Yuya Nishihara <yuya@tcha.org> [Fri, 18 May 2018 21:32:05 +0900] rev 38039
hgweb: do not try to replace signal handlers while locking
According to the issue 5889, mod_wsgi issues a warning on signal.signal()
call, and we wouldn't want to see it in error log. The problem addressed
by
d77c3b023393 could potentially occur in web session, but that would be
less likely than in user processes.
Yuya Nishihara <yuya@tcha.org> [Fri, 18 May 2018 21:24:06 +0900] rev 38038
lock: add internal config to not replace signal handlers while locking
signal.signal() is blocked in some WSGI environments, and a horrible warning
is sent to the server log. So we need a way to disable it, and I think
abusing ui.config is the simplest workaround.
Augie Fackler <augie@google.com> [Tue, 22 May 2018 21:51:20 -0400] rev 38037
merge with i18n
Wagner Bruna <wbruna@softwareexpress.com.br> [Fri, 04 May 2018 18:55:57 -0300] rev 38036
i18n-pt_BR: synchronized with
32a75a8a5b0f
Wagner Bruna <wbruna@softwareexpress.com.br> [Fri, 04 May 2018 18:55:29 -0300] rev 38035
i18n-ja: fix block indentation
FUJIWARA Katsunori <foozy@lares.dti.ne.jp> [Tue, 01 May 2018 18:22:52 +0900] rev 38034
i18n-ja: synchronized with
32a75a8a5b0f
Boris Feld <boris.feld@octobus.net> [Mon, 21 May 2018 15:14:46 +0200] rev 38033
httppeer: declare 'dbg' at the function level
As we just saw in the previous changeset, having the variable defined into a
branch creates bug. This is a cheap to move it at the function level.
Boris Feld <boris.feld@octobus.net> [Fri, 04 May 2018 19:06:46 +0200] rev 38032
httppeer: properly gate debug usage behind debug flag check
The "dbg" local variable is only defined if the 'debugflag' is set to True.
However, it was used indiscriminately later in the function. We hide its usage
behind the 'debugflag' value to avoid raising a NameError.
Yuya Nishihara <yuya@tcha.org> [Sun, 13 May 2018 11:09:53 +0900] rev 38031
test-http-branchmap: fix stdio mode on Windows
Yuya Nishihara <yuya@tcha.org> [Sun, 13 May 2018 11:08:35 +0900] rev 38030
test-http-branchmap: fix encoding test to wrap the server stream
Wrapping sys.std* streams has no effect since
39d13b8c101d, "py3: bulk
replace sys.stdin/out/err by util's."
Yuya Nishihara <yuya@tcha.org> [Thu, 17 May 2018 21:55:00 +0900] rev 38029
test-merge-tools: create repo directory to free $TESTTMP for temporary files
Joerg Sonnenberger <joerg@bec.de> [Sat, 12 May 2018 23:44:08 +0200] rev 38028
notify: add option to include function names in the diff output
This is a localized version of diff.showfunc.
Differential Revision: https://phab.mercurial-scm.org/D3553
Martin von Zweigbergk <martinvonz@google.com> [Wed, 16 May 2018 10:34:31 -0700] rev 38027
tests: update py3 test since json fields are now sorted
The fields have been sorted since
814151cd8c4a (logcmdutil: rewrite
jsonchangeset printer to be backed by jsonformatter, 2018-04-13), and
the test case has thus been failing since that commit.
Differential Revision: https://phab.mercurial-scm.org/D3570
Kyle Lippincott <spectral@google.com> [Wed, 16 May 2018 15:48:12 -0700] rev 38026
tests: test-parseindex.t works just fine with chg
In
538e850ae737 I annotated tests that failed with chg with "#require no-chg",
but did not investigate why the tests were failing. This test was failing for
unrelated reasons (I forgot to rebuild hg) and so should not be marked as
no-chg.
Test Plan:
make local
pushd contrib/chg
make clean; make
popd
cd tests
for f in '' --with-chg=../contrib/chg/chg; do
./run-tests.py --local $f test-parseindex.t
done
Differential Revision: https://phab.mercurial-scm.org/D3571
Pulkit Goyal <7895pulkit@gmail.com> [Fri, 27 Apr 2018 21:55:02 +0530] rev 38025
py3: convert the report to bytes
Differential Revision: https://phab.mercurial-scm.org/D3569
Martin von Zweigbergk <martinvonz@google.com> [Fri, 11 May 2018 10:36:28 -0700] rev 38024
json: reject unicode on py2 as well
This makes it consistent with the behavior on py3.
Differential Revision: https://phab.mercurial-scm.org/D3536
Yuya Nishihara <yuya@tcha.org> [Wed, 16 May 2018 20:22:23 +0900] rev 38023
dispatch: mask negative exit code recorded in blackbox log
That's what we do for the exit code delivered to the environment.
Yuya Nishihara <yuya@tcha.org> [Wed, 16 May 2018 20:17:50 +0900] rev 38022
dispatch: fix exit code of unhandled exception recorded in blackbox log
Spotted by Martin von Zweigbergk.
We might want to change the exit code to -1 (i.e. 255) because 1 means
non-abort error in hg, but that's another issue.
Kyle Lippincott <spectral@google.com> [Tue, 15 May 2018 11:52:43 -0700] rev 38021
tests: mark tests that fail when using chg as #require no-chg
As far as I can tell, most of these failures are due to using $HGPORT, which it
seems chg might be using itself? I don't know enough to debug these failures to
fix them properly.
Differential Revision: https://phab.mercurial-scm.org/D3562
Pulkit Goyal <7895pulkit@gmail.com> [Fri, 27 Apr 2018 21:54:37 +0530] rev 38020
py3: use pycompat.bytestr() instead of str() in extensions.py
Differential Revision: https://phab.mercurial-scm.org/D3568
Pulkit Goyal <7895pulkit@gmail.com> [Wed, 16 May 2018 07:58:02 +0530] rev 38019
py3: use pycompat.bytestr() on bytes before %r-ing it
Differential Revision: https://phab.mercurial-scm.org/D3567
Pulkit Goyal <7895pulkit@gmail.com> [Thu, 15 Feb 2018 17:15:21 +0530] rev 38018
py3: add b'' prefixes in tests/test-extension.t
# skip-blame because just b'' prefixes
Differential Revision: https://phab.mercurial-scm.org/D3564
Pulkit Goyal <7895pulkit@gmail.com> [Fri, 27 Apr 2018 20:55:10 +0530] rev 38017
py3: suppress the value returned by .write() calls
These return values are not present on Python 2.
Differential Revision: https://phab.mercurial-scm.org/D3563
Martin von Zweigbergk <martinvonz@google.com> [Tue, 15 May 2018 09:57:58 -0700] rev 38016
tests: test failure reporting in blackbox code
Differential Revision: https://phab.mercurial-scm.org/D3560
David Demelier <markand@malikania.fr> [Mon, 14 May 2018 12:53:13 +0200] rev 38015
export: add -B option to select a bookmark
Just like `hg email -B`, `hg strip -B`, supports -B in export to select a list
of changesets reachable from a bookmark.
Gregory Szorc <gregory.szorc@gmail.com> [Sat, 12 May 2018 18:05:50 -0700] rev 38014
packaging: move most packaging targets to own Makefile
Since we're putting everything packaging under one roof, let's
define the make targets related to packaging there as well.
I didn't move the "osx" target because it is non-trivial.
Some targets did change slightly as part of the move. But it
was mostly around path normalization.
Differential Revision: https://phab.mercurial-scm.org/D3552
Gregory Szorc <gregory.szorc@gmail.com> [Sat, 12 May 2018 12:27:51 -0700] rev 38013
packaging: make packaging scripts less reliant on pwd
The scripts currently assume they are executed from the repo root.
This feels like an arbitrary restriction. Let's fix that.
Differential Revision: https://phab.mercurial-scm.org/D3551
Gregory Szorc <gregory.szorc@gmail.com> [Sat, 12 May 2018 10:57:04 -0700] rev 38012
packaging: move contrib/macosx to contrib/packaging/
And update various references throughout the code.
Differential Revision: https://phab.mercurial-scm.org/D3550