Mercurial > hg
view mercurial/fancyopts.py @ 42044:bb271ec2fbfb
compression: introduce a `storage.revlog.zstd.level` configuration
This option control the zstd compression level used when compressing revlog
chunk. The usage of zstd for revlog compression has not graduated from
experimental yet, but we intend to fix that soon.
The option name for the compression level is more straight forward to pick, so
this changesets comes first. Having a dedicated option for each compression
engine is useful because they don't support the same range of values.
I ran the same measurement as for the zlib compression level (in the parent
changesets). The variation in repository size is stay mostly in the same (small)
range. The "read/write" performance see smallish variation, but are overall much
better than zlib. Write performance show the same tend of having better write
performance for when reaching high-end compression.
Again, we don't intend to change the default zstd compression level (currently:
3) in this series. However this is worth investigating in the future.
The Performance comparison of zlib vs zstd is quite impressive. The repository
size stay in the same range, but the performance are much better in all
situations.
Comparison summary
==================
We are looking at:
- performance range for zlib
- performance range for zstd
- comparison of default zstd (level-3) to default zlib (level 6)
- comparison of the slowest zstd time to the fastest zlib time
Read performance:
-----------------
| zlib | zstd | cmp | f2s
mercurial | 0.170159 - 0.189219 | 0.144127 - 0.149624 | 80% | 88%
pypy | 2.679217 - 2.768691 | 1.532317 - 1.705044 | 60% | 63%
netbeans | 122.477027 - 141.620281 | 72.996346 - 89.731560 | 58% | 73%
mozilla | 147.867662 - 170.572118 | 91.700995 - 105.853099 | 56% | 71%
Write performance:
------------------
| zlib | zstd | cmp | f2s
mercurial | 53.250304 - 56.2936129 | 40.877025 - 45.677286 | 75% | 86%
pypy | 460.721984 - 476.589918 | 270.545409 - 301.002219 | 63% | 65%
netbeans | 520.560316 - 715.930400 | 370.356311 - 428.329652 | 55% | 82%
mozilla | 739.803002 - 987.056093 | 505.152906 - 591.930683 | 57% | 80%
Raw data
--------
repo alg lvl .hg/store size 00manifest.d read write
mercurial zlib 1 49,402,813 5,963,475 0.170159 53.250304
mercurial zlib 6 47,197,397 5,875,730 0.182820 56.264320
mercurial zlib 9 47,121,596 5,849,781 0.189219 56.293612
mercurial zstd 1 49,737,084 5,966,355 0.144127 40.877025
mercurial zstd 3 48,961,867 5,895,208 0.146376 42.268142
mercurial zstd 5 48,200,592 5,938,676 0.149624 43.162875
mercurial zstd 10 47,833,520 5,913,353 0.145185 44.012489
mercurial zstd 15 47,314,604 5,728,679 0.147686 45.677286
mercurial zstd 20 47,330,502 5,830,539 0.145789 45.025407
mercurial zstd 22 47,330,076 5,830,539 0.143996 44.690460
pypy zlib 1 370,830,572 28,462,425 2.679217 460.721984
pypy zlib 6 340,112,317 27,648,747 2.768691 467.537158
pypy zlib 9 338,360,736 27,639,003 2.763495 476.589918
pypy zstd 1 362,377,479 27,916,214 1.532317 270.545409
pypy zstd 3 354,137,693 27,905,988 1.686718 294.951509
pypy zstd 5 342,640,043 27,655,774 1.705044 301.002219
pypy zstd 10 334,224,327 27,164,493 1.567287 285.186239
pypy zstd 15 329,000,363 26,645,965 1.637729 299.561332
pypy zstd 20 324,534,039 26,199,547 1.526813 302.149827
pypy zstd 22 324,530,595 26,198,932 1.525718 307.821218
netbeans zlib 1 1,281,847,810 165,495,457 122.477027 520.560316
netbeans zlib 6 1,205,284,353 159,161,207 139.876147 715.930400
netbeans zlib 9 1,197,135,671 155,034,586 141.620281 678.297064
netbeans zstd 1 1,259,581,737 160,840,613 72.996346 370.356311
netbeans zstd 3 1,232,978,122 157,691,551 81.622317 396.733087
netbeans zstd 5 1,208,034,075 160,246,880 83.080549 364.342626
netbeans zstd 10 1,188,624,176 156,083,417 79.323935 403.594602
netbeans zstd 15 1,176,973,589 153,859,477 89.731560 428.329652
netbeans zstd 20 1,162,958,258 151,147,535 82.842667 392.335349
netbeans zstd 22 1,162,707,029 151,150,220 82.565695 402.840655
mozilla zlib 1 2,775,497,186 298,527,987 147.867662 751.263721
mozilla zlib 6 2,596,856,420 286,597,671 170.572118 987.056093
mozilla zlib 9 2,587,542,494 287,018,264 163.622338 739.803002
mozilla zstd 1 2,723,159,348 286,617,532 91.700995 570.042751
mozilla zstd 3 2,665,055,001 286,152,013 95.240155 561.412805
mozilla zstd 5 2,607,819,817 288,060,030 101.978048 505.152906
mozilla zstd 10 2,558,761,085 283,967,648 104.113481 497.771202
mozilla zstd 15 2,526,216,060 275,581,300 105.853099 591.930683
mozilla zstd 20 2,485,114,806 266,478,859 95.268795 576.515389
mozilla zstd 22 2,484,869,080 266,456,505 94.429282 572.785537
author | Pierre-Yves David <pierre-yves.david@octobus.net> |
---|---|
date | Wed, 27 Mar 2019 18:35:59 +0100 |
parents | 39e5e346eba7 |
children | 2372284d9457 |
line wrap: on
line source
# fancyopts.py - better command line parsing # # Copyright 2005-2009 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. from __future__ import absolute_import import abc import functools from .i18n import _ from . import ( error, pycompat, ) # Set of flags to not apply boolean negation logic on nevernegate = { # avoid --no-noninteractive 'noninteractive', # These two flags are special because they cause hg to do one # thing and then exit, and so aren't suitable for use in things # like aliases anyway. 'help', 'version', } def _earlyoptarg(arg, shortlist, namelist): """Check if the given arg is a valid unabbreviated option Returns (flag_str, has_embedded_value?, embedded_value, takes_value?) >>> def opt(arg): ... return _earlyoptarg(arg, b'R:q', [b'cwd=', b'debugger']) long form: >>> opt(b'--cwd') ('--cwd', False, '', True) >>> opt(b'--cwd=') ('--cwd', True, '', True) >>> opt(b'--cwd=foo') ('--cwd', True, 'foo', True) >>> opt(b'--debugger') ('--debugger', False, '', False) >>> opt(b'--debugger=') # invalid but parsable ('--debugger', True, '', False) short form: >>> opt(b'-R') ('-R', False, '', True) >>> opt(b'-Rfoo') ('-R', True, 'foo', True) >>> opt(b'-q') ('-q', False, '', False) >>> opt(b'-qfoo') # invalid but parsable ('-q', True, 'foo', False) unknown or invalid: >>> opt(b'--unknown') ('', False, '', False) >>> opt(b'-u') ('', False, '', False) >>> opt(b'-ufoo') ('', False, '', False) >>> opt(b'--') ('', False, '', False) >>> opt(b'-') ('', False, '', False) >>> opt(b'-:') ('', False, '', False) >>> opt(b'-:foo') ('', False, '', False) """ if arg.startswith('--'): flag, eq, val = arg.partition('=') if flag[2:] in namelist: return flag, bool(eq), val, False if flag[2:] + '=' in namelist: return flag, bool(eq), val, True elif arg.startswith('-') and arg != '-' and not arg.startswith('-:'): flag, val = arg[:2], arg[2:] i = shortlist.find(flag[1:]) if i >= 0: return flag, bool(val), val, shortlist.startswith(':', i + 1) return '', False, '', False def earlygetopt(args, shortlist, namelist, gnu=False, keepsep=False): """Parse options like getopt, but ignores unknown options and abbreviated forms If gnu=False, this stops processing options as soon as a non/unknown-option argument is encountered. Otherwise, option and non-option arguments may be intermixed, and unknown-option arguments are taken as non-option. If keepsep=True, '--' won't be removed from the list of arguments left. This is useful for stripping early options from a full command arguments. >>> def get(args, gnu=False, keepsep=False): ... return earlygetopt(args, b'R:q', [b'cwd=', b'debugger'], ... gnu=gnu, keepsep=keepsep) default parsing rules for early options: >>> get([b'x', b'--cwd', b'foo', b'-Rbar', b'-q', b'y'], gnu=True) ([('--cwd', 'foo'), ('-R', 'bar'), ('-q', '')], ['x', 'y']) >>> get([b'x', b'--cwd=foo', b'y', b'-R', b'bar', b'--debugger'], gnu=True) ([('--cwd', 'foo'), ('-R', 'bar'), ('--debugger', '')], ['x', 'y']) >>> get([b'--unknown', b'--cwd=foo', b'--', '--debugger'], gnu=True) ([('--cwd', 'foo')], ['--unknown', '--debugger']) restricted parsing rules (early options must come first): >>> get([b'--cwd', b'foo', b'-Rbar', b'x', b'-q', b'y'], gnu=False) ([('--cwd', 'foo'), ('-R', 'bar')], ['x', '-q', 'y']) >>> get([b'--cwd=foo', b'x', b'y', b'-R', b'bar', b'--debugger'], gnu=False) ([('--cwd', 'foo')], ['x', 'y', '-R', 'bar', '--debugger']) >>> get([b'--unknown', b'--cwd=foo', b'--', '--debugger'], gnu=False) ([], ['--unknown', '--cwd=foo', '--', '--debugger']) stripping early options (without loosing '--'): >>> get([b'x', b'-Rbar', b'--', '--debugger'], gnu=True, keepsep=True)[1] ['x', '--', '--debugger'] last argument: >>> get([b'--cwd']) ([], ['--cwd']) >>> get([b'--cwd=foo']) ([('--cwd', 'foo')], []) >>> get([b'-R']) ([], ['-R']) >>> get([b'-Rbar']) ([('-R', 'bar')], []) >>> get([b'-q']) ([('-q', '')], []) >>> get([b'-q', b'--']) ([('-q', '')], []) '--' may be a value: >>> get([b'-R', b'--', b'x']) ([('-R', '--')], ['x']) >>> get([b'--cwd', b'--', b'x']) ([('--cwd', '--')], ['x']) value passed to bool options: >>> get([b'--debugger=foo', b'x']) ([], ['--debugger=foo', 'x']) >>> get([b'-qfoo', b'x']) ([], ['-qfoo', 'x']) short option isn't separated with '=': >>> get([b'-R=bar']) ([('-R', '=bar')], []) ':' may be in shortlist, but shouldn't be taken as an option letter: >>> get([b'-:', b'y']) ([], ['-:', 'y']) '-' is a valid non-option argument: >>> get([b'-', b'y']) ([], ['-', 'y']) """ parsedopts = [] parsedargs = [] pos = 0 while pos < len(args): arg = args[pos] if arg == '--': pos += not keepsep break flag, hasval, val, takeval = _earlyoptarg(arg, shortlist, namelist) if not hasval and takeval and pos + 1 >= len(args): # missing last argument break if not flag or hasval and not takeval: # non-option argument or -b/--bool=INVALID_VALUE if gnu: parsedargs.append(arg) pos += 1 else: break elif hasval == takeval: # -b/--bool or -s/--str=VALUE parsedopts.append((flag, val)) pos += 1 else: # -s/--str VALUE parsedopts.append((flag, args[pos + 1])) pos += 2 parsedargs.extend(args[pos:]) return parsedopts, parsedargs class customopt(object): """Manage defaults and mutations for any type of opt.""" __metaclass__ = abc.ABCMeta def __init__(self, defaultvalue): self._defaultvalue = defaultvalue def _isboolopt(self): return False def getdefaultvalue(self): """Returns the default value for this opt. Subclasses should override this to return a new value if the value type is mutable.""" return self._defaultvalue @abc.abstractmethod def newstate(self, oldstate, newparam, abort): """Adds newparam to oldstate and returns the new state. On failure, abort can be called with a string error message.""" class _simpleopt(customopt): def _isboolopt(self): return isinstance(self._defaultvalue, (bool, type(None))) def newstate(self, oldstate, newparam, abort): return newparam class _callableopt(customopt): def __init__(self, callablefn): self.callablefn = callablefn super(_callableopt, self).__init__(None) def newstate(self, oldstate, newparam, abort): return self.callablefn(newparam) class _listopt(customopt): def getdefaultvalue(self): return self._defaultvalue[:] def newstate(self, oldstate, newparam, abort): oldstate.append(newparam) return oldstate class _intopt(customopt): def newstate(self, oldstate, newparam, abort): try: return int(newparam) except ValueError: abort(_('expected int')) def _defaultopt(default): """Returns a default opt implementation, given a default value.""" if isinstance(default, customopt): return default elif callable(default): return _callableopt(default) elif isinstance(default, list): return _listopt(default[:]) elif type(default) is type(1): return _intopt(default) else: return _simpleopt(default) def fancyopts(args, options, state, gnu=False, early=False, optaliases=None): """ read args, parse options, and store options in state each option is a tuple of: short option or '' long option default value description option value label(optional) option types include: boolean or none - option sets variable in state to true string - parameter string is stored in state list - parameter string is added to a list integer - parameter strings is stored as int function - call function with parameter customopt - subclass of 'customopt' optaliases is a mapping from a canonical option name to a list of additional long options. This exists for preserving backward compatibility of early options. If we want to use it extensively, please consider moving the functionality to the options table (e.g separate long options by '|'.) non-option args are returned """ if optaliases is None: optaliases = {} namelist = [] shortlist = '' argmap = {} defmap = {} negations = {} alllong = set(o[1] for o in options) for option in options: if len(option) == 5: short, name, default, comment, dummy = option else: short, name, default, comment = option # convert opts to getopt format onames = [name] onames.extend(optaliases.get(name, [])) name = name.replace('-', '_') argmap['-' + short] = name for n in onames: argmap['--' + n] = name defmap[name] = _defaultopt(default) # copy defaults to state state[name] = defmap[name].getdefaultvalue() # does it take a parameter? if not defmap[name]._isboolopt(): if short: short += ':' onames = [n + '=' for n in onames] elif name not in nevernegate: for n in onames: if n.startswith('no-'): insert = n[3:] else: insert = 'no-' + n # backout (as a practical example) has both --commit and # --no-commit options, so we don't want to allow the # negations of those flags. if insert not in alllong: assert ('--' + n) not in negations negations['--' + insert] = '--' + n namelist.append(insert) if short: shortlist += short if name: namelist.extend(onames) # parse arguments if early: parse = functools.partial(earlygetopt, gnu=gnu) elif gnu: parse = pycompat.gnugetoptb else: parse = pycompat.getoptb opts, args = parse(args, shortlist, namelist) # transfer result to state for opt, val in opts: boolval = True negation = negations.get(opt, False) if negation: opt = negation boolval = False name = argmap[opt] obj = defmap[name] if obj._isboolopt(): state[name] = boolval else: def abort(s): raise error.Abort(_('invalid value %r for option %s, %s') % (pycompat.maybebytestr(val), opt, s)) state[name] = defmap[name].newstate(state[name], val, abort) # return unparsed args return args