Matt Harbison <matt_harbison@yahoo.com> [Wed, 16 Dec 2015 13:33:43 -0500] rev 27436
windows: correct the import of win32
This module is relative, and was overlooked when converting to absolute_import
in 6daa795ed32f.
Pascal Quantin <pascal.quantin@gmail.com> [Thu, 17 Dec 2015 19:33:44 +0100] rev 27435
win32: add internals help topics to Inno Setup installer
Augie Fackler <augie@google.com> [Fri, 18 Dec 2015 09:57:35 -0500] rev 27434
changegroups: add documentation for cg3
Mike Edgar <adgar@google.com> [Mon, 14 Dec 2015 15:55:12 -0500] rev 27433
changegroup: add flags field to cg3 delta header
This lets revlog flags be transmitted over the wire. Right now this is
useful for censored nodes and for narrowhg's ellipsis nodes.
Augie Fackler <augie@google.com> [Fri, 11 Dec 2015 11:23:49 -0500] rev 27432
changegroup: introduce cg3, which has support for exchanging treemanifests
I'm not entirely happy with using a trailing / on a "file" entry for
transferring a treemanifest. We've discussed putting some flags on
each file header[0], but I'm unconvinced that's actually any better:
if we were going to add another feature to the cg format we'd still be
doing a version bump anyway to cg4, so I'm inclined to not spend time
coming up with a more sophisticated format until we actually know what
the next feature we want to stuff in a changegroup will be.
Test changes outside test-treemanifest.t are only due to the new CG3
bundlecap showing up in the wire protocol.
Many thanks to adgar@google.com and martinvonz@google.com for helping
me with various odd corners of the changegroup and treemanifest API.
0: It's not hard refactoring, nor is it a lot of work. I'm just
disinclined to do speculative work when it's not clear what the
customer would actually be.
Matt Mackall <mpm@selenic.com> [Fri, 18 Dec 2015 14:40:11 -0600] rev 27431
merge with stable
Gregory Szorc <gregory.szorc@gmail.com> [Thu, 17 Dec 2015 17:16:02 -0800] rev 27430
revlog: seek to end of file before writing (issue4943)
Revlogs were recently refactored to open file handles in "a+" and use a
persistent file handle for reading and writing. This drastically
reduced the number of file handles being opened.
Unfortunately, it appears that some versions of Solaris lose the file
offset when performing a write after the handle has been seeked.
The simplest workaround is to seek to EOF on files opened in a+ mode
before writing to them, which is what this patch does.
Ideally, this code would exist in the vfs layer. However, this would
require creating a proxy class for file objects in order to provide a
custom implementation of write(). This would add overhead. Since
revlogs are the only files we open in a+ mode, the one-off workaround
in revlog.py should be sufficient.
This patch appears to have little to no impact on performance on my
Linux machine.
Augie Fackler <augie@google.com> [Mon, 14 Dec 2015 20:57:21 -0500] rev 27429
manifestmerge: have manifest do matching before diffing
This means that the diff code does less work, potentially
significantly less in the case of treemanifests. It also should ease
implementation with narrowed clone cases (such as narrowhg) when we
don't always have the entire set of treemanifest revlogs locally.
As far as I can tell, this codepath is currently only used by record,
so it'll probably die in the near future, and then narrowhg won't have
to worry about composing with some unknown matching system.
timeless <timeless@mozdev.org> [Thu, 17 Dec 2015 14:57:20 +0000] rev 27428
branch: reword help text
We're not necessarily talking about *the* active branch,
just any old branch.
timeless <timeless@mozdev.org> [Thu, 17 Dec 2015 14:56:32 +0000] rev 27427
archive: adjust help text
timeless <timeless@mozdev.org> [Thu, 17 Dec 2015 14:54:47 +0000] rev 27426
annotate: add missing period to help
timeless <timeless@mozdev.org> [Thu, 17 Dec 2015 14:54:20 +0000] rev 27425
addremove: make help match add
timeless <timeless@mozdev.org> [Thu, 17 Dec 2015 14:53:40 +0000] rev 27424
add: mention .hgignore in help
timeless <timeless@mozdev.org> [Thu, 17 Dec 2015 15:05:25 +0000] rev 27423
bundle: warn for --base with --all
timeless <timeless@mozdev.org> [Thu, 17 Dec 2015 15:03:45 +0000] rev 27422
bundle: fix error for --all with destination
Before it complained about --base
timeless <timeless@mozdev.org> [Thu, 17 Dec 2015 14:59:11 +0000] rev 27421
bundle: fix grammar in help text
timeless <timeless@mozdev.org> [Thu, 17 Dec 2015 14:58:52 +0000] rev 27420
bundle: clarify help text
The file might not be compressed; the interactions between
-a, --base, and a named or default repository weren't clear.
Yuya Nishihara <yuya@tcha.org> [Sun, 13 Dec 2015 20:07:19 +0900] rev 27419
cmdutil: do not duplicate stdout by makefileobj()
It made output order unpredictable because two separate buffers are flushed
individually. Let's use a thin wrapper that just sends close() to black hole.
Yuya Nishihara <yuya@tcha.org> [Sun, 13 Dec 2015 20:01:11 +0900] rev 27418
cmdutil: reimplement file wrapper that disables close()
There's no need to dynamically create wrappedfileobj class and define
close() as lambda. Also, __iter__() was missing.
Yuya Nishihara <yuya@tcha.org> [Sun, 13 Dec 2015 19:57:54 +0900] rev 27417
export: remove useless comparison between file object and string
It was introduced at 31aa2e5b0750, where "template" argument could be a file
object. After that, a306837f8c87 added "len(template)", so "template" must be
a string now. Therefore, "fp != template" should always be True.
It seems 31aa2e5b0750 was intended to work around a bug in TortoiseHg, and
I'm sure I've fixed it completely in TortoiseHg source.
https://selenic.com/pipermail/mercurial-devel/2011-February/028467.html
Yuya Nishihara <yuya@tcha.org> [Sun, 13 Dec 2015 19:47:46 +0900] rev 27416
export: do not print '<fdopen>' as an output filename
Because makefileobj() duplicates or wraps stdout, "fp != sys.stdout" didn't
work correctly. Python doc states that special file objects are named in the
form '<...>', and absolute filenames should never start with '<', we can
ignore names start with '<'. We can't test fp.fileno() because fp may be a
command-server channel.
https://docs.python.org/2.7/library/stdtypes.html#file.name
In the test output, "exporting patch:" line is printed after patch content.
This is caused by fdopen() and will be fixed by the subsequent patch.
Yuya Nishihara <yuya@tcha.org> [Sun, 13 Dec 2015 19:32:01 +0900] rev 27415
commandserver: implement name() to clarify channel is not a plain file
Because unknown attributes are delegated to the underlying file object,
commandserver channels said they were '<stdout>' or '<stdin>' even though
they weren't. This patch makes them say '<X-channel>'.
Mateusz Kwapich <mitrandir@fb.com> [Tue, 15 Dec 2015 13:27:09 -0800] rev 27414
histedit: delete to drop
The default behaviour to forbid this makes a lot of sense for novice users
because it's safeguarding them from dangerous behavior but making it
configurable will be apprieciated by power users in at least one big
organization.
It allows an user to look an histedit rules from declarative perspective and
make the rules reflect the state after histedit. If we can move lines t move
commits why can't we drop lines to drop commits?
Let's put this behind config knob and inform users about this feature the very
moment they are trying to use it so they can choose desired behaviour.
Mathias De Maré <mathias.demare@gmail.com> [Wed, 16 Dec 2015 12:33:54 +0100] rev 27413
clonebundles: fix bundlespec typo
timeless <timeless@mozdev.org> [Thu, 17 Dec 2015 20:13:29 +0000] rev 27412
record: fix hunk handling to remember the current function
timeless <timeless@mozdev.org> [Thu, 17 Dec 2015 14:38:22 +0000] rev 27411
record: turn on showfunc
Always try to give diff context when doing an interactive record
Bryan O'Sullivan <bos@serpentine.com> [Thu, 17 Dec 2015 13:07:34 -0800] rev 27410
parsers: use PyTuple_Pack instead of manual list-filling
Suggested by Yuya.
Thu Trang Pham <thu@fb.com> [Thu, 17 Dec 2015 15:23:36 -0800] rev 27409
tags: mention --quiet switch in help (issue4920)
Matt Mackall <mpm@selenic.com> [Thu, 17 Dec 2015 17:27:32 -0600] rev 27408
merge with stable
timeless <timeless@mozdev.org> [Mon, 14 Dec 2015 23:04:17 +0000] rev 27407
histedit: omit useless message from update (edit)
specifically:
0 files updated, 0 files merged, 0 files removed, 0 files unresolved
timeless <timeless@mozdev.org> [Mon, 14 Dec 2015 22:37:31 +0000] rev 27406
histedit: omit useless message from update (_histedit)
specifically:
0 files updated, 0 files merged, 0 files removed, 0 files unresolved
timeless <timeless@mozdev.org> [Mon, 14 Dec 2015 21:43:16 +0000] rev 27405
histedit: omit useless message from update (histeditaction)
specifically:
0 files updated, 0 files merged, 0 files removed, 0 files unresolved
timeless <timeless@mozdev.org> [Mon, 14 Dec 2015 23:14:06 +0000] rev 27404
update: add quietempty flag to _showstats
if called with quietempty=True, suppress:
0 files updated, 0 files merged, 0 files removed, 0 files unresolved
timeless <timeless@mozdev.org> [Mon, 14 Dec 2015 22:08:14 +0000] rev 27403
histedit: omit useless message from abort
specifically:
0 files updated, 0 files merged, 0 files removed, 0 files unresolved
timeless <timeless@mozdev.org> [Mon, 14 Dec 2015 23:13:25 +0000] rev 27402
hg: add quietempty flag to _showstats
if called with quietempty=True, suppress:
0 files updated, 0 files merged, 0 files removed, 0 files unresolved
Mateusz Kwapich <mitrandir@fb.com> [Thu, 17 Dec 2015 11:00:06 -0800] rev 27401
patch: disable nobinary when HGPLAIN=1
The diff output without binaries is definitely great for interactive users - a
binary patch is not meaningful for them. Although setting diff.nobinary flag
can break the automation. Let's force full output for automation.
Yuya Nishihara <yuya@tcha.org> [Thu, 17 Dec 2015 22:29:41 +0900] rev 27400
help: add missed last new line to "internals" topic
Caught by test-gendoc.t.
Matt Mackall <mpm@selenic.com> [Wed, 16 Dec 2015 20:58:26 -0600] rev 27399
dirstate: make delaywrite sleep until the next multiple of n seconds
Rather than sleep for 2 seconds, we sleep until the next even-numbered
second, which has the same effect, but makes tests faster. This
removes test-largefiles-update as the long pole of the test suite.
Matt Mackall <mpm@selenic.com> [Wed, 16 Dec 2015 20:49:18 -0600] rev 27398
dirstate: only invoke delaywrite if relevant
This avoids a significant amount of sleeping in some of our longest
tests.
Matt Mackall <mpm@selenic.com> [Wed, 16 Dec 2015 20:46:53 -0600] rev 27397
dirstate: move delaywrite logic from write to _write
This will allow us to be smarter in upcoming patches.
Matt Mackall <mpm@selenic.com> [Mon, 07 Dec 2015 16:16:06 -0600] rev 27396
run-tests: show scheduling with --showchannels
This gives one line of output per second with one column per -j level
that allows analyzing test scheduling problems. First 24 seconds of
output at -j 30 looks like this:
0 .
1 = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = s.
2 c c o c r l g r s s = c p = c h c a h c g c h c b c c l l c ss
3 h o b o e a e u u u c o a h o e o c g o l h g h u o = a o = s
4 e n s n b r n n b b m t g n l n l w n o e w e n n e r g i .
5 c t o = a g d - r r = m c w v p v . e v g c e c d v x g . m
6 k r l r s e o t e e b a h e e . e . b e . k b k l e t e . p
7 - i e e e f c e p p u n b b r . r . - r . - - - e r e f . o .
8 p b t v - i . s o o n d o d t . t . c t . c s = 2 t n i . r
9 y - e s c l . t - . d - m i - . - . o - . o y r - - s l . t
10 3 p - e h e . s s . l t b r s . s . m s . d m e f s i e . .
11 - e c t e s . . v . e e . . v . v . m v . e r n o v o s . .
12 c r h . c - . . n . 2 m . . n . n . a n . . e a r n n . . .
13 o f e . k u . . . . - p . . - . - . n - . . v m m - . . . .
14 m . c . - p . . . . e l . . s . m . d s . . . e a e . . . .
15 p . k . r d . . . . x a . . i . o . s o . . . - t n . . . .
16 a . h . e a . . . . c t . . n . v . . u . . . m . c . . . .
17 t . e . s t . . . . h e . . k . e . . r . . . e . o . . . .
18 . . a . t e . . . . a . . . . . . . . c . . . r . d . . . .
19 . . d . o . . . . . n . . . . . . . . e . . . g . i . . . .
20 . . s . r . . . . . g . . . . . . . . . . . . e . n . . . .
21 . . . . e . . . . . e . . . . . . . . . . . . 2 . g . . . .
22 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
23 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
24 . . . . . . . . . . . . . . . . . . . . . . . . . = . . . . ^C
Test names read off vertically, beginning with '='. Idle time (not
shown) appears as blank space.
Matt Mackall <mpm@selenic.com> [Sun, 06 Dec 2015 15:14:01 -0600] rev 27395
tests: avoid duplicate install steps in test-run-tests
At several seconds each, this is significantly slowing down the test.
Matt Mackall <mpm@selenic.com> [Fri, 04 Dec 2015 17:05:20 -0600] rev 27394
run-tests: add more scheduling weight hints
The scheduler would like to order test execution by expected run-time,
but doesn't know much about how long a test will run. It thus uses
test size as a proxy for run-time. By tweaking these weights we can
keep CPUs more evenly busy and thus finish sooner.
In particular, this change pushes the three currently longest-running
tests closer to the beginning:
test-largefiles-update.t
test-run-tests.t
test-gendoc.t
As the largefiles test is currently the long pole of the test suite
with higher -j factors, the sooner it's started, the sooner the tests
can end.
We also up the weight on some shorter but long-running tests that
could have previously delayed completion with low -j factors by
running very close to the end.
Matt Mackall <mpm@selenic.com> [Fri, 04 Dec 2015 14:55:10 -0600] rev 27393
run-tests: report timeouts in a less alarming fashion
Rather than report timed-out tests like this:
ERROR: test-convert-svn-sink.t output changed
!
..simply put a 't' rather than a '.' in the stream.
Matt Mackall <mpm@selenic.com> [Mon, 30 Nov 2015 13:47:29 -0600] rev 27392
ui: try to handle $$ more robustly in prompts (issue4970)
Matt Mackall <mpm@selenic.com> [Wed, 16 Dec 2015 17:40:01 -0600] rev 27391
merge with stable
timeless <timeless@mozdev.org> [Tue, 15 Dec 2015 07:57:04 +0000] rev 27390
import: reorder help text
Try to place key concepts early+together.
timeless <timeless@mozdev.org> [Tue, 15 Dec 2015 07:56:03 +0000] rev 27389
import: add word to help text
timeless <timeless@mozdev.org> [Tue, 15 Dec 2015 07:54:01 +0000] rev 27388
import: refactor exact flag
timeless <timeless@mozdev.org> [Mon, 14 Dec 2015 05:29:55 +0000] rev 27387
help: filter extension commands
timeless <timeless@mozdev.org> [Mon, 14 Dec 2015 06:00:32 +0000] rev 27386
test-help: tighten grep patterns
Help should output section headings, but no debug commands
Laurent Charignon <lcharignon@fb.com> [Mon, 14 Dec 2015 11:19:48 -0800] rev 27385
summary: add troubles list to the output of hg summary
This patch adds troubles information to the output of hg summary.
Example line displayed in hg summary:
unstable: 1 changeset
Matt Mackall <mpm@selenic.com> [Wed, 16 Dec 2015 17:17:12 -0600] rev 27384
wix: add missing template
Matt Harbison <matt_harbison@yahoo.com> [Sun, 13 Sep 2015 22:54:51 -0400] rev 27383
tests: add coverage to ensure Wix tracks 'help' and 'templates' files
This would have caught the problem fixed by 65d2538ac993. There are
other *.wxs files that can be checked, but they appear to be more
complicated. For example, locale.wxs has what appears to be foreach
loop support, as well as variable substitution.
By checking `hg files` to determine tracked file, this is able to avoid false
failures when other junk is present in the filesystem, like *.orig files.
I can't tell if the map-cmdline.status file is not included on purpose, but I
don't see the purpose of excluding it. The missing help files seem reasonable
for Windows.
Matt Harbison <matt_harbison@yahoo.com> [Sun, 13 Dec 2015 18:13:44 -0500] rev 27382
perf: adjust perfstartup() for Windows
The /dev/null redirect was causing the following error:
The system cannot find the path specified.
Adjusting HGRCPATH as part of the command line causes the system to try to
execute 'HGRCPATH'.
Matt Harbison <matt_harbison@yahoo.com> [Sun, 13 Dec 2015 15:36:20 -0500] rev 27381
test-hgignore: conditionalize an illegal Windows filename
Matt Mackall <mpm@selenic.com> [Sat, 12 Dec 2015 21:36:21 -0600] rev 27380
mac: fix percent-encoding of non-utf-8 characters (issue4999)
Gregory Szorc <gregory.szorc@gmail.com> [Sun, 13 Dec 2015 11:19:55 -0800] rev 27379
help: support loading sub-topics
If a sub-topic/section is requested and the main topic corresponds to
a topic with sub-topics, we now look for and return content for a
sub-topic if found.
With this patch, `hg help internals.X` now works. hgweb does not yet
render sub-topics, however.
Gregory Szorc <gregory.szorc@gmail.com> [Sun, 13 Dec 2015 11:29:01 -0800] rev 27378
help: pass sub-topic into help query functions
While we will likely only use this variable in helptopic(), all these
functions are called with the same arguments, so we have to be
consistent.
Gregory Szorc <gregory.szorc@gmail.com> [Sun, 13 Dec 2015 11:04:45 -0800] rev 27377
help: pass subtopic into help()
Now that we have multiple directories where help topics can live,
we need a mechanism to access them. We already use "." to
separate topic from section. So it seems logical to also use "." to
denote the sub-directory of a topic.
This patch teaches the help command to parse out the possible
sub-topic and pass it to the help system.
Gregory Szorc <gregory.szorc@gmail.com> [Sun, 13 Dec 2015 10:35:03 -0800] rev 27376
help: add "internals" topic
We introduce the "internals" help topic, which renders an index of
available sub-topics. The sub-topics themselves are still not
reachable via the help system.
Gregory Szorc <gregory.szorc@gmail.com> [Sun, 13 Dec 2015 10:45:27 -0800] rev 27375
help: teach loaddoc to load from a different directory
The help system currently only supports showing help topics from a
single directory. We'll need to teach it to show results from
different directories in order to show the internals topics.
The first step is to teach loaddoc() to load documentation from
a sub-directory.
Gregory Szorc <gregory.szorc@gmail.com> [Sun, 13 Dec 2015 11:34:04 -0800] rev 27374
setup.py: package internals help files
mpm says internal docs should be visible via `hg help` and hgweb. They
need to be in the distribution for this to work. Package them.
Gregory Szorc <gregory.szorc@gmail.com> [Sun, 13 Dec 2015 11:27:52 -0800] rev 27373
help: add documentation for bundle types
Bundle types and the high-level data format of each bundle isn't
documented anywhere. Let's document this as well.
Obviously there are many more details about bundles that could be
written about. But you have to start somewhere.
Gregory Szorc <gregory.szorc@gmail.com> [Sun, 25 Oct 2015 00:19:45 +0100] rev 27372
help: add documentation for changegroup formats
There is no formal location for spec-like technical/internal docs. The
repository makes sense as such a location because spec-like
documentation should be reviewed (ruling out a wiki). mpm has also
stated that he would like this documentation to be part of the
built-in help system. So, we establish an "internals" sub-directory
to hold this class of documentation.
The format of changegroups does not appear to be documented anywhere,
even in source code. It therefore seemed like an appropriate first thing
to document.
This patch adds low-level documentation of versions 1 and 2 of the
changegroup foromat. It currently only describes the raw data format.
There is probably room to write higher-level documentation on strategies
for producing and consuming the data. We'll leave that for another day.
The added file is not yet accessible via `hg help` nor via hgweb.
Support for this will follow in subsequent patches.
Gregory Szorc <gregory.szorc@gmail.com> [Sun, 06 Dec 2015 19:04:10 -0800] rev 27371
util: reimplement lrucachedict
As part of attempting to more aggressively use the existing
lrucachedict, collections.deque operations were frequently
showing up in profiling output, negating benefits of caching.
Searching the internet seems to tell me that the most efficient
way to implement an LRU cache in Python is to have a dict indexing
the cached entries and then to use a doubly linked list to track
freshness of each entry. So, this patch replaces our existing
lrucachedict with a version using such a pattern.
The recently introduced perflrucachedict command reveals the
following timings for 10,000 operations for the following cache
sizes for the existing cache:
n=4 init=0.004079 gets=0.003632 sets=0.005188 mixed=0.005402
n=8 init=0.004045 gets=0.003998 sets=0.005064 mixed=0.005328
n=16 init=0.004011 gets=0.004496 sets=0.005021 mixed=0.005555
n=32 init=0.004064 gets=0.005611 sets=0.005188 mixed=0.006189
n=64 init=0.003975 gets=0.007684 sets=0.005178 mixed=0.007245
n=128 init=0.004121 gets=0.012005 sets=0.005422 mixed=0.009471
n=256 init=0.004143 gets=0.020295 sets=0.005227 mixed=0.013612
n=512 init=0.004039 gets=0.036703 sets=0.005243 mixed=0.020685
n=1024 init=0.004193 gets=0.068142 sets=0.005251 mixed=0.033064
n=2048 init=0.004070 gets=0.133383 sets=0.005160 mixed=0.050359
n=4096 init=0.004053 gets=0.265194 sets=0.004868 mixed=0.048352
n=8192 init=0.004087 gets=0.542218 sets=0.004562 mixed=0.032753
n=16384 init=0.004106 gets=1.064055 sets=0.004179 mixed=0.020367
n=32768 init=0.004034 gets=2.097620 sets=0.004260 mixed=0.013031
n=65536 init=0.004108 gets=4.106390 sets=0.004268 mixed=0.010191
As the data shows, the existing cache's retrieval performance
diminishes linearly with cache size. (Keep in mind the microbenchmark
is testing 100% cache hit rate.)
The new cache implementation reveals the following:
n=4 init=0.006665 gets=0.006541 sets=0.005733 mixed=0.006876
n=8 init=0.006649 gets=0.006374 sets=0.005663 mixed=0.006899
n=16 init=0.006570 gets=0.006504 sets=0.005799 mixed=0.007057
n=32 init=0.006854 gets=0.006459 sets=0.005747 mixed=0.007034
n=64 init=0.006580 gets=0.006495 sets=0.005740 mixed=0.006992
n=128 init=0.006534 gets=0.006739 sets=0.005648 mixed=0.007124
n=256 init=0.006669 gets=0.006773 sets=0.005824 mixed=0.007151
n=512 init=0.006701 gets=0.007061 sets=0.006042 mixed=0.007372
n=1024 init=0.006641 gets=0.007620 sets=0.006387 mixed=0.007464
n=2048 init=0.006517 gets=0.008598 sets=0.006871 mixed=0.008077
n=4096 init=0.006720 gets=0.010933 sets=0.007854 mixed=0.008663
n=8192 init=0.007383 gets=0.015969 sets=0.010288 mixed=0.008896
n=16384 init=0.006660 gets=0.025447 sets=0.011208 mixed=0.008826
n=32768 init=0.006658 gets=0.044390 sets=0.011192 mixed=0.008943
n=65536 init=0.006836 gets=0.082736 sets=0.011151 mixed=0.008826
Let's go through the results.
The new cache takes longer to construct. ~6.6ms vs ~4.1ms. However,
this is measuring 10,000 __init__ calls, so the difference is
~0.2us/instance. We currently only create lrucachedict for manifest
instances, so this regression is not likely relevant.
The new cache is slightly slower for retrievals for cache sizes
< 1024. It's worth noting that the only existing use of lurcachedict
is in manifest.py and the default cache size is 4. This regression
is worrisome. However, for n=4, the delta is ~2.9s for 10,000 lookups,
or ~0.29us/op. Again, this is a marginal regression and likely not
relevant in the real world. Timing `hg log -p -l 100` for
mozilla-central reveals that cache lookup times are dominated by
decompression and fulltext resolution (even with lz4 manifests).
The new cache is significantly faster for retrievals at larger
capacities. Whereas the old implementation has retrieval performance
linear with cache capacity, the new cache is constant time until much
larger values. And, when it does start to increase significantly, it
is a few magnitudes faster than the current cache.
The new cache does appear to be slower for sets when capacity is large.
However, performance is similar for smaller capacities. Of course,
caches should generally be optimized for retrieval performance because
if a cache is getting more sets than gets, it doesn't really make
sense to cache. If this regression is worrisome, again, taking the
largest regression at n=65536 of ~6.9ms for 10,000 results in a
regression of ~0.68us/op. This is not significant in the grand scheme
of things.
Overall, the new cache is performant at retrievals at much larger
capacity values which makes it a generally more useful cache backend.
While there are regressions, their absolute value is extremely small.
Since we aren't using lrucachedict aggressively today, these
regressions should not be relevant. The improved scalability of
lrucachedict should enable us to more aggressively utilize
lrucachedict for more granular caching (read: higher capacity caches)
in the near future. The impetus for this patch is to establish a cache
of decompressed revlog revisions, notably manifest revisions. And since
delta chains can grow to >10,000 and cache hit rate can be high, the
improved retrieval performance of lrucachedict should be relevant.
Siddharth Agarwal <sid0@fb.com> [Sat, 12 Dec 2015 10:58:05 -0800] rev 27370
record: don't dereference symlinks while copying over stat data
Previously, we could be calling os.utime or os.chflags (via shutil.copystat) on
a symlink. These functions dereference symlinks, so this would have caused the
timestamp of the target to be set. On a read-only or similarly weird
filesystem, this might cause an exception to be raised.
This is pretty hard to test because conjuring up a read-only filesystem for
test purposes is non-trivial.
Siddharth Agarwal <sid0@fb.com> [Sat, 12 Dec 2015 11:00:04 -0800] rev 27369
copyfile: add an optional parameter to copy other stat data
Contrary to the comment, I didn't see any evidence that we were copying
atime/mtime at all. This adds a parameter to copyfile to optionally copy it and
other stat data, with the default being to not copy it.
Many systems don't support changing the timestamp of a symlink, but we don't
need that in general anyway -- copystat is mostly useful for editors, most of
which will dereference symlinks anyway.
Pierre-Yves David <pierre-yves.david@fb.com> [Sat, 05 Dec 2015 22:49:39 -0800] rev 27368
tests: move the '-hg' postfix for all style tests
We had them on 'test-check-code-hg.t' to avoid collision with the test checking
'check-code' itself. Now that this one have been rename, we can safely remove
this suffix for all of them. This get them in line with 'check-pyflakes.t'.
Pierre-Yves David <pierre-yves.david@fb.com> [Sat, 05 Dec 2015 22:47:26 -0800] rev 27367
test: rename 'check-code' own test to 'test-contrib-check-code.t'
This test (making sure the 'check-code' script run as intended) have been
confused with the test making that the mercurial code base comply with our
coding still by multiple generations of contributors.
We are moving it out of the way so that all tests starting with
'test-check' are now doing compliance testing.
Bryan O'Sullivan <bos@serpentine.com> [Mon, 14 Dec 2015 10:47:27 -0800] rev 27366
parsers: add a missed PyErr_NoMemory
Bryan O'Sullivan <bos@serpentine.com> [Mon, 14 Dec 2015 10:47:26 -0800] rev 27365
parsers: check results of PyInt_FromLong (issue4771)
Bryan O'Sullivan <bos@serpentine.com> [Mon, 14 Dec 2015 10:47:24 -0800] rev 27364
parsers: simplify error logic in compute_phases_map_sets
Since Py_XDECREF and free both accept NULL pointers, we can get by
with just two exit paths: one for success, and one for error.
This considerably simplifies reasoning about the possible ways to
exit from this function.
Yuya Nishihara <yuya@tcha.org> [Sun, 13 Dec 2015 18:48:35 +0900] rev 27363
util: rename argument of isatty()
In general, "fd" is a file descriptor, but isatty() expects a file object.
We should call it "fp" or "fh".
Yuya Nishihara <yuya@tcha.org> [Sat, 21 Nov 2015 16:21:52 +0900] rev 27362
posix: remove unixdomainserver class
It's no longer used since the removal of the inotify extension.
Gregory Szorc <gregory.szorc@gmail.com> [Sat, 12 Dec 2015 23:22:18 -0800] rev 27361
revlog: use absolute_import
Gregory Szorc <gregory.szorc@gmail.com> [Sat, 12 Dec 2015 23:19:38 -0800] rev 27360
windows: use absolute_import
Gregory Szorc <gregory.szorc@gmail.com> [Sat, 12 Dec 2015 23:17:22 -0800] rev 27359
similar: use absolute_import
Gregory Szorc <gregory.szorc@gmail.com> [Sat, 12 Dec 2015 23:14:08 -0800] rev 27358
util: use absolute_import
Gregory Szorc <gregory.szorc@gmail.com> [Sat, 12 Dec 2015 23:30:37 -0500] rev 27357
util: make hashlib import unconditional
hashlib was added in Python 2.5. As far as I can tell, SHA-512 is always
available in 2.6+. So move the hashlib import to the top of the file and
remove the one-off handling of SHA-512.
Gregory Szorc <gregory.szorc@gmail.com> [Sat, 12 Dec 2015 23:26:12 -0800] rev 27356
encoding: use double backslash
In Python 2, '\u' == '\\u'. However, in Python 3, '\u' results in:
SyntaxError: (unicode error) 'unicodeescape' codec can't decode bytes in
position 0-1: truncated \uXXXX escape
The minor change in this patch allows Python 3 to ast parse encoding.py.
Gregory Szorc <gregory.szorc@gmail.com> [Sat, 12 Dec 2015 22:57:48 -0500] rev 27355
encoding: use absolute_import
Gregory Szorc <gregory.szorc@gmail.com> [Sat, 12 Dec 2015 22:20:29 -0500] rev 27354
hg: establish function for performing post-share actions
As part of writing an extension that wished to share an arbitrary piece
of data among shared repos, I had to reimplement a significant part of
hg.share in order to obtain localrepository instances for the source
and destination.
This patch establishes a function in hg.py that will be called after a
share is performed. It is passed localrepository instances so extensions
can easily perform additional actions at share time. We move hgrc and
shared file writing there because this function is a logical place for
it.
A side effect of the refactor is writing of the shared file now occurs
before updating. This seems more appropriate and shouldn't have any
impact on real world behavior.
Gregory Szorc <gregory.szorc@gmail.com> [Sat, 12 Dec 2015 22:07:40 -0500] rev 27353
share: pass named arguments
They are defined as named arguments and previous called as positional
arguments. As part of wrapping hg.share in an extension, I had to
extract arguments using some hacky techniques. Using named arguments
makes wrapping much simpler.
Yuya Nishihara <yuya@tcha.org> [Tue, 24 Nov 2015 23:03:54 +0900] rev 27352
commandserver: cut import cycle by itself
We generally make modules importable from the front-end layer, dispatch ->
commands -> x. So the import cycle to dispatch should be resolved by the
commandserver module.
Yuya Nishihara <yuya@tcha.org> [Tue, 24 Nov 2015 22:58:40 +0900] rev 27351
commandserver: use absolute_import
timeless <timeless@mozdev.org> [Mon, 14 Dec 2015 22:21:30 +0000] rev 27350
tests: histedit-helpers fixbundle should not complain about no input
timeless <timeless@mozdev.org> [Mon, 14 Dec 2015 22:34:30 +0000] rev 27349
tests: relax histedit issue4251 and issue3893 backups
I'm globbing these because some are globbed, and this pair
gets in the way of the main parts of the series.
Gregory Szorc <gregory.szorc@gmail.com> [Sun, 13 Dec 2015 09:40:53 -0800] rev 27348
setup.py: use bytes literals
The b() helper was needed because Python < 2.6 didn't support bytes
literals (b''). Now that we don't support Python < 2.6, we no longer
need this helper.
Mathias De Maré <mathias.demare@gmail.com> [Tue, 15 Dec 2015 07:17:15 +0100] rev 27347
clonebundles: fix typo
Augie Fackler <augie@google.com> [Mon, 14 Dec 2015 20:47:22 -0500] rev 27346
merge: rework manifestmerge to use a matcher
This opens the door to working slightly more closely with the manifest
type and letting it optimize out some of the diff comparisons for us,
and also makes life significantly easier for narrowhg.
Augie Fackler <augie@google.com> [Mon, 14 Dec 2015 20:37:41 -0500] rev 27345
merge: restate calculateupdates in terms of a matcher
Once we get a matcher down into manifestmerge, we can make narrowhg
work more easily and potentially let manifest.match().diff() do less
work in manifestmerge.
Augie Fackler <augie@google.com> [Mon, 14 Dec 2015 18:54:03 -0500] rev 27344
merge: have merge.update use a matcher instead of partial fn
This is relatively rarely used functionality, but migrating this to a
matcher will make future work on narrow clones more feasible.
Martin von Zweigbergk <martinvonz@google.com> [Sat, 12 Dec 2015 09:57:05 -0800] rev 27343
treemanifest: don't iterate entire matching submanifests on match()
Before 2773540c3650 (match: remove unnecessary optimization where
visitdir() returns 'all', 2015-05-06), match.visitdir() used to return
the special value 'all' to indicate that it was known that all
subdirectories would also be included in the match. The purpose for
that value was to avoid calling the matcher on all the paths. It
turned out that calling the matcher was not a problem, so the special
return value was removed and the code was simplified. However, if we
use the same special value for not just avoiding calling the matcher
on each file, but to avoid iterating over each file, it's a much
bigger win. On commands like
hg st --rev .^ --rev . dom/
we run the matcher (dom/) on the two manifests, then diff the narrowed
manifest. If the size of the match is much larger than the size of the
diff, this is wasteful. In the above case, we would end up iterating
over the 15k-or-so files in dom/ for each of the manifests, only to
later discover that they are mostly the same. This means that runningt
the command above is usually slower than getting the status for the
entire repo, because that code avoids calling treemanifest.match() and
only calls treemanifest.diff(), which loads only what's needed for the
diff.
Let's fix this by reintroducing the 'all' value in match.visitdir()
and making treemanifest.match() return a lazy copy of the manifest
from dom/ and down (in the above case). This speeds up the above
command on the Firefox repo from 0.357s to 0.137s (best of 5). The
wider the match, the bigger the speedup.
Bryan O'Sullivan <bos@serpentine.com> [Sat, 12 Dec 2015 20:59:37 -0800] rev 27342
pathencode: remove an unused assignment
Bryan O'Sullivan <bos@serpentine.com> [Sat, 12 Dec 2015 20:57:01 -0800] rev 27341
parsers: narrow scope of a variable to be less confusing