sshrepo: be more careful while reading data This should avoid some tracebacks when the server sends some garbage.

fix markup of templates/{,old/}header.tmpl

templates: move filters to their own module This eliminates just about all Mercurial dependencies in templater.py

templater: move email function to util

templater: remove node import

patchbomb: undo backout and fix bugs in the earlier patch

added an entry about the prefix option inside the hgrc man page

tests for hg serve prefix option

hgweb: fixes to make hg serve prefix handling more robust

Merge with -stable

graphlog: skip filectx parents in other filelogs

Provide better context for custom Python encode/decode filters. While some can function with just some text and an optional command name, others may want a repository object, a ui object, and a file path. Use the enhanced information to good effect in win32text.dumbdecode's warning.

Register data filters in a localrepo instead of util - Changing data filters implementation is easier, adddatafilter() can rewrap filter after inspecting their prototype - Custom data filters really belongs to localrepo, mixing them with generic wrapper like "pipefilter" or "tempfilter" looks wrong. - util.filtertable should not be accessed from extensions

hgweb: forgot to centralize the req.write() calls in hgwebdir

hgweb: centralize req.write() calls

hgweb: explicitly check if requested command exists

hgweb: get rid of raw-related code in hgweb.filerevision()

keyword: detect restricted commands thru variable

webcommands: add missing hg import

Merge with crew-stable

convert: follow svn module parent moves

convert: prevent svn branches to leave the root module tree

convert: checkout svn root revisions Using changed files is enough if whole history is converted. Checkouts are required to restrict the conversion to user-supplied subtree.

convert: fix svn_source.latest()

convert: display source revision id with --verbose

Permitting the import command to accept a --user option. The prose section of the help text for the command already said that -u and -m are accepted, but -u was not listed in the table of options, and did not work. Useful when accepting patches from other people made by hg diff rather than hg export. For completeness, also accepting -d DATE. [CHANGES: rebased against d8878742a924, --no-commit option.]

Merge with crew

Simplify test-paths

merge with crew

Permit glob patterns to use nested curly braces.

Merge after backout

Backed out changeset dc6ed2736c81

keyword: improve use of dicts - actually use iteritems() when we iter over items of a dictionary - store commit hooks in dictionary (safe, as hook.hook() sorts before execution)

keyword: no expansion on qfold (bugfix)

Merge with crew

Ensure that absolutized paths from hgrc do not contain ../ segments.

glog shows at most one file: correct synopsis

Update usage message in test-glog.out AKA: Remind Thomas to run tests even after trivial changes.

Add back elipsis to [OPTION] that was stolen by 56e8a54bf71d

graphlog: add test for filelog graph

graphlog: add filelog revision grapher Graph a filelog's DAG, e.g.: hg glog mercurial/util.py

Test case for hg import --no-commit

Add --no-commit option to import command

win32: remove excessive flushing Our Windows file code was flushing buffers from the operating system to disk, rather than from the application to the operating system.

convert: hg.clonebranches must pull missing parents (issue941)

qrecord: add tests

qrecord: record complements commit, so qrecord should complement qnew Also there is an idea flying around to create something that will complement qrefresh: - maybe 'qammend'? or - 'qrefresh --interactive'? If we settle on '--interactive', then it would be conveniet to add this flag to 'commit' and 'qnew' as well.

Remove unneeded "import os.path"

send conservatively capitalized HTTP headers

demandimport: handling new relative imports Mercurial does not work on python2.6 because __import__ takes an additional argument called level. This patch merely calls the built-in __import__ when level is passed.

hgweb: move HTTP content types out of header templates This removes the ability for templates to add custom HTTP headers, which can easily be re-added if someone needs it. Thanks to asak for repeatedly reviewing this patch and helping to iron out the quirks.

hgweb: revert to showing file instead of offering for download

hgweb: be sure to send a valid content-type for raw files

hgweb: use ErrorResponse instead of custom response

better error reporting for hg serve errors in tests

hgweb: fix a stupid KeyError introduced in a0e20a5eba3c Thanks to Christian Ebert for finding this. Also added a test to prevent this from happening again.

hgweb: cleanup buglet introduced in 956afc025c0f

merge with crew-stable

CVS convert: Find correct parent for new branch (issue704) Previously the parent was determined by the last changeset where the branched file was changed even if the branch is based on an earlier revision. Fix written by mpm.

Test for converting a CVS branch

convert: mercurial sink must be local

Fix double import of 're'.

convert: Do not abort with TypeError if svn commit message is None (issue934)

Prefer i in d over d.has_key(i)

fix test-strip-cross on solaris Apparently "ENVVAR=value command" doesn't work when command is a shell function. Reported by Richard Lowe.

Merge with crew-stable

test-strip-cross: test handling of linkrev crosses in the manifest

Fix test-strip-cross on Solaris Solaris /sbin/sh doesn't do brace expansion

simplify revlog.strip interface and callers; add docstring Also, strip files only after the changelog and the manifest.

strip: calculate list of extra nodes to save and pass it to changegroupsubset When we remove revision N from the repository, all revisions >= N are affected: either it's a descendant from N and will also be removed, or it's not a descendant of N and will be renumbered. As a consequence, we have to (at least temporarily) remove all filelog and manifest revisions that have a linkrev >= N, readding some of them later. Unfortunately, it's possible to have a revlog with two revisions r1 and r2 such that r1 < r2, but linkrev(r1) > linkrev(r2). If we try to strip revision linkrev(r1) from the repository, we'll also lose revision r2 when we truncate this revlog. We already use changegroupsubset to create a temporary changegroup containing the revisions that have to be restored, but that function is unable to detect that we also wanted to save the r2 in the case above. So we manually calculate these extra nodes and pass it to changegroupsubset. This should fix issue764.

changegroupsubset: accept list of per-revlog nodes to include This will allow strip to include in the temporary changegroup some extra file/manifest revisions that should be restored after the truncations. This code doesn't allow specification of changelog nodes since I won't need that right now, the code wouldn't be tested and it's probably possible to do something similar enough by using the bases/heads arguments.

addchangegroup: add option to skip check for empty changelog group

allow the creation of bundles with empty changelog/manifest chunks

repair.py: don't use nested functions.

repair.py: use revs in limitheads

repair.py: add a docstring to bundle; use repo.ui

repair.py: split stripall into two functions; clean it up a bit

repair.py: rename chlog to cl

repair.py: nodes are nodes, revs are revs Use "striprev" instead of just "rev" to make the patch a bit less confusing. This is also a bit clearer than just "revnum".

repair.py: use node.* directly

repair.py: don't import commands.py

test-no-symlinks: fix unbundle output

Merge with crew-stable

Merge with crew-stable

keyword: add very simple mq test

Fix typo.

Fix test output

Merge with crew

hgweb: fast path for sending raw files

hgweb: just return iterables instead of yielding them

hgweb: separate out start_response() calling

hgweb: return iterable, add deprecation note

hgweb: remove some legacy code

keyword: avoid extra diffs when not comparing against working dir When diffing against specified revisions expanded keywords only get in the way. Update test output.

keyword: support mq; handle (q)record more gracefully mq: Ensure that expanded keywords do not make it into patches. - disable expansion when reading filelog - shrink expanded keywords when reading from working dir (wread) (q)record: Avoid additional hunks due to expanded keywords. However this is still a compromise, as keyword expansions are not updated in working directory because record should not overwrite files. Mention above shortcomings and "hg kwexpand" workaround in help and update test output. system argument parsing: Command detection might be slightly more expensive with dispatch._parse, but we will need this for improving "hg diff" output.

convert: enforce svn_source.getchanges() files unicity