test-static-http.t: get kill actually working - signal handlers take two arguments, not one - add missing import sys Before this patch, the $ kill $! at the end of the test just caused a hidden traceback, sys.exit(0) was not executed. The swallowed traceback was: Traceback (most recent call last): File "dumb.py", line 10, in <module> run() File "dumb.py", line 7, in run httpd.serve_forever() File "/usr/lib/python2.6/SocketServer.py", line 224, in serve_forever r, w, e = select.select([self], [], [], poll_interval) TypeError: <lambda>() takes exactly 1 argument (2 given)

hooks: sort any dictionaries set in the environment The actual order of dictionary items is implementation-defined in Python, and differs between CPython and PyPy. With this change, test-hooks.t passes with PyPy.

windows.rename: eliminate temp name race (issue2571) On Windows, os.rename reliably raises OSError with errno.EEXIST if the destination already exists (even on shares served by Samba). Windows does *not* silently overwrite the destination of a rename. So there is no need to first call os.path.exists on the chosen temp path. Trusting os.path.exists is actually harmful, since using it enables the following racy sequence of actions: 1) os.path.exists(temp) returns False 2) some evil other process creates a file with name temp 3) os.rename(dst, temp) now fails because temp has been taken Not using os.path.exists and directly trying os.rename(dst, temp) eliminates this race.

merge with stable

checknlink: use two testfiles (issue2543) Preventing file loss repository corruption (e.g. vanished changelog.i) when Mercurial pushes to repositories on Windows shares served by Samba. This is a workaround for Samba bug 7863, which is present in current latest stable Samba 3.5.6 and various prior versions down to 3.0.26a (the oldest one I tested). Of course this should be fixed in Samba, but there probably aren't that many other applications who use hardlinks that extensively and keep files open like Mercurial, so the pressure to fix this on Samba is probably not that high. And even if the Samba project should be able to fix their bug within a month or two, it will take quite some time until users upgrade their Samba installs.

help: correct documentation for branches keyword

notify: use util.ellipsis() to truncate long subject

smtp: fix for server doesn't support starttls extension Currently we only support enabling TLS by using SMTP STARTTLS extension. But not all the servers support it. With this patch, user can choose which way to enable TLS: * Default: tls = none port = 25 * To use STARTTLS: tls = starttls port = 465 * To use SMTP over SSL: tls = smtps port = 465 To keep backward compatibility, when tls = true, we use STARTTLS to enable TLS. Signed-off-by: Zhigang Wang <w1z2g3@gmail.com>

avoid .split() in for loops and use tuples instead split can be more readable for longer lists like the list in dirstate.invalidate. As dirstate.invalidate is used in wlock() and therefoe used heavily, I think it's worth avoiding a split there too.

hgweb: added revision date to annotate line data The only revision information yielded by the annotate view was the revision number itself. The patch allows the use of per-line revision dates in the corresponding templates.

patchbomb: save introductory message in .hg/last-email.txt

opener: forbid paths ending with directory separator (issue2507) If Linux is asked to open a filename with a trailing directory separator, e.g. "foo/", the open fails with EISDIR. On AIX, the open succeeds, opening file "foo". This causes test-mq-qnew to fail on AIX. Fix by adding 'ends with directory separator' to the conditions checked by the path auditor. Change test to expect auditor fail message.

record: clean up command table The --force option to qnew has become a no-op, so qrecord doesn't need to use it. This allows record's command table to be simplified; in the process of doing so, this patch also cleans up the cmdtable visually.

record: clean up comments and docstrings Rewrap comments and docstrings to a width of 72 chars and copy-edit.

test-doctest: test url.py again, removed by 7cc4263e07a9

merge with stable

https: use web.cacerts configuration from local repo to validate remote repo

extensions: warn about invalid extensions when listing disabled commands Invalid extensions in hgext/ could in some cases cause a crash when searching for unknown commands in disabled extensions. With this change we issue a warning if extracting commands from the extensions fails. Traceback is available on request. Reported on https://bugzilla.redhat.com/show_bug.cgi?id=663183 with forest.py.

convert: subversion should use util.quotecommand to wrap args to popen2 All other callers of util.popen2 and util.popen3 do this, as well as direct callers of subprocess.Popen.

merge with stable

util: concentrate quoting knowledge to windows.py quotecommand() This fixes all callers of util.quotecommand() and place special knowledge of the bugfix in 2.7.1 in a single platform specific place.

templater: clarify engine caching

merge with crew

tests: clean up test-rollback.t The last comment is on the last file line and is useless, seems like it was introduced in 8fdc11fec6ae during unification.

merge with stable

url: fix trailing whitespace

subrepo: basic support for status of git subrepos

subrepo: fix subrelpath for git subrepos This patch conforms gitsubrepo code to the variable naming scheme of the other subrepo implementations. All user-facing path references should be relative to the root repository.

subrepo: fix git archive parsing of directories and symfiles

subrepo: support ignoreupdate in gitsubrepo's dirty()

subrepo: backout 519ac79d680b Unfortunately git 1.6 does not support the upstream parameter in for-each-ref.

url: fix check-code whitespace complaint

templater: use the parser.py parser to extend the templater syntax

templater: give slightly nicer error for unknown map entries

mq: update .hgsubstate if subrepos are clean (issue2499) This patch prevents MQ from creating an inconsistent subrepo state. If the .hgsub file has been changed, and none of the subrepos have uncommitted changes, creating or updating a patch (using qnew, qrefresh, or qrecord) will update .hgsubstate accordingly. If any subrepos _do_ have uncommitted changes, qnew/qrefresh/qrecord will abort. Thanks to pmezard for proposing this solution.

backout f08df4d38442

subrepo: backout f02d7a562a21 backing out f02d7a562a21 because it introduced a bug in .hgsubstate handling.

merge with i18n

merge with stable

fncachestore: copy dh directory before the manifest Before this patch, the copy order on clone was: requires 00changelog.i store\data store\00manifest.d store\00manifest.i store\00changelog.d store\00changelog.i store\dh store\fncache Which provides a theoretical non-zero probability of a race during clone where a very early reader might see a repository with missing revlog files if it sees 00changelog.i before all files inside dh have been copied. The dh directory is similar to the data directory -- just for files with long names (which are hashed). The manifest refers to files in data *and* dh, so dh should be copied before the manifest. This patch improves the copy order to: requires 00changelog.i store\data store\dh store\fncache store\00manifest.d store\00manifest.i store\00changelog.d store\00changelog.i I'm putting fncache to before the manifest while I'm at it, since fncache provides a mechanism to enumerate all repository files without visiting the manifest revisions. fncache depends only on data and dh. Note that data must be copied first, since copying data triggers the creation of the repository write lock in the destination repo (see hg.clone).

merge with i18n

i18n: merge with stable

i18n-da: synchronize with 5dac0d04b838

i18n-de: translated strings for the purge extension

merge with stable

https: warn when server certificate isn't verified Mercurial will verify HTTPS server certificates if web.cacerts is configured, but it will by default silently not verify any certificates. We now warn the user that when the certificate isn't verified she won't get the security she might expect from https: warning: localhost certificate not verified (check web.cacerts config setting) Self-signed certificates can be accepted silently by configuring web.cacerts to point to a suitable certificate file.

merge: document some internal return values.

check-code: catch os.path.relpath

check-code: catch "except as"

tests: eliminate fast-forward merge in test-tag

merge: fast-forward merge with descendant issue2538 gives a case where a changeset is merged with its child (which is on another branch), and to my surprise the result is a real merge with two parents, not just a "fast forward" "merge" with only the child as parent. That is essentially the same as issue619. Is the existing behaviour as intended and correct? Or is the following fix correct? Some extra "created new head" pops up with this fix, but it seems to me like they could be considered correct. The old branch head has been superseeded by changes on the other branch, and when the changes on the other branch is merged back to the branch it will introduce a new head not directly related to the previous branch head. (I guess the intention with existing behaviour could be to ensure that the changesets on the branch are directly connected and that no new heads pops up on merges.)

record: teach parsepatch() about non-git style headers These changes are not useful to record itself, since it is hard coded to always generate git style diffs. But it makes parsepatch() more generally useful for parsing normal patch files.

template: add showbranch template for {branch} Like showbranches, but always yields exactly one branch. Replaces the less correct {branches|nonempty}.

subrepo: avoids empty commit when .hgsubstate is dirty (issue2403) This patch avoids empty commit when .hgsubstate is dirty. Empty commit was caused by .hgsubstate being updated back to the state of the working copy parent when committing, if a user had changed it manually and not made any changes in subrepositories. The subrepository state from the working copies parent is compared with the state calculated as a result of trying to commit the subrepositories. If the two states are the same, then return None otherwise the commit is just done. The line: "committing subrepository x" will be written if there is nothing committed, but .hgsubstate is dirty for x subrepository.

progress: don't compute estimate without a total Without this, computing an estimate crashes. Test included.

subrepo: use low-level git-diff-index for dirty() Despite its name, git-diff-index compares a revision to the files in the working directory. This seems way less sketchy and more future proof than parsing human-readable git-status.

subrepo: defer determination of git's current branch

subrepo: incorporate tracking branches into gitbranchmap

subrepo: use low-level git-for-each-ref command in branchmap This command's output doesn't change across versions, and it also disambiguates cases where there are slashes in local branch names.

progress using tests: disable time estimates to avoid flakiness

progress: include time estimate as part of the default progress format

progress: only show time estimate when progress format contains 'estimate'

progress: fix adding format elements after the progress bar Prior to this change, format elements after the progress bar would show up in the wrong order.

test-progress: test completion estimates and progress bar delay

subrepo: add progress bar support to archive

archive: add support for progress extension

progress: test setting progress.width This also makes the tests readable in a 80-character terminal which is why I made the change to both tests.

tests: add filtercr.py helper for progress tests The two tests that used filtercr.py already used a slightly different version, which explains why test-convert-svn-move.t changed after switching to using $TESTDIR/filtercr.py.

test-archive: whitespace cleanup

progress: explain format strings to translators

subrepo: re-backout 2245fcd0e160, remove test 7397a53219c9 re-removed the bad fix but not the test.

subrepo: re-backout 2245fcd0e160 This bug got introduced into default simultaneously with its backout, so future merges didn't eradicate it.

merge with stable

tag: abort if not at a branch head (issue2552) Since it's usually only desirable to make tag commits on top of branch heads, abort if the working dir parent is not a branch head. -f/--force may be passed to commit at a non-head anyway. Does not abort if working dir parent is a named branch head but not a topological head.

tag: fix uncommitted merge check and error message (issue2542) This patch corrects the check for tagging on an uncommitted merge. We should never commit a new tag changeset on an uncommitted merge, whether or not --rev is specified. It also changes the error message from: abort: cannot partially commit a merge (do not specify files or patterns) to the much more accurate (and terse): abort: uncommitted merge Local tags are ok.

tag: don't check .hgtags status if --local passed Local tags don't create a commit, so we don't need to check the status of .hgtags.

progress: refactor for readability and show XXs instead of 0mXXs.

progress: Add estimated time remaining for long tasks Output looks roughly like this: updating [======================> ] 1547/4842 0m13s output will either show h:m or m:s

progress: react more reasonably to nested progress topics Previously, we'd reset the entire progress bar state when a topic was completed, even if it wasn't the outermost progress topic. Now we print the state of the next progress topic on the stack if one is left rather than reset the progress bar.

Merge with stable

util: work around behavior change in Python 2.7.1

merge with stable

merge default heads in crew and main

merge stable heads in crew and main

eol: improve help on whether EOLs are changed in working copy or repository Improved help to make it more clear for users which changes are only changes in the working copy and which changes that will go into the repository (on the following commit). Futhermore a note on when the rules will be applied to the working directory.

churn: ignore trailing and leading spaces (issue2546)

merge with stable

make_file: always return a fresh file handle that can be closed Currently, cmdutil.make_file() will return a freshly made file handle, except when given a pattern of '-'. If callers would want to close the handle, they would have to make sure that it's neither sys.stdin or sys.stdout. Instead, returning a duplicate of either of the two ensures that make_file() lives up to its name and creates a new file handle regardless of the input.

merge with stable

hook: fix import path handling for repo=None

hook: assume relative path to hook is given from repo root

subrepo: backed out changeset 2245fcd0e160

bundle progress: offer best-guess deterministic progress information This uses the same strategy as progress for pulls, estimating manifests based on changeset count and estimating file count by files list in each changeset.

httpsendfile: record progress information during read() This allows us to provide deterministic progress information during transfer of bundle data over HTTP. This is required because we currently buffer the bundle data to local disk prior to transfer since wsgiref lacks chunked transfer-coding support.

templatekw.showdiffstat: use ctx.diff() instead of calling patch

localrepo: remove unneeded os.unlink call in wwrite The opener already unlinks the filename before 'w'riting, for both the symlink and the normal file case. It also now resets the flags for normal files on 'w'rite, which makes this os.unlink call completely redundant. For Windows, removing this extra unlink call helps to avoid tripping issue2524 (os.unlink followed by a write).

opener: always reset flags on 'w'rite only the patcher needs to preserve flags on write

subrepo: silence git output when ui.quiet is set

subrepo: show git command with --debug Also removing the clutter of --no-pager, since none of these commands use the pager.

subrepo: speed up git push logic In many common cases where the subrepo is up to date with the remote repo, hg push will only need to run one git-branch command to see that nothing needs to be pushed.

subrepo: use subprocess.Popen without the shell As well as simplifying the code, this makes subprocess calls about 25% faster. Tested on a couple linux boxes. python -mtimeit -s'import subprocess' 'subprocess.Popen( "git version", shell=True, stdout=subprocess.PIPE).wait()' python -mtimeit -s'import subprocess' 'subprocess.Popen( ["git","version"], stdout=subprocess.PIPE).wait()'

subrepo: treat git error code 1 as success At least status, commit, merge-base, and diff all return 1 when not failing.

subrepo: gitsubrepo should inherit from abstractsubrepo

subrepo: initialize subrepo relative default paths relative to their root

merge with stable

bookmarks: create undo.bookmarks using repo.opener instead of util.copyfile This more closely matches how the other undo files are created, and we don't care about settings permissions or times on the file, which can fail if the user running hg doesn't own the file.

archival: don't set gzip filename header when there's no filename This mainly affects hgweb, which can generate tar.gz archives without filenames. Without this change, the header would be set to ".gz", which can confuse Safari into extracting the file and renaming it to "gz" when "Open 'safe' files after downloading" is enabled. file(1) before: hg-crew-5e51254ad4d4.tar.gz: gzip compressed data, was ".gz", last modified: Thu Dec 2 11:46:20 2010, max compression after: hg-crew-5e51254ad4d4.tar.gz: gzip compressed data, last modified: Thu Dec 2 11:46:20 2010, max compression

build: don't delete precious version information on 'make clean' setup.py rebuilds version information if it's locally available, regardless if file already exists.

patch: write .rej files without rewriting EOLs Do not pass reject file content to patchfile.writelines() to: - Avoid line endings transformations - Avoid polluting overriding implementations with unrelated data. They should override write_rej() to deal or ignore reject files properly. Bug report, analysis and original patch and test by Shun-ichi GOTO <shunichi.goto@gmail.com>

record: move copystat() hack out of util.copyfile() and into record b2410ed2cbe9 updated copyfile to also copy over atimes and mtimes. That behavior is specifically to trick editors into thinking files that hg record has modified haven't changed. We don't really care about preserving times in the general case.

util: make atomicfiles closable

subrepo: drop arguments unsupported by old git

subrepo: work around old git's remote naming scheme Older git versions use 'origin/master' Current git uses 'remotes/origin/master'

subrepo: use environment variable instead of git commit's --date Older git versions do not have a --date flag.

subrepo: parse git status's human-readable output Older git versions do not have a machine-readable output, but the default format has not changed over time.

subrepo: use subprocess's cwd instead of git's --work-tree Some older git commands (e.g. git merge) do not properly recognize the --work-tree argument, so abstract that away from git.

test-subrepo-git.t: make gitroot pushable earlier Older git versions get confused when the git repo becomes bare.

test-subrepo-git.t: silence git output Older git versions have different ideas of what output is considered quiet, so the tests can be more version-blind.

bdiff: Fix bogus NULL return

bdiff: dynamically allocate hunks Make the hunk list a singly-linked list rather than a large array.