5166b7a84b72fdf8d65f0ec8f3310af26a25978a: censor: use absolute_import
Gregory Szorc <gregory.szorc@gmail.com> - Tue, 09 Feb 2016 17:33:10 -0800 - rev 30201
Push 184 by gszorc@mozilla.com at Tue, 23 Feb 2016 19:37:02 +0000
censor: use absolute_import
2f0384242b35253850b8bd40666d07f9c08c21f0: bugzilla: use absolute_import
Gregory Szorc <gregory.szorc@gmail.com> - Tue, 09 Feb 2016 17:31:50 -0800 - rev 30200
Push 184 by gszorc@mozilla.com at Tue, 23 Feb 2016 19:37:02 +0000
bugzilla: use absolute_import
8113c88b8e6dece3c2c7f3818dac5e58c2e57302: blackbox: use absolute_import
Gregory Szorc <gregory.szorc@gmail.com> - Tue, 09 Feb 2016 17:30:38 -0800 - rev 30199
Push 184 by gszorc@mozilla.com at Tue, 23 Feb 2016 19:37:02 +0000
blackbox: use absolute_import
a1163ee26e4a94d713f2355c90f4c317fb4e9494: acl: use absolute_import
Gregory Szorc <gregory.szorc@gmail.com> - Tue, 09 Feb 2016 17:29:39 -0800 - rev 30198
Push 184 by gszorc@mozilla.com at Tue, 23 Feb 2016 19:37:02 +0000
acl: use absolute_import Continuing the march towards Python 3.
19424f960bf5891a1a84e8bd1bc1ccbcdfb913d1: checkunknown: audit path before checking if it's a file or link
Durham Goode <durham@fb.com> - Thu, 11 Feb 2016 17:23:10 -0800 - rev 30197
Push 184 by gszorc@mozilla.com at Tue, 23 Feb 2016 19:37:02 +0000
checkunknown: audit path before checking if it's a file or link Previously we would lstat the file to see if it was a file or a link before attempting to process it. If the file happened to exist across a symlink, and if that symlink was pointing to a network file system, that check could be very expensive. The new logic audit's the path to avoid symlinks before performing the lstat on the file itself. In our situation, this shaved 10 minutes off of certain hg updates. 300 files * (2 seconds - the network filesystem lookup time)
0b7ce0b16d8a629cee8610c62c16307ccefa5893: pathauditor: change parts verification order to be root first
Durham Goode <durham@fb.com> - Thu, 11 Feb 2016 17:04:33 -0800 - rev 30196
Push 184 by gszorc@mozilla.com at Tue, 23 Feb 2016 19:37:02 +0000
pathauditor: change parts verification order to be root first Previously, when we verified the parts of a path in the auditor, we would validate the deepest directory first, then it's parent, and so on up to the root. If there happened to be a symlink in the chain, that meant our first check would likely traverse that symlink. In some cases that symlink might point to a network filesystem that is expensive, and therefore this simple check could be very slow. The fix is to check the path parts starting at the root and working our way down. This has a minor performance difference in that we used to be able to short circuit from the audit if we reached a directory that had already been checked. Now we can't, but the cost is N dictionary look ups, where N is the number of parts in the path, which should be fairly minor.
65d24ca3549625aee4ccd4713bb745336610be40: chg: forward job control signals to worker process (issue5051)
Yuya Nishihara <yuya@tcha.org> - Tue, 19 Jan 2016 22:31:59 +0900 - rev 30195
Push 184 by gszorc@mozilla.com at Tue, 23 Feb 2016 19:37:02 +0000
chg: forward job control signals to worker process (issue5051) This is necessary to suspend/resume long pulls, interactive curses session, etc. The implementation is based on emacsclient, but our version doesn't test if chg process is foreground or not before propagating SIGCONT. This is because chg isn't always an interactive session. If we copy the SIGTTIN/SIGTTOU emulation from emacsclient, non-interactive session can't be moved to a background job. $ chg pull ^Z suspended $ bg %1 [1] continued [1] suspended (tty input) # wrong https://github.com/emacs-mirror/emacs/blob/0e96320/lib-src/emacsclient.c#L1094
c0d1bf1b26b71e747e07bb78cbccbfc433ff1da4: chg: verify return value of sigaction() and sigemptyset()
Yuya Nishihara <yuya@tcha.org> - Fri, 29 Jan 2016 22:52:16 +0900 - rev 30194
Push 184 by gszorc@mozilla.com at Tue, 23 Feb 2016 19:37:02 +0000
chg: verify return value of sigaction() and sigemptyset() They should never fail, but it couldn't hurt to be a paranoid.
3fc45956c978f5e2fbb7ad912a75fbffc80e8b4a: chg: initialize sigaction fields more reliably
Yuya Nishihara <yuya@tcha.org> - Fri, 29 Jan 2016 22:42:22 +0900 - rev 30193
Push 184 by gszorc@mozilla.com at Tue, 23 Feb 2016 19:37:02 +0000
chg: initialize sigaction fields more reliably It seems calling memset() and sigemptyset() is common pattern to initialize sigaction. And strictly speaking, sigset_t must be initialized by sigemptyset() or sigfillset(). I saw git and uwsgi do that way, so let's follow them.
550097d01ca3573626ae2eebe6c63b53f6084cde: tests: confirm that a badly documented extension doesn't cause a crash
Simon Farnsworth <simonfar@fb.com> - Fri, 12 Feb 2016 06:25:05 -0800 - rev 30192
Push 184 by gszorc@mozilla.com at Tue, 23 Feb 2016 19:37:02 +0000
tests: confirm that a badly documented extension doesn't cause a crash An external extension whose docstring doesn't conform to Mercurial standards used to cause crashes. Test that we omit such extensions when you do a keyword search.
db90cb1e89c8b7aa5cba363ba2f8d9c08fd5e07e: bookmarks: avoid creating a nested repository during testing
Martijn Pieters <mjpieters@fb.com> - Fri, 12 Feb 2016 14:24:48 +0000 - rev 30191
Push 184 by gszorc@mozilla.com at Tue, 23 Feb 2016 19:37:02 +0000
bookmarks: avoid creating a nested repository during testing This helps the test to pass with hgwatchman, which would otherwise need to be taught about a nested .hg directory. hgwatchman already blacklists test-nested-repo.t which covers the actual usecase
a6344df5108e0333b057967db3d298a1d5cf1982: dispatch: strip command line options like config file options
Tony Tung <ttung@fb.com> - Mon, 08 Feb 2016 15:35:30 -0800 - rev 30190
Push 184 by gszorc@mozilla.com at Tue, 23 Feb 2016 19:37:02 +0000
dispatch: strip command line options like config file options Currently, whitespace in command line --config options are considered significant while whitespace in config files are not considered significant. This diff strips the leading and trailing whitespace from command line config options.
37b818cad14664f7cf262a3cfb1f81e616dd1f4d: hook: for python hook ImportErrors, add note to run with --traceback
Siddharth Agarwal <sid0@fb.com> - Thu, 11 Feb 2016 22:52:23 -0800 - rev 30189
Push 184 by gszorc@mozilla.com at Tue, 23 Feb 2016 19:37:02 +0000
hook: for python hook ImportErrors, add note to run with --traceback I personally found it completely non-obvious that --traceback prints out stack traces for failed imports.
0c9e914029be47569f58cd39ce1f1763d2b03633: hook: fewer parentheses for hook load errors
Siddharth Agarwal <sid0@fb.com> - Thu, 11 Feb 2016 22:41:20 -0800 - rev 30188
Push 184 by gszorc@mozilla.com at Tue, 23 Feb 2016 19:37:02 +0000
hook: fewer parentheses for hook load errors This matches 'hook failed' warnings. We're also going to add hints to some of the hook load errors. Without this change we'd have two pairs of parens for a single error message, which looks really cluttered.
2058e1a894f2ca6c4a971f4cdd540df8a49f91da: hook: use sys.exc_info rather than the deprecated equivalents
Siddharth Agarwal <sid0@fb.com> - Thu, 11 Feb 2016 22:02:52 -0800 - rev 30187
Push 184 by gszorc@mozilla.com at Tue, 23 Feb 2016 19:37:02 +0000
hook: use sys.exc_info rather than the deprecated equivalents sys.exc_type etc have been deprecated since Python 1.5.
27ae22a4f9f9ee6c01a2f82489c1aaacd59d88f0: doc: describe full help document hierarchy to create a valid link in HTML
FUJIWARA Katsunori <foozy@lares.dti.ne.jp> - Thu, 11 Feb 2016 23:15:34 +0900 - rev 30186
Push 184 by gszorc@mozilla.com at Tue, 23 Feb 2016 19:37:02 +0000
doc: describe full help document hierarchy to create a valid link in HTML For example, ":hg:`help config.default-push`" creates an invalid link to "hgrc.5.html#default-push" in HTML, but ":hg:`help config.paths.default-push`" creates a valid link to "hgrc.5.html#paths".
18c6b271579b190472560a8667c0cff3ea129a34: doc: translate from :hg:`help config.SECTION` to a valid link to hgrc.5.html
FUJIWARA Katsunori <foozy@lares.dti.ne.jp> - Thu, 11 Feb 2016 23:15:34 +0900 - rev 30185
Push 184 by gszorc@mozilla.com at Tue, 23 Feb 2016 19:37:02 +0000
doc: translate from :hg:`help config.SECTION` to a valid link to hgrc.5.html Before this patch, ":hg:`help config.SECTION`" in online help text is translated to a link to "hg.1.html#config.SECTION" in HTML unintentionally. This patch translates from :hg:`help config.SECTION` in online help text to a valid link to "hgrc.5.html#SECTION" in HTML. This patch ignores element(s) under "SECTION" (e.g. "ITEM" of ":hg:`help config.SECTION.ITEM`"), because there is no way to refer directly to it in HTML, yet.
63eae465095e3acd63a1894693d6701a48918156: doc: translate from :hg:`help config` to a valid link to hgrc.5.html
FUJIWARA Katsunori <foozy@lares.dti.ne.jp> - Thu, 11 Feb 2016 23:15:34 +0900 - rev 30184
Push 184 by gszorc@mozilla.com at Tue, 23 Feb 2016 19:37:02 +0000
doc: translate from :hg:`help config` to a valid link to hgrc.5.html Before this patch, ":hg:`help config`" in online help text is translated to a link to "hg.1.html#config" in HTML, even though actual "hg help config" shows not help for "hg config" command but "config" help topic, and all of current ":hg:`help config`" expects the latter. This patch translates from ":hg:`help config`" in online help text to a link to "hgrc.5.html" in HTML as expected. This patch also allows ":hg:`help -c COMMAND`" style to link "hg.1.html#COMMAND" for readability.
a1924bc6e267b818fe77ae6bbb617971bef32955: i18n: calculate correct line number in source of messages to be translated
FUJIWARA Katsunori <foozy@lares.dti.ne.jp> - Thu, 11 Feb 2016 23:15:34 +0900 - rev 30183
Push 184 by gszorc@mozilla.com at Tue, 23 Feb 2016 19:37:02 +0000
i18n: calculate correct line number in source of messages to be translated Before this patch, line number in source of the message to be translated is wrong in hg.pot, if corresponded message is placed after ".. DIRECTIVE::", because number of lines related to such directive isn't added to variable "delta", which holds number of untranslated lines in given text. This patch always adds "2" to "delta", because text block is split into translation units by "\n\n".
c4bec3c45ec9fe292965be3867b28db691ede818: revsetbenchmark: handle exception case
Durham Goode <durham@fb.com> - Wed, 10 Feb 2016 12:39:25 -0800 - rev 30182
Push 184 by gszorc@mozilla.com at Tue, 23 Feb 2016 19:37:02 +0000
revsetbenchmark: handle exception case If the revset being benchmarked has an exception, the handling code was encountering an error because the exception did not always have an "output" attribute (I think it's a python 2.7 thing).
(0) -30000 -10000 -3000 -1000 -300 -20 +20 +50 +100 +1000 +3000 +10000 tip