with: use context manager for wlock in overriderevert

with: use context manager for wlock in removelargefiles

with: use context manager for wlock in addlargefiles

with: use context manager for wlock in updatelfiles

with: use context manager for wlock in kw_dorecord

with: use context manager for wlock in kw_copy

with: use context manager for wlock in kw_amend

with: use context manager for wlock in keyword demo

with: use context manager for wlock in _kwfwrite

with: use context manager for wlock in sign

with: use context manager for wlock in checklookup

with: use context manager for wlock in copy

with: use context manager for wlock in workingctx.undelete

with: use context manager for wlock in workingctx.forget

with: use context manager for wlock in workingctx.add

with: use context manager for wlock in graft

with: use context manager for wlock in debugsetparents

with: use context manager for wlock in debugrebuilddirstate

with: use context manager for wlock in copy

with: use context manager for wlock in branch

with: use context manager for wlock in revert

with: use context manager for wlock in remove

with: use context manager for wlock in recordinwlock

with: use context manager for wlock in _writeactive

with: use context manager for wlock in _writerepo

sshpeer: make remotelock a context manager

lock: turn a lock into a Python context manager This lets us greatly simply acquire/release cycles. Code pattern before: try: lock = repo.lock() # zillions of lines of code finally: lock.release() And after: with repo.lock(): # ...

with: use context manager for I/O in changedfiles in patch

with: use context manager for file I/O in memusage

i18n: don't translate a transaction name

chgserver: mark as a built-in extension

chgserver: import background server extension from cHg This extension is copied from https://bitbucket.org/yuja/chg/ -r 86feb5f2e971 It could be imported as mercurial/chgserver.py, but in that case, we would have to resolve circular import between chgserver and commandserver. So I decided to keep it as an extension. chgserver.chgcmdserver -> commandserver.server commandserver._servicemap -> chgserver.chgunixservice

check-code: export needs a space to avoid false positives

rebase: small refactoring to allow better extensibility from extensions Inhibit, one of evolve's extension, would like to change the way rebase works with obsolete changesets. During a rebase with inhibit, the inhibition of the obsolescence markers should be lifted for the rebase. With this small refactoring, inhibit and can wrap the _filterobsoleterevs function to lift inhibition cleanly and at the same time this change makes rebases' code more legible.

tests: non-GNU mv can't move a file onto a hardlink of itself On Solaris, when you try to mv a file to another path that is a hardlink to the original, it complains that they're identical. GNU mv doesn't complain, but it's simpler to just remove the original file instead.

buildrpm: move creation of RPM directories from dockerrpm Handling the creation of the RPM directories in buildrpm is more consistent and takes care of non-docker builds as well.

i18n: fix a test-gendoc.t failure

test-ctxmanager: fix Python 2.6 compatibility problem

util: rename ctxmanager's __call__ method to enter

obsolete: make _computeobsoleteset much faster This patch makes _computeobsoleteset much faster by looping over the draft and secrets as opposed to looping over the successors. This works because "number of draft and secret" is typically way smaller(<100) than the number of successor in the repo (~90k in my checkout of core mercurial as of today). And also because it is very fast to compute "not public()". I timed the code with the following setup: """ from mercurial import hg, ui, obsolete ui = ui.ui() repo = hg.repository(ui, "~/hg") l = repo.obsstore.successors # This caches the result """ With about 90k successors. k=obsolete._computeobsoleteset(repo) before this patch: 10 loops, best of 3: 33.9 ms per loop k=obsolete._computeobsoleteset(repo) after this patch: 10000 loops, best of 3: 83.3 µs per loop

check-commit: sort errors by line number

check-commit: try to fix multiline handling The old code did not understand the difference between the first line of the summary, and a random line in the summary that happened to include a #, or a random line in the changes that happened to include it. 965788d9ae09 is an example where it fails

check-commit: support REVs as commandline arguments usage: * HG_NODE=REV check-commit * hg export REV | check-commit * check-commit REV ...

check-commit: modularize

check-commit: fix summary length regexp

util: simplify file I/O functions using context managers

run-tests: use a context manager for file I/O in TextTestRunner

bundlerepo: use context manager for file I/O in _writetempbundle

shelve: use a context manager for file I/O in listcmd

largefiles: use util.readfile in lfconvert

run-tests: use a context manager for file I/O

largefiles: use util.readfile in overrides

commands: use a context manager for file I/O in debugdiscovery

check-seclevel: use a context manager for file I/O

largefiles: use a context manager in _getfile

util: replace file I/O with readfile