Tue, 04 Dec 2018 17:03:43 +0000 rust: make clean takes care of rust/target
Georges Racinet <gracinet@anybox.fr> [Tue, 04 Dec 2018 17:03:43 +0000] rev 40833
rust: make clean takes care of rust/target This isn't the prettiest way of doing it, but it doesn't require looking up cargo, or wondering whether that should be part of setup.py clean. Differential Revision: https://phab.mercurial-scm.org/D5369
Tue, 04 Dec 2018 14:01:13 +0000 rust: rename local variables in AncestorsIterator::next
Georges Racinet <georges@racinet.fr> [Tue, 04 Dec 2018 14:01:13 +0000] rev 40832
rust: rename local variables in AncestorsIterator::next It was confusing to have p1 and parents.1 ; (p1, p2) is clearer. Differential Revision: https://phab.mercurial-scm.org/D5365
Tue, 04 Dec 2018 17:34:10 -0500 tests: stabilize test-inherit-mode.t on FreeBSD and macOS again
Augie Fackler <augie@google.com> [Tue, 04 Dec 2018 17:34:10 -0500] rev 40831
tests: stabilize test-inherit-mode.t on FreeBSD and macOS again This is the same fix as 90e26ef4cbb1, just repeated on the new file location. Differential Revision: https://phab.mercurial-scm.org/D5371
Tue, 04 Dec 2018 08:50:33 -0800 tests: drop redundant "#if execbit" from test-rebase-inmemory.t
Martin von Zweigbergk <martinvonz@google.com> [Tue, 04 Dec 2018 08:50:33 -0800] rev 40830
tests: drop redundant "#if execbit" from test-rebase-inmemory.t The whole file is already guarded by "#require execbit". Differential Revision: https://phab.mercurial-scm.org/D5366
Mon, 03 Dec 2018 16:56:09 -0800 narrow: don't resurrect old commits when narrowing (don't strip obsmarkers)
Martin von Zweigbergk <martinvonz@google.com> [Mon, 03 Dec 2018 16:56:09 -0800] rev 40829
narrow: don't resurrect old commits when narrowing (don't strip obsmarkers) If you have an old obsolescence-chain of commits that has been pruned and you narrow your repo so that some of those commits get stripped (because they affected the removed paths), then we would currently resurrect the commit that came before (along the obsmarker chain) the last stripped commit. That happens by the usual rules for obsmarker-stripping. However, it's quite surprising when it happens when you narrow your repo. This patch makes narrowing not strip obsmarkers. Differential Revision: https://phab.mercurial-scm.org/D5364
Sun, 18 Nov 2018 19:55:53 +0900 commandserver: turn server debug messages into logs
Yuya Nishihara <yuya@tcha.org> [Sun, 18 Nov 2018 19:55:53 +0900] rev 40828
commandserver: turn server debug messages into logs They were ui.debug() just because commandserver.log() was noop if no client connected.
Sun, 18 Nov 2018 19:47:04 +0900 commandserver: add config knob for various logging options
Yuya Nishihara <yuya@tcha.org> [Sun, 18 Nov 2018 19:47:04 +0900] rev 40827
commandserver: add config knob for various logging options The default rotating options are copied from the blackbox extension.
Sat, 10 Nov 2018 19:27:47 +0900 commandserver: expand log path for convenience
Yuya Nishihara <yuya@tcha.org> [Sat, 10 Nov 2018 19:27:47 +0900] rev 40826
commandserver: expand log path for convenience This allows us to set the log path relative to $XDG_RUNTIME_DIR, for instance. [cmdserver] log = $XDG_RUNTIME_DIR/chg/server.log
Sat, 10 Nov 2018 19:09:37 +0900 commandserver: switch logging facility to ui.log() interface
Yuya Nishihara <yuya@tcha.org> [Sat, 10 Nov 2018 19:09:37 +0900] rev 40825
commandserver: switch logging facility to ui.log() interface The "pager subcommand" message is removed since ui isn't accessible there. I think that's okay as cmdtable[cmd]() will call attachio() and some debug message will be printed.
Sat, 10 Nov 2018 19:00:17 +0900 commandserver: install logger to record server events through canonical API
Yuya Nishihara <yuya@tcha.org> [Sat, 10 Nov 2018 19:00:17 +0900] rev 40824
commandserver: install logger to record server events through canonical API The global commandserver.log() will be replaced with this.
(0) -30000 -10000 -3000 -1000 -300 -100 -10 +10 +100 +300 +1000 +3000 +10000 tip