Nicolas Dumazet <nicdumz.commits@gmail.com> [Fri, 11 Dec 2009 15:58:09 +0900] rev 10090
inotify: do not rely on stat(.hg/dirstate) to invalidate our dirstate
stat() is not reliable when several events happen quickly. Which means
that if two hg actions occur in the same second, stat() result will not
reflect the second change. And only _one_ invalidate() call was done.
Also ignore the events that occur when wlock is held, since wlock release
will trigger a full rescan anyway.
Fixes 17 run-tests.py --inotify tests.
Nicolas Dumazet <nicdumz.commits@gmail.com> [Fri, 27 Nov 2009 09:23:10 +0900] rev 10089
inotify: follow new files if they are regular or a symlink.
Fixes run-tests.py --inotify symlink-basic
Mads Kiilerich <mads@kiilerich.com> [Wed, 16 Dec 2009 12:09:25 +0100] rev 10088
run-tests.py: Show paths to failing tests, .err and .out
I think this is very convenient when investigating and fixing test failures,
but perhaps it must be tweaked further to show paths if and only if
it is appropriate.
Matt Mackall <mpm@selenic.com> [Wed, 16 Dec 2009 16:46:12 -0600] rev 10087
Merge with stable
Matt Mackall <mpm@selenic.com> [Wed, 16 Dec 2009 16:42:58 -0600] rev 10086
convert: make hg sink cleanup safer
Martin Geisler <mg@lazybytes.net> [Wed, 16 Dec 2009 22:58:57 +0100] rev 10085
Merge with stable
Nicolas Dumazet <nicdumz.commits@gmail.com> [Fri, 11 Dec 2009 15:25:33 +0900] rev 10084
graphlog: fix output when both a limit and a path are provided
Limit was interpreted as absolute, from the topmost revision, without
counting the number of revisions matching a given file.
Which caused "glog -lN file" to show sometimes less than N csets if
the file was not modified in all of the N previous csets.
glog will now match the behavior of log.
Patrick Mezard <pmezard@gmail.com> [Wed, 16 Dec 2009 22:18:15 +0100] rev 10083
mercurial.iss: update locale location
Mads Kiilerich <mads@kiilerich.com> [Wed, 16 Dec 2009 12:10:21 +0100] rev 10082
test-share: Solaris sed can't handle never-ending lines
Mads Kiilerich <mads@kiilerich.com> [Wed, 16 Dec 2009 12:10:21 +0100] rev 10081
test-rollback: Ignore hook exit code
Failing hooks returned 255 on Solaris
Patrick Mezard <pmezard@gmail.com> [Wed, 16 Dec 2009 13:19:42 +0100] rev 10080
Merge with crew-stable
Mads Kiilerich <mads@kiilerich.com> [Wed, 16 Dec 2009 12:10:21 +0100] rev 10079
test-hook: Don't use extended regular expressions with grep
Using egrep fixes failure on Solaris
Brendan Cully <brendan@kublai.com> [Tue, 15 Dec 2009 12:33:04 -0800] rev 10078
hgweb: Make get_mtime use repository to find store path.
It was calculating it directly, which is redundant and caused
it to break with shared repositories.
Ronny Pfannschmidt <Ronny.Pfannschmidt@gmx.de> [Sun, 13 Dec 2009 18:29:16 +0100] rev 10077
make ui.interactive() return false in case stdin lacks isatty
Matt Mackall <mpm@selenic.com> [Tue, 15 Dec 2009 10:45:24 -0600] rev 10076
Merge with stable
Matt Mackall <mpm@selenic.com> [Tue, 15 Dec 2009 10:44:38 -0600] rev 10075
Merge with crew
Matt Mackall <mpm@selenic.com> [Tue, 15 Dec 2009 10:44:10 -0600] rev 10074
Merge -stable heads
Martin Geisler <mg@lazybytes.net> [Mon, 14 Dec 2009 23:14:50 +0100] rev 10073
Makefile: update "make clean" target after f91e5630ce7e
Patrick Mezard <pmezard@gmail.com> [Mon, 14 Dec 2009 22:15:46 +0100] rev 10072
Merge with crew-stable
Patrick Mezard <pmezard@gmail.com> [Mon, 14 Dec 2009 22:09:49 +0100] rev 10071
convert/svn: better handling of hg recursive call failure
Alexander Solovyov <piranha@piranha.org.ua> [Mon, 14 Dec 2009 21:21:34 +0200] rev 10070
schemes: url parts are counted from 1
Matt Mackall <mpm@selenic.com> [Mon, 14 Dec 2009 15:02:02 -0600] rev 10069
Merge with -stable
Matt Mackall <mpm@selenic.com> [Mon, 14 Dec 2009 15:01:13 -0600] rev 10068
subrepo: add default path to new clones
Martin Geisler <mg@lazybytes.net> [Mon, 14 Dec 2009 00:26:28 +0100] rev 10067
win32mbcs: fix typos and reST syntax
Martin Geisler <mg@lazybytes.net> [Mon, 14 Dec 2009 00:02:04 +0100] rev 10066
help/templates: use hanging indent for field list
Now that the field list indentation is computed automatically, we no
longer need to align all fields the same. Using a simple hanging
indent makes it easy to edit the text in an editor like Emacs, which
will automatically recognized the hang and keep it.
The fields are still output with no blank lines between them.
Martin Geisler <mg@lazybytes.net> [Sun, 13 Dec 2009 23:49:53 +0100] rev 10065
minirst: improve layout of field lists
Before, we used the padding following the key to compute where to wrap
the text. Long keys would thus give a big indentation. It also
required careful alignment of the source text, making it cumbersome to
items to the list.
We now compute the maximum key length and use that for all items in
the list. We also put a cap on the indentation: keys longer than 10
characters are put on their own line. This is similar to how rst2html
handles large keys: it uses 14 as the cutoff point, but I felt that 10
was better for monospaced text in the console.
Martin Geisler <mg@lazybytes.net> [Sun, 13 Dec 2009 22:37:30 +0100] rev 10064
minirst: don't test regexps twice
We know the regexps match since splitparagraphs used them too.
Martin Geisler <mg@lazybytes.net> [Sun, 13 Dec 2009 19:24:24 +0100] rev 10063
minirst: run inlineliterals too in debug mode
Patrick Mezard <pmezard@gmail.com> [Sun, 13 Dec 2009 18:54:54 +0100] rev 10062
test-share: handle sed/GNU-sed discrepancies
Patrick Mezard <pmezard@gmail.com> [Sun, 13 Dec 2009 18:06:24 +0100] rev 10061
Make {file_copies} usable as a --template key
Before this, to display file copies in templates, you had to write something
like {file_copies%filecopy}. For some reason, the {file_copy} subtemplate was
used by default but not defined by default in changeset_templater, while styles
were already using it. Here we define {file_copy} in changeset_templater, and
change the templater to handle formatting strings like {file_copies%filecopy}
with already expanded keys (in this case {file_copies}), for backward
compatibility.