Mercurial > public > mercurial-scm > hg
view mercurial/minirst.py @ 45892:06b64fabf91c
copies: cache the ancestor checking call when tracing copy
A good share of the time spent in this function is spent doing ancestors
checking. To avoid spending time in duplicated call, we cache the result of
calls.
In the slower case, this provide a quite significant performance boost. Below
are the result for a set of selected pairs (many of them pathological):
(And further down is another table that summarize the current state of filelog
based vs changeset base copy tracing)
The benchmark have been configured to be killed after 6 minutes of runtime,
which mean that any detect slower than 2 minutes will be marked as "killed".
This drop some useful information about how much slower these case are? but also
prevent 99% of the benchmark time to be spent on case that can be labelled "very
slow" anyway.
Repo Case Source-Rev Dest-Rev Old-Time New-Time Difference Factor
------------------------------------------------------------------------------------------------------------------------------------
mercurial x_revs_x_added_0_copies ad6b123de1c7 39cfcef4f463 : 0.000044 s, 0.000044 s, +0.000000 s, ? 1.0000
mercurial x_revs_x_added_x_copies 2b1c78674230 0c1d10351869 : 0.000138 s, 0.000138 s, +0.000000 s, ? 1.0000
mercurial x000_revs_x000_added_x_copies 81f8ff2a9bf2 dd3267698d84 : 0.005067 s, 0.005052 s, -0.000015 s, ? 0.9970
pypy x_revs_x_added_0_copies aed021ee8ae8 099ed31b181b : 0.000218 s, 0.000219 s, +0.000001 s, ? 1.0046
pypy x_revs_x000_added_0_copies 4aa4e1f8e19a 359343b9ac0e : 0.000053 s, 0.000055 s, +0.000002 s, ? 1.0377
pypy x_revs_x_added_x_copies ac52eb7bbbb0 72e022663155 : 0.000125 s, 0.000128 s, +0.000003 s, ? 1.0240
pypy x_revs_x00_added_x_copies c3b14617fbd7 ace7255d9a26 : 0.001098 s, 0.001089 s, -0.000009 s, ? 0.9918
pypy x_revs_x000_added_x000_copies df6f7a526b60 a83dc6a2d56f : 0.017546 s, 0.017407 s, -0.000139 s, ? 0.9921
pypy x000_revs_xx00_added_0_copies 89a76aede314 2f22446ff07e : 0.096723 s, 0.094175 s, -0.002548 s, ? 0.9737
pypy x000_revs_x000_added_x_copies 8a3b5bfd266e 2c68e87c3efe : 0.271796 s, 0.238009 s, -0.033787 s, ? 0.8757
pypy x000_revs_x000_added_x000_copies 89a76aede314 7b3dda341c84 : 0.128602 s, 0.125876 s, -0.002726 s, ? 0.9788
pypy x0000_revs_x_added_0_copies d1defd0dc478 c9cb1334cc78 : 7.086742 s, 3.581556 s, -3.505186 s, ? 0.5054
pypy x0000_revs_xx000_added_0_copies bf2c629d0071 4ffed77c095c : 0.016634 s, 0.016721 s, +0.000087 s, ? 1.0052
pypy x0000_revs_xx000_added_x000_copies 08ea3258278e d9fa043f30c0 : 0.254225 s, 0.242367 s, -0.011858 s, ? 0.9534
netbeans x_revs_x_added_0_copies fb0955ffcbcd a01e9239f9e7 : 0.000166 s, 0.000165 s, -0.000001 s, ? 0.9940
netbeans x_revs_x000_added_0_copies 6f360122949f 20eb231cc7d0 : 0.000118 s, 0.000114 s, -0.000004 s, ? 0.9661
netbeans x_revs_x_added_x_copies 1ada3faf6fb6 5a39d12eecf4 : 0.000296 s, 0.000296 s, +0.000000 s, ? 1.0000
netbeans x_revs_x00_added_x_copies 35be93ba1e2c 9eec5e90c05f : 0.001137 s, 0.001124 s, -0.000013 s, ? 0.9886
netbeans x000_revs_xx00_added_0_copies eac3045b4fdd 51d4ae7f1290 : 0.014133 s, 0.013060 s, -0.001073 s, ? 0.9241
netbeans x000_revs_x000_added_x_copies e2063d266acd 6081d72689dc : 0.016988 s, 0.017112 s, +0.000124 s, ? 1.0073
netbeans x000_revs_x000_added_x000_copies ff453e9fee32 411350406ec2 : 0.676361 s, 0.660350 s, -0.016011 s, ? 0.9763
netbeans x0000_revs_xx000_added_x000_copies 588c2d1ced70 1aad62e59ddd : 12.515149 s, 10.032499 s, -2.482650 s, ? 0.8016
mozilla-central x_revs_x_added_0_copies 3697f962bb7b 7015fcdd43a2 : 0.000186 s, 0.000189 s, +0.000003 s, ? 1.0161
mozilla-central x_revs_x000_added_0_copies dd390860c6c9 40d0c5bed75d : 0.000459 s, 0.000462 s, +0.000003 s, ? 1.0065
mozilla-central x_revs_x_added_x_copies 8d198483ae3b 14207ffc2b2f : 0.000273 s, 0.000270 s, -0.000003 s, ? 0.9890
mozilla-central x_revs_x00_added_x_copies 98cbc58cc6bc 446a150332c3 : 0.001503 s, 0.001474 s, -0.000029 s, ? 0.9807
mozilla-central x_revs_x000_added_x000_copies 3c684b4b8f68 0a5e72d1b479 : 0.004862 s, 0.004806 s, -0.000056 s, ? 0.9885
mozilla-central x_revs_x0000_added_x0000_copies effb563bb7e5 c07a39dc4e80 : 0.088291 s, 0.085150 s, -0.003141 s, ? 0.9644
mozilla-central x000_revs_xx00_added_0_copies 6100d773079a 04a55431795e : 0.007113 s, 0.007064 s, -0.000049 s, ? 0.9931
mozilla-central x000_revs_x000_added_x_copies 9f17a6fc04f9 2d37b966abed : 0.004687 s, 0.004741 s, +0.000054 s, ? 1.0115
mozilla-central x000_revs_x000_added_x000_copies 7c97034feb78 4407bd0c6330 : 0.198710 s, 0.190133 s, -0.008577 s, ? 0.9568
mozilla-central x0000_revs_xx000_added_0_copies 9eec5917337d 67118cc6dcad : 0.036068 s, 0.035651 s, -0.000417 s, ? 0.9884
mozilla-central x0000_revs_xx000_added_x000_copies f78c615a656c 96a38b690156 : 0.465362 s, 0.440694 s, -0.024668 s, ? 0.9470
mozilla-central x00000_revs_x0000_added_x0000_copies 6832ae71433c 4c222a1d9a00 : 24.519684 s, 18.454163 s, -6.065521 s, ? 0.7526
mozilla-central x00000_revs_x00000_added_x000_copies 76caed42cf7c 1daa622bbe42 : 42.711897 s, 31.562719 s, -11.149178 s, ? 0.7390
mozilla-try x_revs_x_added_0_copies aaf6dde0deb8 9790f499805a : 0.001201 s, 0.001189 s, -0.000012 s, ? 0.9900
mozilla-try x_revs_x000_added_0_copies d8d0222927b4 5bb8ce8c7450 : 0.001216 s, 0.001204 s, -0.000012 s, ? 0.9901
mozilla-try x_revs_x_added_x_copies 092fcca11bdb 936255a0384a : 0.000595 s, 0.000586 s, -0.000009 s, ? 0.9849
mozilla-try x_revs_x00_added_x_copies b53d2fadbdb5 017afae788ec : 0.001856 s, 0.001845 s, -0.000011 s, ? 0.9941
mozilla-try x_revs_x000_added_x000_copies 20408ad61ce5 6f0ee96e21ad : 0.064936 s, 0.063822 s, -0.001114 s, ? 0.9828
mozilla-try x_revs_x0000_added_x0000_copies effb563bb7e5 c07a39dc4e80 : 0.090601 s, 0.088038 s, -0.002563 s, ? 0.9717
mozilla-try x000_revs_xx00_added_0_copies 6100d773079a 04a55431795e : 0.007510 s, 0.007389 s, -0.000121 s, ? 0.9839
mozilla-try x000_revs_x000_added_x_copies 9f17a6fc04f9 2d37b966abed : 0.004911 s, 0.004868 s, -0.000043 s, ? 0.9912
mozilla-try x000_revs_x000_added_x000_copies 1346fd0130e4 4c65cbdabc1f : 0.233231 s, 0.222450 s, -0.010781 s, ? 0.9538
mozilla-try x0000_revs_x_added_0_copies 63519bfd42ee a36a2a865d92 : 0.419989 s, 0.370675 s, -0.049314 s, ? 0.8826
mozilla-try x0000_revs_x_added_x_copies 9fe69ff0762d bcabf2a78927 : 0.401521 s, 0.358020 s, -0.043501 s, ? 0.8917
mozilla-try x0000_revs_xx000_added_x_copies 156f6e2674f2 4d0f2c178e66 : 0.179555 s, 0.145235 s, -0.034320 s, ? 0.8089
mozilla-try x0000_revs_xx000_added_0_copies 9eec5917337d 67118cc6dcad : 0.038004 s, 0.037606 s, -0.000398 s, ? 0.9895
mozilla-try x0000_revs_xx000_added_x000_copies 89294cd501d9 7ccb2fc7ccb5 : 52.838482 s, 7.382439 s, -45.456043 s, ? 0.1397
mozilla-try x0000_revs_x0000_added_x0000_copies e928c65095ed e951f4ad123a : 8.705874 s, 7.273506 s, -1.432368 s, ? 0.8355
mozilla-try x00000_revs_x00000_added_0_copies dc8a3ca7010e d16fde900c9c : 1.126708 s, 1.074593 s, -0.052115 s, ? 0.9537
mozilla-try x00000_revs_x0000_added_x0000_copies 8d3fafa80d4b eb884023b810 : 83.854020 s, 27.746195 s, -56.107825 s, ? 0.3309
Below is a table comparing the runtime of the current "filelog centric"
algorithm, with the "changeset centric" one, we just modified.
The changeset centric algorithm is a significant win in many scenario, but they
are still various cases where it is quite slower. When many revision has to be
considered the cost of retrieving the copy information, creating new
dictionaries, merging dictionaries and checking if revision are ancestors of
each other can slow things down.
The rest of this series, will introduce a rust version of the copy tracing code
to deal with most of theses issues.
Repo Case Source-Rev Dest-Rev filelog sidedata Difference Factor
---------------------------------------------------------------------------------------------------------------------------------------
mercurial x_revs_x_added_0_copies ad6b123de1c7 39cfcef4f463 : 0.000914 s, 0.000044 s, - 0.000870 s, ? 0.048140
mercurial x_revs_x_added_x_copies 2b1c78674230 0c1d10351869 : 0.001812 s, 0.000138 s, - 0.001674 s, ? 0.076159
mercurial x000_revs_x000_added_x_copies 81f8ff2a9bf2 dd3267698d84 : 0.017954 s, 0.005052 s, - 0.012902 s, ? 0.281386
pypy x_revs_x_added_0_copies aed021ee8ae8 099ed31b181b : 0.001509 s, 0.000219 s, - 0.001290 s, ? 0.145129
pypy x_revs_x000_added_0_copies 4aa4e1f8e19a 359343b9ac0e : 0.206881 s, 0.000055 s, - 0.206826 s, ? 0.000266
pypy x_revs_x_added_x_copies ac52eb7bbbb0 72e022663155 : 0.016951 s, 0.000128 s, - 0.016823 s, ? 0.007551
pypy x_revs_x00_added_x_copies c3b14617fbd7 ace7255d9a26 : 0.019096 s, 0.001089 s, - 0.018007 s, ? 0.057028
pypy x_revs_x000_added_x000_copies df6f7a526b60 a83dc6a2d56f : 0.762506 s, 0.017407 s, - 0.745099 s, ? 0.022829
pypy x000_revs_xx00_added_0_copies 89a76aede314 2f22446ff07e : 1.179211 s, 0.094175 s, - 1.085036 s, ? 0.079863
pypy x000_revs_x000_added_x_copies 8a3b5bfd266e 2c68e87c3efe : 1.249058 s, 0.238009 s, - 1.011049 s, ? 0.190551
pypy x000_revs_x000_added_x000_copies 89a76aede314 7b3dda341c84 : 1.614107 s, 0.125876 s, - 1.488231 s, ? 0.077985
pypy x0000_revs_x_added_0_copies d1defd0dc478 c9cb1334cc78 : 0.001064 s, 3.581556 s, + 3.580492 s, ? 3366.124060
pypy x0000_revs_xx000_added_0_copies bf2c629d0071 4ffed77c095c : 1.061275 s, 0.016721 s, - 1.044554 s, ? 0.015756
pypy x0000_revs_xx000_added_x000_copies 08ea3258278e d9fa043f30c0 : 1.341119 s, 0.242367 s, - 1.098752 s, ? 0.180720
netbeans x_revs_x_added_0_copies fb0955ffcbcd a01e9239f9e7 : 0.027803 s, 0.000165 s, - 0.027638 s, ? 0.005935
netbeans x_revs_x000_added_0_copies 6f360122949f 20eb231cc7d0 : 0.130014 s, 0.000114 s, - 0.129900 s, ? 0.000877
netbeans x_revs_x_added_x_copies 1ada3faf6fb6 5a39d12eecf4 : 0.024990 s, 0.000296 s, - 0.024694 s, ? 0.011845
netbeans x_revs_x00_added_x_copies 35be93ba1e2c 9eec5e90c05f : 0.052201 s, 0.001124 s, - 0.051077 s, ? 0.021532
netbeans x000_revs_xx00_added_0_copies eac3045b4fdd 51d4ae7f1290 : 0.037642 s, 0.013060 s, - 0.024582 s, ? 0.346953
netbeans x000_revs_x000_added_x_copies e2063d266acd 6081d72689dc : 0.197086 s, 0.017112 s, - 0.179974 s, ? 0.086825
netbeans x000_revs_x000_added_x000_copies ff453e9fee32 411350406ec2 : 0.935148 s, 0.660350 s, - 0.274798 s, ? 0.706145
netbeans x0000_revs_xx000_added_x000_copies 588c2d1ced70 1aad62e59ddd : 3.920674 s, 10.032499 s, + 6.111825 s, ? 2.558871
mozilla-central x_revs_x_added_0_copies 3697f962bb7b 7015fcdd43a2 : 0.024232 s, 0.000189 s, - 0.024043 s, ? 0.007800
mozilla-central x_revs_x000_added_0_copies dd390860c6c9 40d0c5bed75d : 0.141483 s, 0.000462 s, - 0.141021 s, ? 0.003265
mozilla-central x_revs_x_added_x_copies 8d198483ae3b 14207ffc2b2f : 0.025775 s, 0.000270 s, - 0.025505 s, ? 0.010475
mozilla-central x_revs_x00_added_x_copies 98cbc58cc6bc 446a150332c3 : 0.084922 s, 0.001474 s, - 0.083448 s, ? 0.017357
mozilla-central x_revs_x000_added_x000_copies 3c684b4b8f68 0a5e72d1b479 : 0.194784 s, 0.004806 s, - 0.189978 s, ? 0.024673
mozilla-central x_revs_x0000_added_x0000_copies effb563bb7e5 c07a39dc4e80 : 2.161103 s, 0.085150 s, - 2.075953 s, ? 0.039401
mozilla-central x000_revs_xx00_added_0_copies 6100d773079a 04a55431795e : 0.089347 s, 0.007064 s, - 0.082283 s, ? 0.079063
mozilla-central x000_revs_x000_added_x_copies 9f17a6fc04f9 2d37b966abed : 0.732171 s, 0.004741 s, - 0.727430 s, ? 0.006475
mozilla-central x000_revs_x000_added_x000_copies 7c97034feb78 4407bd0c6330 : 1.157287 s, 0.190133 s, - 0.967154 s, ? 0.164292
mozilla-central x0000_revs_xx000_added_0_copies 9eec5917337d 67118cc6dcad : 6.726568 s, 0.035651 s, - 6.690917 s, ? 0.005300
mozilla-central x0000_revs_xx000_added_x000_copies f78c615a656c 96a38b690156 : 3.266229 s, 0.440694 s, - 2.825535 s, ? 0.134924
mozilla-central x00000_revs_x0000_added_x0000_copies 6832ae71433c 4c222a1d9a00 : 15.860534 s, 18.454163 s, + 2.593629 s, ? 1.163527
mozilla-central x00000_revs_x00000_added_x000_copies 76caed42cf7c 1daa622bbe42 : 20.450475 s, 31.562719 s, +11.112244 s, ? 1.543373
mozilla-try x_revs_x_added_0_copies aaf6dde0deb8 9790f499805a : 0.080442 s, 0.001189 s, - 0.079253 s, ? 0.014781
mozilla-try x_revs_x000_added_0_copies d8d0222927b4 5bb8ce8c7450 : 0.497672 s, 0.001204 s, - 0.496468 s, ? 0.002419
mozilla-try x_revs_x_added_x_copies 092fcca11bdb 936255a0384a : 0.021183 s, 0.000586 s, - 0.020597 s, ? 0.027664
mozilla-try x_revs_x00_added_x_copies b53d2fadbdb5 017afae788ec : 0.230991 s, 0.001845 s, - 0.229146 s, ? 0.007987
mozilla-try x_revs_x000_added_x000_copies 20408ad61ce5 6f0ee96e21ad : 1.118461 s, 0.063822 s, - 1.054639 s, ? 0.057062
mozilla-try x_revs_x0000_added_x0000_copies effb563bb7e5 c07a39dc4e80 : 2.206083 s, 0.088038 s, - 2.118045 s, ? 0.039907
mozilla-try x000_revs_xx00_added_0_copies 6100d773079a 04a55431795e : 0.089404 s, 0.007389 s, - 0.082015 s, ? 0.082647
mozilla-try x000_revs_x000_added_x_copies 9f17a6fc04f9 2d37b966abed : 0.733043 s, 0.004868 s, - 0.728175 s, ? 0.006641
mozilla-try x000_revs_x000_added_x000_copies 1346fd0130e4 4c65cbdabc1f : 1.163367 s, 0.222450 s, - 0.940917 s, ? 0.191212
mozilla-try x0000_revs_x_added_0_copies 63519bfd42ee a36a2a865d92 : 0.085456 s, 0.370675 s, + 0.285219 s, ? 4.337612
mozilla-try x0000_revs_x_added_x_copies 9fe69ff0762d bcabf2a78927 : 0.083601 s, 0.358020 s, + 0.274419 s, ? 4.282485
mozilla-try x0000_revs_xx000_added_x_copies 156f6e2674f2 4d0f2c178e66 : 7.366614 s, 0.145235 s, - 7.221379 s, ? 0.019715
mozilla-try x0000_revs_xx000_added_0_copies 9eec5917337d 67118cc6dcad : 6.664464 s, 0.037606 s, - 6.626858 s, ? 0.005643
mozilla-try x0000_revs_xx000_added_x000_copies 89294cd501d9 7ccb2fc7ccb5 : 7.467836 s, 7.382439 s, - 0.085397 s, ? 0.988565
mozilla-try x0000_revs_x0000_added_x0000_copies e928c65095ed e951f4ad123a : 9.801294 s, 7.273506 s, - 2.527788 s, ? 0.742097
mozilla-try x00000_revs_x_added_0_copies 6a320851d377 1ebb79acd503 : 0.091886 s, killed
mozilla-try x00000_revs_x00000_added_0_copies dc8a3ca7010e d16fde900c9c : 26.491140 s, 1.074593 s, -25.416547 s, ? 0.040564
mozilla-try x00000_revs_x_added_x_copies 5173c4b6f97c 95d83ee7242d : 0.092863 s, killed
mozilla-try x00000_revs_x000_added_x_copies 9126823d0e9c ca82787bb23c : 0.226823 s, killed
mozilla-try x00000_revs_x0000_added_x0000_copies 8d3fafa80d4b eb884023b810 : 18.914630 s, 27.746195 s, + 8.831565 s, ? 1.466917
mozilla-try x00000_revs_x00000_added_x0000_copies 1b661134e2ca 1ae03d022d6d : 21.198903 s, killed
mozilla-try x00000_revs_x00000_added_x000_copies 9b2a99adc05e 8e29777b48e6 : 24.952268 s, killed
Differential Revision: https://phab.mercurial-scm.org/D9296
author | Pierre-Yves David <pierre-yves.david@octobus.net> |
---|---|
date | Mon, 02 Nov 2020 11:03:56 +0100 |
parents | 687b865b95ad |
children | 89a2afe31e82 |
line wrap: on
line source
# minirst.py - minimal reStructuredText parser # # Copyright 2009, 2010 Matt Mackall <mpm@selenic.com> and others # # This software may be used and distributed according to the terms of the # GNU General Public License version 2 or any later version. """simplified reStructuredText parser. This parser knows just enough about reStructuredText to parse the Mercurial docstrings. It cheats in a major way: nested blocks are not really nested. They are just indented blocks that look like they are nested. This relies on the user to keep the right indentation for the blocks. Remember to update https://mercurial-scm.org/wiki/HelpStyleGuide when adding support for new constructs. """ from __future__ import absolute_import import re from .i18n import _ from . import ( encoding, pycompat, url, ) from .utils import stringutil def section(s): return b"%s\n%s\n\n" % (s, b"\"" * encoding.colwidth(s)) def subsection(s): return b"%s\n%s\n\n" % (s, b'=' * encoding.colwidth(s)) def subsubsection(s): return b"%s\n%s\n\n" % (s, b"-" * encoding.colwidth(s)) def subsubsubsection(s): return b"%s\n%s\n\n" % (s, b"." * encoding.colwidth(s)) def subsubsubsubsection(s): return b"%s\n%s\n\n" % (s, b"'" * encoding.colwidth(s)) def replace(text, substs): ''' Apply a list of (find, replace) pairs to a text. >>> replace(b"foo bar", [(b'f', b'F'), (b'b', b'B')]) 'Foo Bar' >>> encoding.encoding = b'latin1' >>> replace(b'\\x81\\\\', [(b'\\\\', b'/')]) '\\x81/' >>> encoding.encoding = b'shiftjis' >>> replace(b'\\x81\\\\', [(b'\\\\', b'/')]) '\\x81\\\\' ''' # some character encodings (cp932 for Japanese, at least) use # ASCII characters other than control/alphabet/digit as a part of # multi-bytes characters, so direct replacing with such characters # on strings in local encoding causes invalid byte sequences. utext = text.decode(pycompat.sysstr(encoding.encoding)) for f, t in substs: utext = utext.replace(f.decode("ascii"), t.decode("ascii")) return utext.encode(pycompat.sysstr(encoding.encoding)) _blockre = re.compile(br"\n(?:\s*\n)+") def findblocks(text): """Find continuous blocks of lines in text. Returns a list of dictionaries representing the blocks. Each block has an 'indent' field and a 'lines' field. """ blocks = [] for b in _blockre.split(text.lstrip(b'\n').rstrip()): lines = b.splitlines() if lines: indent = min((len(l) - len(l.lstrip())) for l in lines) lines = [l[indent:] for l in lines] blocks.append({b'indent': indent, b'lines': lines}) return blocks def findliteralblocks(blocks): """Finds literal blocks and adds a 'type' field to the blocks. Literal blocks are given the type 'literal', all other blocks are given type the 'paragraph'. """ i = 0 while i < len(blocks): # Searching for a block that looks like this: # # +------------------------------+ # | paragraph | # | (ends with "::") | # +------------------------------+ # +---------------------------+ # | indented literal block | # +---------------------------+ blocks[i][b'type'] = b'paragraph' if blocks[i][b'lines'][-1].endswith(b'::') and i + 1 < len(blocks): indent = blocks[i][b'indent'] adjustment = blocks[i + 1][b'indent'] - indent if blocks[i][b'lines'] == [b'::']: # Expanded form: remove block del blocks[i] i -= 1 elif blocks[i][b'lines'][-1].endswith(b' ::'): # Partially minimized form: remove space and both # colons. blocks[i][b'lines'][-1] = blocks[i][b'lines'][-1][:-3] elif ( len(blocks[i][b'lines']) == 1 and blocks[i][b'lines'][0].lstrip(b' ').startswith(b'.. ') and blocks[i][b'lines'][0].find(b' ', 3) == -1 ): # directive on its own line, not a literal block i += 1 continue else: # Fully minimized form: remove just one colon. blocks[i][b'lines'][-1] = blocks[i][b'lines'][-1][:-1] # List items are formatted with a hanging indent. We must # correct for this here while we still have the original # information on the indentation of the subsequent literal # blocks available. m = _bulletre.match(blocks[i][b'lines'][0]) if m: indent += m.end() adjustment -= m.end() # Mark the following indented blocks. while i + 1 < len(blocks) and blocks[i + 1][b'indent'] > indent: blocks[i + 1][b'type'] = b'literal' blocks[i + 1][b'indent'] -= adjustment i += 1 i += 1 return blocks _bulletre = re.compile(br'(\*|-|[0-9A-Za-z]+\.|\(?[0-9A-Za-z]+\)|\|) ') _optionre = re.compile( br'^(-([a-zA-Z0-9]), )?(--[a-z0-9-]+)' br'((.*) +)(.*)$' ) _fieldre = re.compile(br':(?![: ])([^:]*)(?<! ):[ ]+(.*)') _definitionre = re.compile(br'[^ ]') _tablere = re.compile(br'(=+\s+)*=+') def splitparagraphs(blocks): """Split paragraphs into lists.""" # Tuples with (list type, item regexp, single line items?). Order # matters: definition lists has the least specific regexp and must # come last. listtypes = [ (b'bullet', _bulletre, True), (b'option', _optionre, True), (b'field', _fieldre, True), (b'definition', _definitionre, False), ] def match(lines, i, itemre, singleline): """Does itemre match an item at line i? A list item can be followed by an indented line or another list item (but only if singleline is True). """ line1 = lines[i] line2 = i + 1 < len(lines) and lines[i + 1] or b'' if not itemre.match(line1): return False if singleline: return line2 == b'' or line2[0:1] == b' ' or itemre.match(line2) else: return line2.startswith(b' ') i = 0 while i < len(blocks): if blocks[i][b'type'] == b'paragraph': lines = blocks[i][b'lines'] for type, itemre, singleline in listtypes: if match(lines, 0, itemre, singleline): items = [] for j, line in enumerate(lines): if match(lines, j, itemre, singleline): items.append( { b'type': type, b'lines': [], b'indent': blocks[i][b'indent'], } ) items[-1][b'lines'].append(line) blocks[i : i + 1] = items break i += 1 return blocks _fieldwidth = 14 def updatefieldlists(blocks): """Find key for field lists.""" i = 0 while i < len(blocks): if blocks[i][b'type'] != b'field': i += 1 continue j = i while j < len(blocks) and blocks[j][b'type'] == b'field': m = _fieldre.match(blocks[j][b'lines'][0]) key, rest = m.groups() blocks[j][b'lines'][0] = rest blocks[j][b'key'] = key j += 1 i = j + 1 return blocks def updateoptionlists(blocks): i = 0 while i < len(blocks): if blocks[i][b'type'] != b'option': i += 1 continue optstrwidth = 0 j = i while j < len(blocks) and blocks[j][b'type'] == b'option': m = _optionre.match(blocks[j][b'lines'][0]) shortoption = m.group(2) group3 = m.group(3) longoption = group3[2:].strip() desc = m.group(6).strip() longoptionarg = m.group(5).strip() blocks[j][b'lines'][0] = desc noshortop = b'' if not shortoption: noshortop = b' ' opt = b"%s%s" % ( shortoption and b"-%s " % shortoption or b'', b"%s--%s %s" % (noshortop, longoption, longoptionarg), ) opt = opt.rstrip() blocks[j][b'optstr'] = opt optstrwidth = max(optstrwidth, encoding.colwidth(opt)) j += 1 for block in blocks[i:j]: block[b'optstrwidth'] = optstrwidth i = j + 1 return blocks def prunecontainers(blocks, keep): """Prune unwanted containers. The blocks must have a 'type' field, i.e., they should have been run through findliteralblocks first. """ pruned = [] i = 0 while i + 1 < len(blocks): # Searching for a block that looks like this: # # +-------+---------------------------+ # | ".. container ::" type | # +---+ | # | blocks | # +-------------------------------+ if blocks[i][b'type'] == b'paragraph' and blocks[i][b'lines'][ 0 ].startswith(b'.. container::'): indent = blocks[i][b'indent'] adjustment = blocks[i + 1][b'indent'] - indent containertype = blocks[i][b'lines'][0][15:] prune = True for c in keep: if c in containertype.split(b'.'): prune = False if prune: pruned.append(containertype) # Always delete "..container:: type" block del blocks[i] j = i i -= 1 while j < len(blocks) and blocks[j][b'indent'] > indent: if prune: del blocks[j] else: blocks[j][b'indent'] -= adjustment j += 1 i += 1 return blocks, pruned _sectionre = re.compile(br"""^([-=`:.'"~^_*+#])\1+$""") def findtables(blocks): '''Find simple tables Only simple one-line table elements are supported ''' for block in blocks: # Searching for a block that looks like this: # # === ==== === # A B C # === ==== === <- optional # 1 2 3 # x y z # === ==== === if ( block[b'type'] == b'paragraph' and len(block[b'lines']) > 2 and _tablere.match(block[b'lines'][0]) and block[b'lines'][0] == block[b'lines'][-1] ): block[b'type'] = b'table' block[b'header'] = False div = block[b'lines'][0] # column markers are ASCII so we can calculate column # position in bytes columns = [ x for x in pycompat.xrange(len(div)) if div[x : x + 1] == b'=' and (x == 0 or div[x - 1 : x] == b' ') ] rows = [] for l in block[b'lines'][1:-1]: if l == div: block[b'header'] = True continue row = [] # we measure columns not in bytes or characters but in # colwidth which makes things tricky pos = columns[0] # leading whitespace is bytes for n, start in enumerate(columns): if n + 1 < len(columns): width = columns[n + 1] - start v = encoding.getcols(l, pos, width) # gather columns pos += len(v) # calculate byte position of end row.append(v.strip()) else: row.append(l[pos:].strip()) rows.append(row) block[b'table'] = rows return blocks def findsections(blocks): """Finds sections. The blocks must have a 'type' field, i.e., they should have been run through findliteralblocks first. """ for block in blocks: # Searching for a block that looks like this: # # +------------------------------+ # | Section title | # | ------------- | # +------------------------------+ if ( block[b'type'] == b'paragraph' and len(block[b'lines']) == 2 and encoding.colwidth(block[b'lines'][0]) == len(block[b'lines'][1]) and _sectionre.match(block[b'lines'][1]) ): block[b'underline'] = block[b'lines'][1][0:1] block[b'type'] = b'section' del block[b'lines'][1] return blocks def inlineliterals(blocks): substs = [(b'``', b'"')] for b in blocks: if b[b'type'] in (b'paragraph', b'section'): b[b'lines'] = [replace(l, substs) for l in b[b'lines']] return blocks def hgrole(blocks): substs = [(b':hg:`', b"'hg "), (b'`', b"'")] for b in blocks: if b[b'type'] in (b'paragraph', b'section'): # Turn :hg:`command` into "hg command". This also works # when there is a line break in the command and relies on # the fact that we have no stray back-quotes in the input # (run the blocks through inlineliterals first). b[b'lines'] = [replace(l, substs) for l in b[b'lines']] return blocks def addmargins(blocks): """Adds empty blocks for vertical spacing. This groups bullets, options, and definitions together with no vertical space between them, and adds an empty block between all other blocks. """ i = 1 while i < len(blocks): if blocks[i][b'type'] == blocks[i - 1][b'type'] and blocks[i][ b'type' ] in (b'bullet', b'option', b'field',): i += 1 elif not blocks[i - 1][b'lines']: # no lines in previous block, do not separate i += 1 else: blocks.insert( i, {b'lines': [b''], b'indent': 0, b'type': b'margin'} ) i += 2 return blocks def prunecomments(blocks): """Remove comments.""" i = 0 while i < len(blocks): b = blocks[i] if b[b'type'] == b'paragraph' and ( b[b'lines'][0].startswith(b'.. ') or b[b'lines'] == [b'..'] ): del blocks[i] if i < len(blocks) and blocks[i][b'type'] == b'margin': del blocks[i] else: i += 1 return blocks def findadmonitions(blocks, admonitions=None): """ Makes the type of the block an admonition block if the first line is an admonition directive """ admonitions = admonitions or _admonitiontitles.keys() admonitionre = re.compile( br'\.\. (%s)::' % b'|'.join(sorted(admonitions)), flags=re.IGNORECASE ) i = 0 while i < len(blocks): m = admonitionre.match(blocks[i][b'lines'][0]) if m: blocks[i][b'type'] = b'admonition' admonitiontitle = blocks[i][b'lines'][0][3 : m.end() - 2].lower() firstline = blocks[i][b'lines'][0][m.end() + 1 :] if firstline: blocks[i][b'lines'].insert(1, b' ' + firstline) blocks[i][b'admonitiontitle'] = admonitiontitle del blocks[i][b'lines'][0] i = i + 1 return blocks _admonitiontitles = { b'attention': _(b'Attention:'), b'caution': _(b'Caution:'), b'danger': _(b'!Danger!'), b'error': _(b'Error:'), b'hint': _(b'Hint:'), b'important': _(b'Important:'), b'note': _(b'Note:'), b'tip': _(b'Tip:'), b'warning': _(b'Warning!'), } def formatoption(block, width): desc = b' '.join(map(bytes.strip, block[b'lines'])) colwidth = encoding.colwidth(block[b'optstr']) usablewidth = width - 1 hanging = block[b'optstrwidth'] initindent = b'%s%s ' % (block[b'optstr'], b' ' * ((hanging - colwidth))) hangindent = b' ' * (encoding.colwidth(initindent) + 1) return b' %s\n' % ( stringutil.wrap( desc, usablewidth, initindent=initindent, hangindent=hangindent ) ) def formatblock(block, width): """Format a block according to width.""" if width <= 0: width = 78 indent = b' ' * block[b'indent'] if block[b'type'] == b'admonition': admonition = _admonitiontitles[block[b'admonitiontitle']] if not block[b'lines']: return indent + admonition + b'\n' hang = len(block[b'lines'][-1]) - len(block[b'lines'][-1].lstrip()) defindent = indent + hang * b' ' text = b' '.join(map(bytes.strip, block[b'lines'])) return b'%s\n%s\n' % ( indent + admonition, stringutil.wrap( text, width=width, initindent=defindent, hangindent=defindent ), ) if block[b'type'] == b'margin': return b'\n' if block[b'type'] == b'literal': indent += b' ' return indent + (b'\n' + indent).join(block[b'lines']) + b'\n' if block[b'type'] == b'section': underline = encoding.colwidth(block[b'lines'][0]) * block[b'underline'] return b"%s%s\n%s%s\n" % (indent, block[b'lines'][0], indent, underline) if block[b'type'] == b'table': table = block[b'table'] # compute column widths widths = [max([encoding.colwidth(e) for e in c]) for c in zip(*table)] text = b'' span = sum(widths) + len(widths) - 1 indent = b' ' * block[b'indent'] hang = b' ' * (len(indent) + span - widths[-1]) for row in table: l = [] for w, v in zip(widths, row): pad = b' ' * (w - encoding.colwidth(v)) l.append(v + pad) l = b' '.join(l) l = stringutil.wrap( l, width=width, initindent=indent, hangindent=hang ) if not text and block[b'header']: text = l + b'\n' + indent + b'-' * (min(width, span)) + b'\n' else: text += l + b"\n" return text if block[b'type'] == b'definition': term = indent + block[b'lines'][0] hang = len(block[b'lines'][-1]) - len(block[b'lines'][-1].lstrip()) defindent = indent + hang * b' ' text = b' '.join(map(bytes.strip, block[b'lines'][1:])) return b'%s\n%s\n' % ( term, stringutil.wrap( text, width=width, initindent=defindent, hangindent=defindent ), ) subindent = indent if block[b'type'] == b'bullet': if block[b'lines'][0].startswith(b'| '): # Remove bullet for line blocks and add no extra # indentation. block[b'lines'][0] = block[b'lines'][0][2:] else: m = _bulletre.match(block[b'lines'][0]) subindent = indent + m.end() * b' ' elif block[b'type'] == b'field': key = block[b'key'] subindent = indent + _fieldwidth * b' ' if len(key) + 2 > _fieldwidth: # key too large, use full line width key = key.ljust(width) else: # key fits within field width key = key.ljust(_fieldwidth) block[b'lines'][0] = key + block[b'lines'][0] elif block[b'type'] == b'option': return formatoption(block, width) text = b' '.join(map(bytes.strip, block[b'lines'])) return ( stringutil.wrap( text, width=width, initindent=indent, hangindent=subindent ) + b'\n' ) def formathtml(blocks): """Format RST blocks as HTML""" out = [] headernest = b'' listnest = [] def escape(s): return url.escape(s, True) def openlist(start, level): if not listnest or listnest[-1][0] != start: listnest.append((start, level)) out.append(b'<%s>\n' % start) blocks = [b for b in blocks if b[b'type'] != b'margin'] for pos, b in enumerate(blocks): btype = b[b'type'] level = b[b'indent'] lines = b[b'lines'] if btype == b'admonition': admonition = escape(_admonitiontitles[b[b'admonitiontitle']]) text = escape(b' '.join(map(bytes.strip, lines))) out.append(b'<p>\n<b>%s</b> %s\n</p>\n' % (admonition, text)) elif btype == b'paragraph': out.append(b'<p>\n%s\n</p>\n' % escape(b'\n'.join(lines))) elif btype == b'margin': pass elif btype == b'literal': out.append(b'<pre>\n%s\n</pre>\n' % escape(b'\n'.join(lines))) elif btype == b'section': i = b[b'underline'] if i not in headernest: headernest += i level = headernest.index(i) + 1 out.append(b'<h%d>%s</h%d>\n' % (level, escape(lines[0]), level)) elif btype == b'table': table = b[b'table'] out.append(b'<table>\n') for row in table: out.append(b'<tr>') for v in row: out.append(b'<td>') out.append(escape(v)) out.append(b'</td>') out.append(b'\n') out.pop() out.append(b'</tr>\n') out.append(b'</table>\n') elif btype == b'definition': openlist(b'dl', level) term = escape(lines[0]) text = escape(b' '.join(map(bytes.strip, lines[1:]))) out.append(b' <dt>%s\n <dd>%s\n' % (term, text)) elif btype == b'bullet': bullet, head = lines[0].split(b' ', 1) if bullet in (b'*', b'-'): openlist(b'ul', level) else: openlist(b'ol', level) out.append(b' <li> %s\n' % escape(b' '.join([head] + lines[1:]))) elif btype == b'field': openlist(b'dl', level) key = escape(b[b'key']) text = escape(b' '.join(map(bytes.strip, lines))) out.append(b' <dt>%s\n <dd>%s\n' % (key, text)) elif btype == b'option': openlist(b'dl', level) opt = escape(b[b'optstr']) desc = escape(b' '.join(map(bytes.strip, lines))) out.append(b' <dt>%s\n <dd>%s\n' % (opt, desc)) # close lists if indent level of next block is lower if listnest: start, level = listnest[-1] if pos == len(blocks) - 1: out.append(b'</%s>\n' % start) listnest.pop() else: nb = blocks[pos + 1] ni = nb[b'indent'] if ni < level or ( ni == level and nb[b'type'] not in b'definition bullet field option' ): out.append(b'</%s>\n' % start) listnest.pop() return b''.join(out) def parse(text, indent=0, keep=None, admonitions=None): """Parse text into a list of blocks""" blocks = findblocks(text) for b in blocks: b[b'indent'] += indent blocks = findliteralblocks(blocks) blocks = findtables(blocks) blocks, pruned = prunecontainers(blocks, keep or []) blocks = findsections(blocks) blocks = inlineliterals(blocks) blocks = hgrole(blocks) blocks = splitparagraphs(blocks) blocks = updatefieldlists(blocks) blocks = updateoptionlists(blocks) blocks = findadmonitions(blocks, admonitions=admonitions) blocks = addmargins(blocks) blocks = prunecomments(blocks) return blocks, pruned def formatblocks(blocks, width): text = b''.join(formatblock(b, width) for b in blocks) return text def formatplain(blocks, width): """Format parsed blocks as plain text""" return b''.join(formatblock(b, width) for b in blocks) def format(text, width=80, indent=0, keep=None, style=b'plain', section=None): """Parse and format the text according to width.""" blocks, pruned = parse(text, indent, keep or []) if section: blocks = filtersections(blocks, section) if style == b'html': return formathtml(blocks) else: return formatplain(blocks, width=width) def filtersections(blocks, section): """Select parsed blocks under the specified section The section name is separated by a dot, and matches the suffix of the full section path. """ parents = [] sections = _getsections(blocks) blocks = [] i = 0 lastparents = [] synthetic = [] collapse = True while i < len(sections): path, nest, b = sections[i] del parents[nest:] parents.append(i) if path == section or path.endswith(b'.' + section): if lastparents != parents: llen = len(lastparents) plen = len(parents) if llen and llen != plen: collapse = False s = [] for j in pycompat.xrange(3, plen - 1): parent = parents[j] if j >= llen or lastparents[j] != parent: s.append(len(blocks)) sec = sections[parent][2] blocks.append(sec[0]) blocks.append(sec[-1]) if s: synthetic.append(s) lastparents = parents[:] blocks.extend(b) ## Also show all subnested sections while i + 1 < len(sections) and sections[i + 1][1] > nest: i += 1 blocks.extend(sections[i][2]) i += 1 if collapse: synthetic.reverse() for s in synthetic: path = [blocks[syn][b'lines'][0] for syn in s] real = s[-1] + 2 realline = blocks[real][b'lines'] realline[0] = b'"%s"' % b'.'.join(path + [realline[0]]).replace( b'"', b'' ) del blocks[s[0] : real] return blocks def _getsections(blocks): '''return a list of (section path, nesting level, blocks) tuples''' nest = b"" names = () secs = [] def getname(b): if b[b'type'] == b'field': x = b[b'key'] else: x = b[b'lines'][0] x = encoding.lower(x).strip(b'"') if b'(' in x: x = x.split(b'(')[0] return x for b in blocks: if b[b'type'] == b'section': i = b[b'underline'] if i not in nest: nest += i level = nest.index(i) + 1 nest = nest[:level] names = names[:level] + (getname(b),) secs.append((b'.'.join(names), level, [b])) elif b[b'type'] in (b'definition', b'field'): i = b' ' if i not in nest: nest += i level = nest.index(i) + 1 nest = nest[:level] for i in range(1, len(secs) + 1): sec = secs[-i] if sec[1] < level: break siblings = [a for a in sec[2] if a[b'type'] == b'definition'] if siblings: siblingindent = siblings[-1][b'indent'] indent = b[b'indent'] if siblingindent < indent: level += 1 break elif siblingindent == indent: level = sec[1] break names = names[:level] + (getname(b),) secs.append((b'.'.join(names), level, [b])) else: if not secs: # add an initial empty section secs = [(b'', 0, [])] if b[b'type'] != b'margin': pointer = 1 bindent = b[b'indent'] while pointer < len(secs): section = secs[-pointer][2][0] if section[b'type'] != b'margin': sindent = section[b'indent'] if len(section[b'lines']) > 1: sindent += len(section[b'lines'][1]) - len( section[b'lines'][1].lstrip(b' ') ) if bindent >= sindent: break pointer += 1 if pointer > 1: blevel = secs[-pointer][1] if section[b'type'] != b[b'type']: blevel += 1 secs.append((b'', blevel, [])) secs[-1][2].append(b) return secs def maketable(data, indent=0, header=False): '''Generate an RST table for the given table data as a list of lines''' widths = [max(encoding.colwidth(e) for e in c) for c in zip(*data)] indent = b' ' * indent div = indent + b' '.join(b'=' * w for w in widths) + b'\n' out = [div] for row in data: l = [] for w, v in zip(widths, row): if b'\n' in v: # only remove line breaks and indentation, long lines are # handled by the next tool v = b' '.join(e.lstrip() for e in v.split(b'\n')) pad = b' ' * (w - encoding.colwidth(v)) l.append(v + pad) out.append(indent + b' '.join(l) + b"\n") if header and len(data) > 1: out.insert(2, div) out.append(div) return out