.diff

match: stop passing files through commitfunc

status: use match helpers for various users

churn: simplify hg status call

match: use helpers for cmdutil

match: add always, never, and exact methods

remove default arg from patkind

resolve: use cmdutil.match

dirstate: remove default from status list_unknown

purge: remove unused 'missing' logic

dirstate: minor status cleanups

status: rename type_ to state

walk: use match.bad in statwalk

walk: use match.dir in statwalk

walk: make dirstate.walk return a single value too

walk: return a single value

walk: remove cmdutil.walk

walk: remove rel and exact returns

walk: remove more old badmatch logic

walk: remove remaining users of cmdutils.matchpats

cmdutil: switch commit to match

walkchangerev: use new match objects

walk: pass match object to cmdutil.walk - introduce cmdutil.match - change args to cmdutil.walk - create match objects for walk calls

walk: begin refactoring badmatch handling

walk: remove files arg from repo.walk and dirstate.walk

walk: introduce match objects

walk: kill util.cmdmatcher and _matcher

purge: use cmdutil.matchpats

purge: eliminate dopurge

walk: eliminate some default variables

coal: fix CSS errors as reported by Firefox

add compression type type parameter to bundle command

add support for HG10GZ bundles to bundlerepo.bundlerevlog()

tests: fix test-highlight output, changed in ccc2481e3954

merge from crew-stable

rename --after: do not unlink source file (issue910) "rename --after a.txt A.txt" on case mangling systems (e.g. Windows) erroneously deleted A.txt (even if modified).

Make mercurial easy installable again

webcommands: pass full author to annotate, fix templates (issue 1054)

sshserver: Don't try to close fp if mkstemp failed

merge from crew-stable

mq: fix qdiff -U (it apparently never really worked)

mergetools.hgrc patch

mergetools.hgrc patch

merge from crew-stable

merge from crew-stable

coal: put link fragment at the end of the url (after session vars)

merge from crew-stable

mq: don't commit local changes on pushing empty patch (issue1087)

remove trailing spaces

mq: add a little documentation on qpush -f

ignore *.pyd files *.c modules on Windows are compiled to *.pyd

Merge with crew-stable, fix test output

Merge with crew-stable

replace usage of os.popen() with util.popen() To make this possible, I added a mode parameter to both implementations of util.popen(), defaulting to 'r' (as it does in the Python stdlib).

test-merge-internal-tools-pattern: make executable

convert: fix svn file copy detection code

convert: fix svn branch source detection corner case It was confused by composite-looking revisions made by copy + remove.

convert: cleanup svn file copy handling

convert: handle past or foreign partial svn copies Subversion allows revisions to be composed of subparts coming from revisions before the parent or from other part of the repository. There is no simple representation for these now, keep the changes but do not track their origins.

convert: more cleanup in svn directory copy handling

convert: svn directory copy handling cleanup

convert: improve subversion branch filtering

convert: rename get_entry_from_path() into an svn_source method

convert: fix subpaths detection in svn source

test-convert-svn-branches: short identifiers are more legible

cmdutil: handle and warn about missing copy revisions

localrepo: forbid nullid copyrev

verify: check copy source revlog and nodeid

Merge with -stable

removing unused local ccache in patch.diff

merge with -stable

merge: replace readline() call, missing from posixfile_nt

Merge with crew-stable

remove trailing whitespace introduced in cfeeac24fc1e

templates/coal: spell class correctly in {fileentry}

repo: add rjoin method

parseurl: make revs optional

Hoist localpath to the top level

test-imerge: disable due to massive brokenness

use internal merge tool when specified for a merge-pattern in hgrc It is possible to specify in the hgrc file that one of the internal merge tools (internal:fail, internal:local or internal:other) be used for performing merges on files matching a given pattern. However, this setting is not being acted upon (the merge tool is not found). This patch fixes that.

test that internal merge tools are used when set for a merge-pattern in hgrc It is possible to specify that one of the internal merge tools (internal:fail, internal:local or internal:other) should be used for performing merges on files matching a given pattern. This test makes sure that such a setting actually works.

fix import with -p0

context: handle untracked files in filectx.fileflags()

resolve: new command - add basic resolve command functionality - point failed update and merge at resolve

python-2.6: use sha wrapper from util for new merge code

fixed missing sorting in test-dumprevlog The test failed at random because the order of files was non-deterministic, which produced diffs in test output.

test-dumprevlog: fix extra path separator under MacOSX The "find .hg/store/ -name '*.i'" command returned paths with extra separators like '.hg/store//00changelog.i'.

Merge from main.

Add a batch file driver for Windows

merge: introduce mergestate

Merge with crew-stable.

win32text: actually import util so util.binary() can be used

issue 1053: heavily qualify the "cp -al" advice for cloning

Get rid of reimplementations of util.binary

Let util.binary check entire data for \0 (issue1066, issue1079)

keyword: check for '\0' in entire data before acting util.binary might not be safe enough, as it fails eg. on certain pdf files (issue1066).

keyword: make overwrite() arguments mandatory

keyword: compact setting of optional arguments

keyword: privatize remaining monkeypatches by moving them into reposetup Make it more transparent if and when monkeypatches are needed.

keyword: collect filename patterns, wrap dispatch._parse in uisetup Store [keyword] config in kwtools. hgcmd defaults to empty string.

Merge

Tidy code, fix typo

Added hgexecutable support for py2exe/frozen scripts

Remove trailing space

merge with crew-stable

python-2.6: md5 import in test-archive helper

python-2.6: deprecation of GzipFile.filename

highlight: backward compatibility with pygments 0.5.1

Fix spacing error introduced in 5efd447a9b8d

Merge with crew-stable

convert: check svn branches are directories

Merge from crew-stable.

tests/test-diff-copy-depth: using sh instead of bash test failed on plain FreeBSD 6.2 due to nonexistent bash

gendoc: fix indentation

hg.1: bogus escaping of .hg* not needed anymore Revert ba40e267762b, 1c1ef561d541 workarounds.

doc: escape .hg at bol in manpages with make (issue1072) This works around a bug in old docbook stylesheets. Remove .*.swp example from hgignore.5.txt. Diagnosis and workaround thanks to Alexis S. L. Carvalho.

highlight: Generate pygments style sheet dynamically This patch allows a per-repository (for example, within a hgwebdir) selection of pygments_style web option. No static .css files required. Test edited by pmezard (compatibility fixes)

Merge with macencode branch

win32text: rename forbidcrlforcr() function

test-mactext: simplify test, coverage is ensured by win32text tests

win32text: Add macencode/macdecode

Merge with crew-stable

dirstate: do not ignore current directory '.' (issue 1078)

patchbomb: fix cStringIO import (spotted by pychecker)

Make tests/test-dumprevlog executable

merge with main

Merge with crew-stable

Merge backout

Backed out changeset 7f4257b5cbfc util.binary heuristic cannot be trusted when decoding working directory files.

mq: add --force option to strip

coal: fix some glitches in annotate header

python 2.6 compatibility: compatibility wrappers for hash functions

python 2.6 compatibility: add __hash__ to classes that have __eq__

Merge with crew-stable

Let --unified default to diff.unified (issue 1076)

contrib: fix binary file issues with dumprevlog on Windows * Fix traceback on dumprevlog: open revlog in mode 'rb' (binary) * Set stdio to binary mode so that redirected data is not modified References: http://selenic.com/pipermail/mercurial/2008-March/018330.html http://selenic.com/repo/index.cgi/hg/rev/a3fe91b4f6eb

add tests for contrib/dumprevlog and undumprevlog

coal: minor icon improvements

coal: more tweaks - move icons inside links - set icon borders to 0 - make active menu button white on black and remove bar

pager: make config info accessible with "hg help pager"

tests: easier error diagnostics for test-serve

Merge from crew-stable.

hgweb: make hgwebdir work in the absence of PATH_INFO Thanks to Andrea Arcangeli for reporting and an initial patch.

Merge from main.

Merge from crew-stable.

Pager extension: switch it off if --debugger is set The pager is preventing the debugger prompt and much of the debugger output to be refreshed. Moreover the pager does not make sense when debugging line by line. (This supersedes the similar ui.debugflag patch. Disabling the pager for debug output doesn't make that much sense, as this is actually when the pager might be useful.)

Pager extension: switch it off if --debugger is set The pager is preventing the debugger prompt and much of the debugger output to be refreshed. Moreover the pager does not make sense when debugging line by line (Thanks to Gilles Moris).

coal: even more tweaks - rearrange file browser columns - add folder and file icons - make browse table uniformly monospaced - use lowercase consistently for column heads - fix search container div

coal: more tweaking - add quirksmode hack in attempt to make MSIE happy - add container divs for more browsers compatibility - fix file log button on file page - link to first line of file source from browser - change revision navigator style

coal: more tweaks - add revision numbers to changeset, browse, and file pages - add revision number to revision number navigator in log - remove node from log and filelog pages - add changeset, browse, and archive buttons back to log page - make buttons more clickable with display: block - fix alignment of some table heads

coal: minor fixes sans -> sans-serif tweak column widths get rid of full changelog

merge with crew-stable

test-churn: force COLUMNS to be set

Consistently 1 space after full stops in command doc strings Update test output accordingly.

patchbomb: Fix mangling of lines beginning with From we should mangle the lines beginning with From only when writing to mbox format.

asciidoc: consistently use 1 space after full stop Note: the command doc strings included in hg (1) are still inconsistent wrt to spaces after full stops.

hgrc.5: fix subsection formatting

imported patch templ.diff

Merge with crew-stable

merge with crew-stable

rollback: unlink files truncated to length 0

make test-diff-copy-depth executable

convert: fix test-convert-svn-* problems with mtime not changing The `svn commit` command does not detect changed files unless their mtime has changed. A quick succession of, for instance, `svn co ...; echo x >> y; svn ci` can thus lead to the change to y being ignored. Edited by pmezard to write in binary mode.

merge with hg-stable

hgweb: fix merge breakage

templates: add coal web theme

optimize PNGs

hgweb: minor improvements for new web style - make renamelink take a context - pass current node to shortlog page rather than tip - pass user/desc/date/rename to filediff - improve template error reporting for foo%bar

add simple dump and undump scripts to contrib/

Merge with stable

copies: refactor symmetricdifference as _findlimit We only need to track the lowest revision seen, which makes things simpler.

copies: teach symmetric difference about working revisions - use changelog.count() as a pseudo revision number - abort early in copies if revs are the same - eliminate working dir hacks in copies - yield results as they're found

symmetricdifference: move back to copies It's too tightly dependent on known revlog ordering to fit well in ancestors

symmetricdifference: change colors to sides

ancestors: simplify symmetric difference - n_wanted/wanted -> interesting - scan colors rather than managing ret

copies: don't double-detect items in the directory copy check

copies: skip directory rename checks when not merging The directory rename checks are not meaningful unless we're moving or copying files across a branch during a merge.

copies: sort old names by depth

backout: reverse changeset belongs on current branch Backing out a changeset that is before a named branch branchpoint was making the reverse changeset the tip of the old branch, which is wrong and very confusing. So instead, we put it on the current named branch.

copies: fix silly precedence bug

churn: allow whitespaces as delimiter in aliases it was exactly on space character before.

merge with crew

hg serve: add clearer message when starting the server with --verbose Explicitly tell the address is the one the process is bound to. Fix the printing of IPv6 addresses.

convert: fix trailing space introduced in 5efd447a9b8d

pager: Add a configuration to enable/disable the pager for certain commands Add the configuration options pager.ignore and pager.attend. You can disable the pager on certain commands by adding them to the pager.ignore setting. To whitelist commands, you can add them to pager.attend. To disable or enable global commands like 'hg version' or 'hg help' you have to use your global .hgrc. (thanks, Matt Mackall)

keyword: regroup monkey patch code, underscore prefix private vars

merge with crew-stable

Fix test-convert-svn-tags on case sensitive filesystems

fix bug introduced by 2540521dc7c1 (thanks pychecker)

merge with -stable

fix error spotted by pychecker

fix bug introduced by 894875eae49b (short not defined)

merge with crew

fix warning from pychecker

fix error spotted by pychecker

Merge with crew-stable

commands: fix shadowed repo module

merge with crew

hgweb: remove dead code

Merge with crew-stable

test-convert-svn-source: remove redundant tests

convert: allow tags detection to be disabled Tags calculation becomes more reliable but slower, measurable on frequent synchronizations.

convert: follow svn tags history (issue953)

Add a test for svn tags conversion

convert: avoid svn log retrieval process cleanup With huge history (like kdelibs), the process termination suddenly consumes a lot of memory (from 700M to 1.3G+). Since the job is done, clean termination is not required, just exit.

test-convert-mtn: test directory deletion

fix const annotation warning

highlight: some small cleanups

hgweb: refactor hgweb code

hgweb: separate out utility functions

Backed out changeset d2bb66a8a435 (temporary template compatibility)

mercurial.el: note new Emacs VC mode support

manifest: improve parsing performance by 8x via a new C extension

Added signature for changeset bae2e9c838e9

Added tag 1.0 for changeset bae2e9c838e9

Tests for issue 1049

commit: when committing the results of a merge, it's all or nothing Previously, it was possible to commit just one file from a merge, which is essentially always the wrong thing to do. This fixes issue 1049.

Skip older monotone versions for tests. test-convert-mtn is known to work with 0.37 and 0.38, but fails with 0.31 (Debian etch). hghave now skips all versions up to 0.31.

run-tests.py: Only one fail message when output changed and error code. This fixes passing back fail messages mistaken for skip messages when running with parallel jobs because run_children() only expects one message per fail.

mq: use repo.changelog.tip() instead of the "tip" tag

Fix missing space in one of the new update messages.

qclone: do not abort if remote hasn't enabled mq (issue1040)

hgweb: add compatibility code for old templates Up to changeset 3340aa5a64f7, HTTP headers were expected to be embedded in the "headers" template. Since that changeset, the content-type is supposed to be defined as the "mimetype" template in the map file. This changeset makes sure the old templates still work.

Merge with crew

test-convert-mtn: test directory move

convert: record deleted files in monotone source

update: better logic and messages for updates - complain about attempts to merge with ancestor - when updating, differentiate between - crossing named branches with no local changes (jump) - crossing named branches with local changes (complain) - nonlinear update on the same named branch, no changes (complain some more) - nonlinear update on the same named branch, changes (different complaining)

hgweb: fix breakage in python < 2.5 introduced in 2c370f08c486

setup.py: hide compiler error messages while searching for inotify

Add a test for monotone conversion

don't use hasattr in repo.invalidate hasattr ends up calling __getattr__ which will instantiate the very attributes we want to remove.

localrepo.commit: grab locks before getting the list of files to commit Somebody may change the dirstate after we've determined the parents of the working dir and run repo.status, but before we called wlock(). This should also fix issue997, where backout would change a file without changing its size and then call repo.commit without passing the list of files. If this happened in less than one second, we wouldn't detect any file changes - the in-memory dirstate still has the cached stat data for that file. Grabbing the wlock early causes the dirstate to be invalidated and we end up reading the dirstate file again, which has that file marked for lookup (size == -1). A better fix would be for backout to give repo.commit the exact list of files, but that'll require some changes to the revert operation. A significant user-visible change is that the precommit hook is always run with both locks grabbed - previously, hg commit would run it before grabbing any locks, but hg import would run it after grabbing locks.

Avoid calling heads() twice on every hg commit. In an extreme case (merging two revisions with very low revision numbers) this could be slower than the previous code, but it should be much faster in the usual cases (parents are near the tip). It also avoids some races in some uninteresting cases (e.g. two concurrent hg commits).

hgweb: better error messages

Remove unexpected "Alternately" word from tip help.

run-tests.py: Allow environment variables to set jobs/timeout/port.

convertrepo: make it work with refactored convert extension

Make tip help more helpful

test-imerge: test resolve subcommand

imerge: replace "merge" with "internal:merge" when non-interactive

hgk: display orphan revisions content (issue 1041)

test-mq-symlinks: skip if symlinks are not supported

test-notify: make it pass on Mac OS X 10.5

test-remove: make tests really pass Ah, the joys of cross-platform shell scripting...

test-remove: fix for non-gnu find

test-remove: make it pass on Mac OS X 10.5

tests: add highlight extension tests

tests: add tests for hgweb to test-keyword

Merge main and crew.

removed trailing spaces

ui: disallow newlines in usernames (issue1034)

update: no -C needed to switch branches without outstanding changes

coverage: deal with symlinked input paths (MacOSX issue) /tmp is symlinked in MacOSX therefore test scripts as well as python modules are installed in a symlinked location. coverage.py uses abspath() to normalize its inputs which fails if these are referencing real paths. Use realpath() instead.

Make churn an official extension

tiny tab cleanup

improved semantics for remove (issue438) - Added files are never deleted (only removed with --force). - Modified files can only be removed with --force. - With --after, only deleted files are removed. - With --after --force, all files are removed but not deleted.

test: fix mixing exec bit on test-hgwebdirsym

tests: make test-rebuildstate repeatable

tests: teach -i about fails list

tests: make test-convertcvs repeatable

test: Add tests for webdir symlinks and walkrepos.

mq: warn when applying a patch to somewhere other than tip

util: check fileno() validity in win32 set_binary()

clone: print "updating working directory" status message With this change, "hg clone" looks like this: % hg clone http://example.com/repo/big big requesting all changes adding changesets adding manifests adding file changes added XXX changesets with XXX changes to XXX files updating working directory XXX files updated, XXX files merged, XXX files removed, XXX files unresolved So the user sees % hg clone http://example.com/repo/big big requesting all changes adding changesets adding manifests adding file changes added XXX changesets with XXX changes to XXX files updating working directory while Mercurial is writing to disk to populate the working directory With this change, "hg clone" looks like this: % hg clone big big-work updating working directory XXX files updated, XXX files merged, XXX files removed, XXX files unresolved

give better error message on non-existent mapfile (issue813)

warn about new heads on commit (issue842)

hgweb: only accept POST requests for unbundle

tab/space cleanup

ui: copy overlay from both direct and indirect parentui

convert: allow missing tools not to stop source type detection

fix test-serve breakage for other hostnames

util: test fileno() availability in win32 set_binary() Fix suggested by Alexander Belchenko <bialix@ukr.net>

tests: fix repeatability for test-filebranch with dirstate granularity

tests: remove some unnecessary sleeps

dirstate: refactor granularity code, add a test - rename option dirstate.granularity - move option reading into .write() - add a simple test

dirstate: ignore stat data for files that were updated too recently This should fix the race where hg commit foo <change foo without changing its size> happens in the same second and status is fooled into thinking foo is clean. A configuration item is used to determine the timeout, since different filesystems may have different requirements (I think VFAT needs 3s, while most Unix filesystems are fine with 1s).

pager: remove pager code from core

pager: further simplify code, clean up comments

Use the pager given by the environment to display long output Unix systems usually have a PAGER environment variable set. If it is set, mercurial will use the pager application to display output. Two configuration variables are available to influence the behaviour of the pager: pager.application sets the application to be used pager.quiet silences Broken Pipe errors that might occur when the user quits the pager before mercurial finished to write the output

hgk: don't exit if mercurial commands only print warnings

tag: allow multiple tags to be added or removed - Example: "hg tag -r 42 build-25 beta-1" will add tags build-25 and beta-1 for rev 42. - The deprecated and undocumented usage "hg tag arg1 arg2" used to emit a warning, then add tag arg1 for rev arg2 (equivalent to "hg tag -r arg2 arg1"). It will now add tags arg1 and arg2 for the current revision. - If one tag triggers an error, no tags are added/removed (all or nothing).

tests: update test-command-template for the new {isodatesec} filter

Add an {isodatesec} template, to show seconds too.

convert: Add convert.cvsps option to set up an alternate cvsps command line.

hgwebdir: Tiny fix for webdir on non-symlink capable platforms.

Add tests for cloning from a all-history bundle

clone: use cancopy

Add ability to directly clone from all-history bundles bundlerepos can be used as clone src, even if CWD is not a repo

remoterepo: no longer needed All users already use repo.local() to test for local, which is false in the repository base class. statichttprepository never derived from this class anyway.

repo classes: remove unused dev() method

Add default local() and cancopy() methods to repository base class

make hgrc man page tell about ~/.hgrc on Windows

ui: make chuild ui feed output to parentui's active buffer

convert: fix test complaint for mtn bits

cleanup monotone conversion and use commandline class

initial version of monotone source for convert extension

notify: print diffstat even if maxline == 0

Make clear that for a changegroup '-r' pulls "up to" a revision If some better wording is found, it can be changed later.

two fixes for the pager stuff - open pipe in binary mode - don't start a pager if stdout is not a tty (this prevents test-ssh from hanging if you set ui.usepager=True in run-tests.py)

Use the pager given by the environment to display long output Unix systems usually have a PAGER environment variable set. If it is set, mercurial will use the pager application to display output. Two configuration variables are available to influence the behaviour of the pager. ui.pager sets the pager application. The pager is only used if ui.usepager is true. By default ui.usepager is disabled.

Fix thinko in test-inherit-mode If we don't change any rwx bit in the last test, hg will skip the calls to chmod since it'll assume they're not needed. This might fix things on BSD systems.

test-serve: replace copy/paste with shell function

revert: update state of files in the "checkout" list This can make a difference when there are filters involved and decode(encode(working-dir-data)) != working-dir-data even though encode(decode(repo-data)) == repo-data An example is a working dir file that uses only \n when you're using the win32text extension.

normallookup: during merges, restore the state saved by remove

dirstate.remove: during merges, remember the previous file state We encode the previous state as a negative file size (AFAICS, previous versions of hg always have size == 0 when state == 'r'). We save the state of 'm'erged and dirty files, because they're the two states that indicate that a file has to be committed on a merge to correctly record per-file history.

debugstate: add --nodates This can be useful for tests.

patch: fix corner case with update + copy patch handling (issue 937) The self patching of files when diffed with a backup is a bit peculiar to me. It makes sense in mpatch, that's less clear in mercurial patching code. Let's document and preserve it for now.

Merge

Backed out changeset 4b81eecc8aa2

Merge

Backed out changeset 91ac1726730a

Automated merge with http://hg.intevation.org/mercurial/crew

issue 1003: send all data properly

Remove unneeded "import os.path" again.

Spacing cleanup

filectx.ancestor: use fctx._repopath to cache filelogs (issue1035) _repopath may be different from _path for workingfilectx when there are renames involved.

Change User-agent field in client to have a lot more info.

Allow hgwebdir collections to follow symlinks.

Revert cfebb175704f, fixing the output of 2 tests

test-mv-cp-st-diff: avoid linkrev collisions The copy detection code relies heavily on the linkrevs, so it's better to avoid collisions in this test.

convert: avoid querying log of foreign svn branches Server configuration may prevent access to other branches, so don't even try to get them, instead of ignoring them afterwards.

patch: check filename is /dev/null for creation or deletion (issue 1033) Otherwise, file creation or deletion is ambiguous with file being filled or emptied.

update tests for copy detection changes

copies: add tests for status -C/diff --git This goes through 42 combinations of copy/rename branch/parent/working dir.

copies: deal with symmetric difference for linear relationships choose the minimum of symmetric difference and the two revs in question

status: find copies and renames beyond the working directory

diff: use copy smarts from copies.py

copies: move findcopies code to its own module - pass in contexts - fold symmetricdifference check into copies.copies

merge: move symmetricdifferences to ancestor.py

merge: simplify some helpers

merge: more simplifications to checkcopies

merge: simplify checkcopies

merge: privatize some functions, unnest some others

merge: notice uncommited copies (issue1000) As mentioned in msg5349 in issue988, "hg update" doesn't take into account uncommitted copies. To reproduce: ---------------- hg init repo cd repo touch foo hg ci -Am 'add foo' echo >> foo hg ci -m 'change foo' hg up -C 0 hg mv foo bar HGMERGE=false thg --debug -y update -------------------------- A similar problem happens with hg merge --force. I'm attaching a possible patch.

Check for patches repo with os.path.isdir not os.path.exists

hook.py: fix redirections introduced by 323b9c55b328 The only non-obvious part is the use of sys.{__stderr__,__stdout__}, which is needed because sshserver overrides sys.stdout. This makes a test that I added back in revision 7939c71f3132 ineffective.

Issue937: error messages from hooks not sent over HTTP. Turns out that stderr - where ui.warn would send messages - was not being proxied over the HTTP connection. stdout was, and it seems you need both. (The streams are interleaved for readability.) Tested on Ubuntu 7.10 with lighttpd on hgweb.cgi with HTTP Basic auth, no SSL, using a changeset failing win32text.forbidcrlf.

revlog.py: remove extra close() atomictempfile.rename() already calls close().

adjust test-http output to non-default HGPORT, e.g. with run-tests.py -j

hgweb: clarify which address and port can/cannot be bound at startup (bug 769) The error message at startup when the address/port could not be bound was confusing: hg serve abort: cannot start server: Address already in use Be more explicit: $ hg serve -a localhost abort: cannot start server at 'localhost:8000': Address already in use Also be more explicit on success, showing hostname and ip address/port: $ hg -v serve -a localhost -p 80 listening at http://localhost/ (127.0.0.1:80) We are careful to handle a missconfigured machine whose hostname does not resolve, falling back to the address given at the command line. Remove a dead-code error message.

revlog: make sure the files are closed after an exception happens This prevents a corruption when the writes happen after the truncate initiated by the transaction rollback.

merge with crew

make sure not to reuse an existing append-file from a previous failed pull

Fix issue995 (copy --after and symlinks pointing to a directory) I haven't looked at other places that call os.path.isdir to make sure they're OK.

dirstate: ignore mode changes if the fs does not supports the exec bit This can make a difference when e.g. the repo is exported through NFS (which support exec bits) and CIFS (which does not).

merge: require --force when there are deleted files

add a test for a375ffc2aa1b

localrepo.commit: normalize commit message even for rawcommit. This normalization consists of: - stripping trailing whitespace - always using "\n" as the line separator I think the main reason rawcommit was skipping this normalization was an attempt to preserve hashes during an hg->hg conversion. While this is a nice goal, it's not particularly interesting in practice. Since SHA-1 is so strong, the only safe way to do it is to have absolutely identical revisions. But: - if the original revision was created with a recent version of hg, the commit message will be the same, with or without that normalization - if it was created with an ancient version of hg that didn't do any normalization, even if the commit message is identical, the file list in the changelog is likely to be different (e.g. no removed files), and there were some old issues with e.g. extra file merging, which will end up changing the hash anyway - in any case, if one *really* has to preserve hashes, it's easier (and faster) to fake a partial conversion using something like: hg clone -U -r rev orig-repo new-repo hg -R new-repo log --template '#node# #node#\n' > new-repo/.hg/shamap Additionally, we've had some reports of problems arising from this lack of normalization - e.g. issue871, and a user that was wondering why hg export/hg import was not preserving hashes when there was nothing unusual going on (it was just import doing the normalization that had been skipped). This also means that it's even more unlikely to get identical revisions when going $VCS->hg->$VCS.

debugancestor: use repo.lookup when no revlog was specified

update output of test-hgweb

setup.py: use a simplified custom version of CCompiler.has_function The original one doesn't remove its temporary files and even creates a temporary file in the CWD.