Durham Goode <durham@fb.com> [Fri, 05 Feb 2016 13:23:24 -0800] rev 28000
copies: optimize forward copy detection logic for rebases
Forward copy detection (i.e. detecting what files have been moved/copied in
commit X since ancestor Y) previously required diff'ing the manifests of both X
and Y. This was expensive since it required reading both entire manifests and
doing a set difference (they weren't already in a set because of the
lazymanifest work). This cost almost 1 second on very large repositories, and
happens N times for a rebase of N commits.
This patch optimizes it for the case of rebase. In a rebase, we are comparing a
commit against it's immediate parent, and therefore we can know what files
changed by looking at ctx.files(). This lets us drastically decrease the size
of the set comparison, and makes it O(# of changes) instead of O(size of
manifest). This makes it take 1ms instead of 1000ms.
Yuya Nishihara <yuya@tcha.org> [Tue, 26 Jan 2016 23:05:19 +0900] rev 27999
tests: pass settings of hypothesis by with statement
given(..., settings=) is no longer available in Hypothesis 2.0.0.
https://github.com/DRMacIver/hypothesis/commit/7712c01
Yuya Nishihara <yuya@tcha.org> [Tue, 26 Jan 2016 22:44:29 +0900] rev 27998
tests: adjust for code move in Hypothesis 2.0.0
It appears that the Settings class was renamed to settings, and because
of this, the settings module was renamed to configuration.
https://github.com/DRMacIver/hypothesis/commit/a0e663b
Yuya Nishihara <yuya@tcha.org> [Tue, 26 Jan 2016 23:33:53 +0900] rev 27997
test-highlight: add normalization rule for Pygments 2.1
It appears that several classes have 'h', '1', '2' suffixes on Pygments 2.1,
which is the current version on Debian sid.
Yuya Nishihara <yuya@tcha.org> [Tue, 26 Jan 2016 23:27:12 +0900] rev 27996
test-highlight: factor out function that normalizes pygments output
Yuya Nishihara <yuya@tcha.org> [Tue, 26 Jan 2016 23:26:05 +0900] rev 27995
test-highlight: unify normalization rule of pygments output
We had two variants for unknown reason, s/mf/mi/g and s/mi/mf/g. This patch
unifies them to s/mf/mi/g so that we can introduce a utility function.
FUJIWARA Katsunori <foozy@lares.dti.ne.jp> [Fri, 05 Feb 2016 22:32:05 +0900] rev 27994
tests: remove protected file forcibly for portability
Before this patch, test-command-template.t is timed out on Solaris,
because "rm" on permission denied file implies prompting "override
protection 0 (yes/no)?" and blocks execution of test script.
FUJIWARA Katsunori <foozy@lares.dti.ne.jp> [Fri, 05 Feb 2016 22:32:05 +0900] rev 27993
tests: dump journal file by python script instead of sed for portability
Before this patch, test-repair-strip.t fails on Solaris, because of
"sed" on it doesn't work as expected if input contains null ('\0')
character.
FUJIWARA Katsunori <foozy@lares.dti.ne.jp> [Fri, 05 Feb 2016 22:32:05 +0900] rev 27992
tests: execute check-config.py without xargs
Before this patch, test-check-config.t fails on Solaris, because
"xargs" doesn't invoke check-config.py with all filenames at once.
"xargs" may invoke specified command multiple times with part of
arguments given from stdin: according to "xargs(1)" man page, this
dividing arguments is system-dependent.
For portability of test-check-config.t, this patch adds "xargs" like
mode to check-config.py and executes it in test-check-config.t without
"xargs".
FUJIWARA Katsunori <foozy@lares.dti.ne.jp> [Fri, 05 Feb 2016 22:32:05 +0900] rev 27991
tests: avoid nested quoting on command line for portability
Before this patch, test-revset.t fails on Solaris using ksh as default
"sh", because nested quoting below isn't acceptable for it.
+-----------------------------+ inner quoting
"`python -c "print '|'.join(['0:1'] * 500)"`"
+-------------------------------------------+ outer quoting
This patch does below for portability.
- omit outer quoting
This should be safe, because generated string contains no white
space character.
- use '+' instead of '|' (for safety)
'|' has special meaning for many shell, but '+' isn't (at least,
for ordinary ones).
liscju <piotr.listkiewicz@gmail.com> [Fri, 05 Feb 2016 13:20:23 +0100] rev 27990
version: verbose list internal and external extension source (
issue4731)
Martin von Zweigbergk <martinvonz@google.com> [Tue, 02 Feb 2016 21:20:04 -0800] rev 27989
check-code: allow "grep pattern filename-containing-dash-a"
We're trying to forbid "grep -a" and unintentionally complained even
if the "-a" was part of the filename. Requiring a space before "-a" to
match is probably good enough.
timeless <timeless@mozdev.org> [Fri, 05 Feb 2016 01:56:46 +0000] rev 27988
rebase: update working directory when aborting (
issue5084)
Yuya Nishihara <yuya@tcha.org> [Tue, 02 Feb 2016 23:49:49 +0900] rev 27987
revset: flatten chained 'list' operations (aka function args) (
issue5072)
Internal _matchfiles() function can take bunch of arguments, which would
lead to a maximum recursion depth error. This patch avoids the excessive
stack use by flattening 'list' nodes beforehand.
Since getlist() no longer takes a nested 'list' nodes, _parsealiasdecl()
also needs to flatten argument list, "aliasname($1, $2, ...)".
André Sintzoff <andre.sintzoff@gmail.com> [Fri, 05 Feb 2016 19:40:57 +0100] rev 27986
help: fix typo in backgroundclose documentation
liscju <piotr.listkiewicz@gmail.com> [Fri, 05 Feb 2016 15:18:40 +0100] rev 27985
revert: makes interactive mode ask to forget added files (
issue4936)
Before this patch revert interactive mode unconditionally forgets
added files. This patch fixes this by asking user if he wants
to forget added file. If user doesn't want to forget given file,
it is added to matcher_opts exclude list, to not reviewing it
later with other modified files.
Durham Goode <durham@fb.com> [Thu, 04 Feb 2016 15:38:04 -0800] rev 27984
obsolete: fix n^2 marker computation behavior
Previously, if you ran obsolete.createmarkers with a bunch of markers that did
not have successors (like when you do a prune), it encountered a n^2 computation
behavior because the loop would read the changelog (to get ctx.parents()), then
add a marker, in a loop. Adding a marker invalidated the computehidden cache,
and reading the changelog recomputed it.
This resulted in pruning 150 commits taking 150+ seconds in a large repo.
The fix is to break the reading part of the loop to be separate from the writing
part.
Durham Goode <durham@fb.com> [Wed, 03 Feb 2016 17:44:11 -0800] rev 27983
memctx: fix memctx manifest file hashes
When memctx is asked for a manifest, it constructs one by merging the p1
manifest, and the changes that are on top. For the changes on top, it was
previously using p1.node() as the file entries parent, which actually returns
the commit node that the p1 linkrev points at! Which is entirely incorrect.
The fix is to use p1.filenode() instead, which returns the parent file node as
desired.
I don't know how to execute this or make it have a visible effect, so I'm not
sure how to test it. It was noticed because asking for the linkrev is an
expensive operation when using the remotefilelog extension and this was causing
performance regressions with commit.
Jason R. Coombs <jaraco@jaraco.com> [Fri, 29 Jan 2016 14:53:44 -0500] rev 27982
schemes: add debugexpandscheme command, resolving a scheme to canonical form
Jason R. Coombs <jaraco@jaraco.com> [Fri, 29 Jan 2016 14:40:41 -0500] rev 27981
schemes: extract scheme expansion as its own method on ShortRepository
Matt Mackall <mpm@selenic.com> [Wed, 03 Feb 2016 16:24:24 -0600] rev 27980
merge with stable
Matt Mackall <mpm@selenic.com> [Wed, 03 Feb 2016 16:00:00 -0600] rev 27979
Added signature for changeset
b698abf971e7
Matt Mackall <mpm@selenic.com> [Wed, 03 Feb 2016 15:59:59 -0600] rev 27978
Added tag 3.7.1 for changeset
b698abf971e7
Siddharth Agarwal <sid0@fb.com> [Wed, 03 Feb 2016 09:23:31 -0800] rev 27977
rebase: don't preserve most extra fields
This backs out changeset
88fde8db5307.
See the previous patches for why.