Mercurial > hg
view mercurial/revsetlang.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 | ddb174511f1b |
children | 2372284d9457 |
line wrap: on
line source
# revsetlang.py - parser, tokenizer and utility for revision set language # # Copyright 2010 Matt Mackall <mpm@selenic.com> # # 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 string from .i18n import _ from . import ( error, node, parser, pycompat, smartset, util, ) from .utils import ( stringutil, ) elements = { # token-type: binding-strength, primary, prefix, infix, suffix "(": (21, None, ("group", 1, ")"), ("func", 1, ")"), None), "[": (21, None, None, ("subscript", 1, "]"), None), "#": (21, None, None, ("relation", 21), None), "##": (20, None, None, ("_concat", 20), None), "~": (18, None, None, ("ancestor", 18), None), "^": (18, None, None, ("parent", 18), "parentpost"), "-": (5, None, ("negate", 19), ("minus", 5), None), "::": (17, "dagrangeall", ("dagrangepre", 17), ("dagrange", 17), "dagrangepost"), "..": (17, "dagrangeall", ("dagrangepre", 17), ("dagrange", 17), "dagrangepost"), ":": (15, "rangeall", ("rangepre", 15), ("range", 15), "rangepost"), "not": (10, None, ("not", 10), None, None), "!": (10, None, ("not", 10), None, None), "and": (5, None, None, ("and", 5), None), "&": (5, None, None, ("and", 5), None), "%": (5, None, None, ("only", 5), "onlypost"), "or": (4, None, None, ("or", 4), None), "|": (4, None, None, ("or", 4), None), "+": (4, None, None, ("or", 4), None), "=": (3, None, None, ("keyvalue", 3), None), ",": (2, None, None, ("list", 2), None), ")": (0, None, None, None, None), "]": (0, None, None, None, None), "symbol": (0, "symbol", None, None, None), "string": (0, "string", None, None, None), "end": (0, None, None, None, None), } keywords = {'and', 'or', 'not'} symbols = {} _quoteletters = {'"', "'"} _simpleopletters = set(pycompat.iterbytestr("()[]#:=,-|&+!~^%")) # default set of valid characters for the initial letter of symbols _syminitletters = set(pycompat.iterbytestr( pycompat.sysbytes(string.ascii_letters) + pycompat.sysbytes(string.digits) + '._@')) | set(map(pycompat.bytechr, pycompat.xrange(128, 256))) # default set of valid characters for non-initial letters of symbols _symletters = _syminitletters | set(pycompat.iterbytestr('-/')) def tokenize(program, lookup=None, syminitletters=None, symletters=None): ''' Parse a revset statement into a stream of tokens ``syminitletters`` is the set of valid characters for the initial letter of symbols. By default, character ``c`` is recognized as valid for initial letter of symbols, if ``c.isalnum() or c in '._@' or ord(c) > 127``. ``symletters`` is the set of valid characters for non-initial letters of symbols. By default, character ``c`` is recognized as valid for non-initial letters of symbols, if ``c.isalnum() or c in '-._/@' or ord(c) > 127``. Check that @ is a valid unquoted token character (issue3686): >>> list(tokenize(b"@::")) [('symbol', '@', 0), ('::', None, 1), ('end', None, 3)] ''' if not isinstance(program, bytes): raise error.ProgrammingError('revset statement must be bytes, got %r' % program) program = pycompat.bytestr(program) if syminitletters is None: syminitletters = _syminitletters if symletters is None: symletters = _symletters if program and lookup: # attempt to parse old-style ranges first to deal with # things like old-tag which contain query metacharacters parts = program.split(':', 1) if all(lookup(sym) for sym in parts if sym): if parts[0]: yield ('symbol', parts[0], 0) if len(parts) > 1: s = len(parts[0]) yield (':', None, s) if parts[1]: yield ('symbol', parts[1], s + 1) yield ('end', None, len(program)) return pos, l = 0, len(program) while pos < l: c = program[pos] if c.isspace(): # skip inter-token whitespace pass elif c == ':' and program[pos:pos + 2] == '::': # look ahead carefully yield ('::', None, pos) pos += 1 # skip ahead elif c == '.' and program[pos:pos + 2] == '..': # look ahead carefully yield ('..', None, pos) pos += 1 # skip ahead elif c == '#' and program[pos:pos + 2] == '##': # look ahead carefully yield ('##', None, pos) pos += 1 # skip ahead elif c in _simpleopletters: # handle simple operators yield (c, None, pos) elif (c in _quoteletters or c == 'r' and program[pos:pos + 2] in ("r'", 'r"')): # handle quoted strings if c == 'r': pos += 1 c = program[pos] decode = lambda x: x else: decode = parser.unescapestr pos += 1 s = pos while pos < l: # find closing quote d = program[pos] if d == '\\': # skip over escaped characters pos += 2 continue if d == c: yield ('string', decode(program[s:pos]), s) break pos += 1 else: raise error.ParseError(_("unterminated string"), s) # gather up a symbol/keyword elif c in syminitletters: s = pos pos += 1 while pos < l: # find end of symbol d = program[pos] if d not in symletters: break if d == '.' and program[pos - 1] == '.': # special case for .. pos -= 1 break pos += 1 sym = program[s:pos] if sym in keywords: # operator keywords yield (sym, None, s) elif '-' in sym: # some jerk gave us foo-bar-baz, try to check if it's a symbol if lookup and lookup(sym): # looks like a real symbol yield ('symbol', sym, s) else: # looks like an expression parts = sym.split('-') for p in parts[:-1]: if p: # possible consecutive - yield ('symbol', p, s) s += len(p) yield ('-', None, s) s += 1 if parts[-1]: # possible trailing - yield ('symbol', parts[-1], s) else: yield ('symbol', sym, s) pos -= 1 else: raise error.ParseError(_("syntax error in revset '%s'") % program, pos) pos += 1 yield ('end', None, pos) # helpers _notset = object() def getsymbol(x): if x and x[0] == 'symbol': return x[1] raise error.ParseError(_('not a symbol')) def getstring(x, err): if x and (x[0] == 'string' or x[0] == 'symbol'): return x[1] raise error.ParseError(err) def getinteger(x, err, default=_notset): if not x and default is not _notset: return default try: return int(getstring(x, err)) except ValueError: raise error.ParseError(err) def getboolean(x, err): value = stringutil.parsebool(getsymbol(x)) if value is not None: return value raise error.ParseError(err) def getlist(x): if not x: return [] if x[0] == 'list': return list(x[1:]) return [x] def getrange(x, err): if not x: raise error.ParseError(err) op = x[0] if op == 'range': return x[1], x[2] elif op == 'rangepre': return None, x[1] elif op == 'rangepost': return x[1], None elif op == 'rangeall': return None, None raise error.ParseError(err) def getintrange(x, err1, err2, deffirst=_notset, deflast=_notset): """Get [first, last] integer range (both inclusive) from a parsed tree If any of the sides omitted, and if no default provided, ParseError will be raised. """ if x and (x[0] == 'string' or x[0] == 'symbol'): n = getinteger(x, err1) return n, n a, b = getrange(x, err1) return getinteger(a, err2, deffirst), getinteger(b, err2, deflast) def getargs(x, min, max, err): l = getlist(x) if len(l) < min or (max >= 0 and len(l) > max): raise error.ParseError(err) return l def getargsdict(x, funcname, keys): return parser.buildargsdict(getlist(x), funcname, parser.splitargspec(keys), keyvaluenode='keyvalue', keynode='symbol') # cache of {spec: raw parsed tree} built internally _treecache = {} def _cachedtree(spec): # thread safe because parse() is reentrant and dict.__setitem__() is atomic tree = _treecache.get(spec) if tree is None: _treecache[spec] = tree = parse(spec) return tree def _build(tmplspec, *repls): """Create raw parsed tree from a template revset statement >>> _build(b'f(_) and _', (b'string', b'1'), (b'symbol', b'2')) ('and', ('func', ('symbol', 'f'), ('string', '1')), ('symbol', '2')) """ template = _cachedtree(tmplspec) return parser.buildtree(template, ('symbol', '_'), *repls) def _match(patspec, tree): """Test if a tree matches the given pattern statement; return the matches >>> _match(b'f(_)', parse(b'f()')) >>> _match(b'f(_)', parse(b'f(1)')) [('func', ('symbol', 'f'), ('symbol', '1')), ('symbol', '1')] >>> _match(b'f(_)', parse(b'f(1, 2)')) """ pattern = _cachedtree(patspec) return parser.matchtree(pattern, tree, ('symbol', '_'), {'keyvalue', 'list'}) def _matchonly(revs, bases): return _match('ancestors(_) and not ancestors(_)', ('and', revs, bases)) def _fixops(x): """Rewrite raw parsed tree to resolve ambiguous syntax which cannot be handled well by our simple top-down parser""" if not isinstance(x, tuple): return x op = x[0] if op == 'parent': # x^:y means (x^) : y, not x ^ (:y) # x^: means (x^) :, not x ^ (:) post = ('parentpost', x[1]) if x[2][0] == 'dagrangepre': return _fixops(('dagrange', post, x[2][1])) elif x[2][0] == 'dagrangeall': return _fixops(('dagrangepost', post)) elif x[2][0] == 'rangepre': return _fixops(('range', post, x[2][1])) elif x[2][0] == 'rangeall': return _fixops(('rangepost', post)) elif op == 'or': # make number of arguments deterministic: # x + y + z -> (or x y z) -> (or (list x y z)) return (op, _fixops(('list',) + x[1:])) elif op == 'subscript' and x[1][0] == 'relation': # x#y[z] ternary return _fixops(('relsubscript', x[1][1], x[1][2], x[2])) return (op,) + tuple(_fixops(y) for y in x[1:]) def _analyze(x): if x is None: return x op = x[0] if op == 'minus': return _analyze(_build('_ and not _', *x[1:])) elif op == 'only': return _analyze(_build('only(_, _)', *x[1:])) elif op == 'onlypost': return _analyze(_build('only(_)', x[1])) elif op == 'dagrangeall': raise error.ParseError(_("can't use '::' in this context")) elif op == 'dagrangepre': return _analyze(_build('ancestors(_)', x[1])) elif op == 'dagrangepost': return _analyze(_build('descendants(_)', x[1])) elif op == 'negate': s = getstring(x[1], _("can't negate that")) return _analyze(('string', '-' + s)) elif op in ('string', 'symbol', 'smartset'): return x elif op == 'rangeall': return (op, None) elif op in {'or', 'not', 'rangepre', 'rangepost', 'parentpost'}: return (op, _analyze(x[1])) elif op == 'group': return _analyze(x[1]) elif op in {'and', 'dagrange', 'range', 'parent', 'ancestor', 'relation', 'subscript'}: ta = _analyze(x[1]) tb = _analyze(x[2]) return (op, ta, tb) elif op == 'relsubscript': ta = _analyze(x[1]) tb = _analyze(x[2]) tc = _analyze(x[3]) return (op, ta, tb, tc) elif op == 'list': return (op,) + tuple(_analyze(y) for y in x[1:]) elif op == 'keyvalue': return (op, x[1], _analyze(x[2])) elif op == 'func': return (op, x[1], _analyze(x[2])) raise ValueError('invalid operator %r' % op) def analyze(x): """Transform raw parsed tree to evaluatable tree which can be fed to optimize() or getset() All pseudo operations should be mapped to real operations or functions defined in methods or symbols table respectively. """ return _analyze(x) def _optimize(x): if x is None: return 0, x op = x[0] if op in ('string', 'symbol', 'smartset'): return 0.5, x # single revisions are small elif op == 'and': wa, ta = _optimize(x[1]) wb, tb = _optimize(x[2]) w = min(wa, wb) # (draft/secret/_notpublic() & ::x) have a fast path m = _match('_() & ancestors(_)', ('and', ta, tb)) if m and getsymbol(m[1]) in {'draft', 'secret', '_notpublic'}: return w, _build('_phaseandancestors(_, _)', m[1], m[2]) # (::x and not ::y)/(not ::y and ::x) have a fast path m = _matchonly(ta, tb) or _matchonly(tb, ta) if m: return w, _build('only(_, _)', *m[1:]) m = _match('not _', tb) if m: return wa, ('difference', ta, m[1]) if wa > wb: op = 'andsmally' return w, (op, ta, tb) elif op == 'or': # fast path for machine-generated expression, that is likely to have # lots of trivial revisions: 'a + b + c()' to '_list(a b) + c()' ws, ts, ss = [], [], [] def flushss(): if not ss: return if len(ss) == 1: w, t = ss[0] else: s = '\0'.join(t[1] for w, t in ss) y = _build('_list(_)', ('string', s)) w, t = _optimize(y) ws.append(w) ts.append(t) del ss[:] for y in getlist(x[1]): w, t = _optimize(y) if t is not None and (t[0] == 'string' or t[0] == 'symbol'): ss.append((w, t)) continue flushss() ws.append(w) ts.append(t) flushss() if len(ts) == 1: return ws[0], ts[0] # 'or' operation is fully optimized out return max(ws), (op, ('list',) + tuple(ts)) elif op == 'not': # Optimize not public() to _notpublic() because we have a fast version if _match('public()', x[1]): o = _optimize(_build('_notpublic()')) return o[0], o[1] else: o = _optimize(x[1]) return o[0], (op, o[1]) elif op == 'rangeall': return 1, x elif op in ('rangepre', 'rangepost', 'parentpost'): o = _optimize(x[1]) return o[0], (op, o[1]) elif op in ('dagrange', 'range'): wa, ta = _optimize(x[1]) wb, tb = _optimize(x[2]) return wa + wb, (op, ta, tb) elif op in ('parent', 'ancestor', 'relation', 'subscript'): w, t = _optimize(x[1]) return w, (op, t, x[2]) elif op == 'relsubscript': w, t = _optimize(x[1]) return w, (op, t, x[2], x[3]) elif op == 'list': ws, ts = zip(*(_optimize(y) for y in x[1:])) return sum(ws), (op,) + ts elif op == 'keyvalue': w, t = _optimize(x[2]) return w, (op, x[1], t) elif op == 'func': f = getsymbol(x[1]) wa, ta = _optimize(x[2]) w = getattr(symbols.get(f), '_weight', 1) m = _match('commonancestors(_)', ta) # Optimize heads(commonancestors(_)) because we have a fast version if f == 'heads' and m: return w + wa, _build('_commonancestorheads(_)', m[1]) return w + wa, (op, x[1], ta) raise ValueError('invalid operator %r' % op) def optimize(tree): """Optimize evaluatable tree All pseudo operations should be transformed beforehand. """ _weight, newtree = _optimize(tree) return newtree # the set of valid characters for the initial letter of symbols in # alias declarations and definitions _aliassyminitletters = _syminitletters | {'$'} def _parsewith(spec, lookup=None, syminitletters=None): """Generate a parse tree of given spec with given tokenizing options >>> _parsewith(b'foo($1)', syminitletters=_aliassyminitletters) ('func', ('symbol', 'foo'), ('symbol', '$1')) >>> _parsewith(b'$1') Traceback (most recent call last): ... ParseError: ("syntax error in revset '$1'", 0) >>> _parsewith(b'foo bar') Traceback (most recent call last): ... ParseError: ('invalid token', 4) """ if lookup and spec.startswith('revset(') and spec.endswith(')'): lookup = None p = parser.parser(elements) tree, pos = p.parse(tokenize(spec, lookup=lookup, syminitletters=syminitletters)) if pos != len(spec): raise error.ParseError(_('invalid token'), pos) return _fixops(parser.simplifyinfixops(tree, ('list', 'or'))) class _aliasrules(parser.basealiasrules): """Parsing and expansion rule set of revset aliases""" _section = _('revset alias') @staticmethod def _parse(spec): """Parse alias declaration/definition ``spec`` This allows symbol names to use also ``$`` as an initial letter (for backward compatibility), and callers of this function should examine whether ``$`` is used also for unexpected symbols or not. """ return _parsewith(spec, syminitletters=_aliassyminitletters) @staticmethod def _trygetfunc(tree): if tree[0] == 'func' and tree[1][0] == 'symbol': return tree[1][1], getlist(tree[2]) def expandaliases(tree, aliases, warn=None): """Expand aliases in a tree, aliases is a list of (name, value) tuples""" aliases = _aliasrules.buildmap(aliases) tree = _aliasrules.expand(aliases, tree) # warn about problematic (but not referred) aliases if warn is not None: for name, alias in sorted(aliases.iteritems()): if alias.error and not alias.warned: warn(_('warning: %s\n') % (alias.error)) alias.warned = True return tree def foldconcat(tree): """Fold elements to be concatenated by `##` """ if (not isinstance(tree, tuple) or tree[0] in ('string', 'symbol', 'smartset')): return tree if tree[0] == '_concat': pending = [tree] l = [] while pending: e = pending.pop() if e[0] == '_concat': pending.extend(reversed(e[1:])) elif e[0] in ('string', 'symbol'): l.append(e[1]) else: msg = _("\"##\" can't concatenate \"%s\" element") % (e[0]) raise error.ParseError(msg) return ('string', ''.join(l)) else: return tuple(foldconcat(t) for t in tree) def parse(spec, lookup=None): try: return _parsewith(spec, lookup=lookup) except error.ParseError as inst: if len(inst.args) > 1: # has location loc = inst.args[1] # Remove newlines -- spaces are equivalent whitespace. spec = spec.replace('\n', ' ') # We want the caret to point to the place in the template that # failed to parse, but in a hint we get a open paren at the # start. Therefore, we print "loc + 1" spaces (instead of "loc") # to line up the caret with the location of the error. inst.hint = spec + '\n' + ' ' * (loc + 1) + '^ ' + _('here') raise def _quote(s): r"""Quote a value in order to make it safe for the revset engine. >>> _quote(b'asdf') "'asdf'" >>> _quote(b"asdf'\"") '\'asdf\\\'"\'' >>> _quote(b'asdf\'') "'asdf\\''" >>> _quote(1) "'1'" """ return "'%s'" % stringutil.escapestr(pycompat.bytestr(s)) def _formatargtype(c, arg): if c == 'd': return '_rev(%d)' % int(arg) elif c == 's': return _quote(arg) elif c == 'r': if not isinstance(arg, bytes): raise TypeError parse(arg) # make sure syntax errors are confined return '(%s)' % arg elif c == 'n': return _quote(node.hex(arg)) elif c == 'b': try: return _quote(arg.branch()) except AttributeError: raise TypeError raise error.ParseError(_('unexpected revspec format character %s') % c) def _formatlistexp(s, t): l = len(s) if l == 0: return "_list('')" elif l == 1: return _formatargtype(t, s[0]) elif t == 'd': return _formatintlist(s) elif t == 's': return "_list(%s)" % _quote("\0".join(s)) elif t == 'n': return "_hexlist('%s')" % "\0".join(node.hex(a) for a in s) elif t == 'b': try: return "_list('%s')" % "\0".join(a.branch() for a in s) except AttributeError: raise TypeError m = l // 2 return '(%s or %s)' % (_formatlistexp(s[:m], t), _formatlistexp(s[m:], t)) def _formatintlist(data): try: l = len(data) if l == 0: return "_list('')" elif l == 1: return _formatargtype('d', data[0]) return "_intlist('%s')" % "\0".join('%d' % int(a) for a in data) except (TypeError, ValueError): raise error.ParseError(_('invalid argument for revspec')) def _formatparamexp(args, t): return ', '.join(_formatargtype(t, a) for a in args) _formatlistfuncs = { 'l': _formatlistexp, 'p': _formatparamexp, } def formatspec(expr, *args): ''' This is a convenience function for using revsets internally, and escapes arguments appropriately. Aliases are intentionally ignored so that intended expression behavior isn't accidentally subverted. Supported arguments: %r = revset expression, parenthesized %d = rev(int(arg)), no quoting %s = string(arg), escaped and single-quoted %b = arg.branch(), escaped and single-quoted %n = hex(arg), single-quoted %% = a literal '%' Prefixing the type with 'l' specifies a parenthesized list of that type, and 'p' specifies a list of function parameters of that type. >>> formatspec(b'%r:: and %lr', b'10 or 11', (b"this()", b"that()")) '(10 or 11):: and ((this()) or (that()))' >>> formatspec(b'%d:: and not %d::', 10, 20) '_rev(10):: and not _rev(20)::' >>> formatspec(b'%ld or %ld', [], [1]) "_list('') or _rev(1)" >>> formatspec(b'keyword(%s)', b'foo\\xe9') "keyword('foo\\\\xe9')" >>> b = lambda: b'default' >>> b.branch = b >>> formatspec(b'branch(%b)', b) "branch('default')" >>> formatspec(b'root(%ls)', [b'a', b'b', b'c', b'd']) "root(_list('a\\\\x00b\\\\x00c\\\\x00d'))" >>> formatspec(b'sort(%r, %ps)', b':', [b'desc', b'user']) "sort((:), 'desc', 'user')" >>> formatspec(b'%ls', [b'a', b"'"]) "_list('a\\\\x00\\\\'')" ''' parsed = _parseargs(expr, args) ret = [] for t, arg in parsed: if t is None: ret.append(arg) elif t == 'baseset': if isinstance(arg, set): arg = sorted(arg) ret.append(_formatintlist(list(arg))) else: raise error.ProgrammingError("unknown revspec item type: %r" % t) return b''.join(ret) def spectree(expr, *args): """similar to formatspec but return a parsed and optimized tree""" parsed = _parseargs(expr, args) ret = [] inputs = [] for t, arg in parsed: if t is None: ret.append(arg) elif t == 'baseset': newtree = ('smartset', smartset.baseset(arg)) inputs.append(newtree) ret.append("$") else: raise error.ProgrammingError("unknown revspec item type: %r" % t) expr = b''.join(ret) tree = _parsewith(expr, syminitletters=_aliassyminitletters) tree = parser.buildtree(tree, ('symbol', '$'), *inputs) tree = foldconcat(tree) tree = analyze(tree) tree = optimize(tree) return tree def _parseargs(expr, args): """parse the expression and replace all inexpensive args return a list of tuple [(arg-type, arg-value)] Arg-type can be: * None: a string ready to be concatenated into a final spec * 'baseset': an iterable of revisions """ expr = pycompat.bytestr(expr) argiter = iter(args) ret = [] pos = 0 while pos < len(expr): q = expr.find('%', pos) if q < 0: ret.append((None, expr[pos:])) break ret.append((None, expr[pos:q])) pos = q + 1 try: d = expr[pos] except IndexError: raise error.ParseError(_('incomplete revspec format character')) if d == '%': ret.append((None, d)) pos += 1 continue try: arg = next(argiter) except StopIteration: raise error.ParseError(_('missing argument for revspec')) f = _formatlistfuncs.get(d) if f: # a list of some type, might be expensive, do not replace pos += 1 islist = (d == 'l') try: d = expr[pos] except IndexError: raise error.ParseError(_('incomplete revspec format character')) if islist and d == 'd' and arg: # we don't create a baseset yet, because it come with an # extra cost. If we are going to serialize it we better # skip it. ret.append(('baseset', arg)) pos += 1 continue try: ret.append((None, f(list(arg), d))) except (TypeError, ValueError): raise error.ParseError(_('invalid argument for revspec')) else: # a single entry, not expensive, replace try: ret.append((None, _formatargtype(d, arg))) except (TypeError, ValueError): raise error.ParseError(_('invalid argument for revspec')) pos += 1 try: next(argiter) raise error.ParseError(_('too many revspec arguments specified')) except StopIteration: pass return ret def prettyformat(tree): return parser.prettyformat(tree, ('string', 'symbol')) def depth(tree): if isinstance(tree, tuple): return max(map(depth, tree)) + 1 else: return 0 def funcsused(tree): if not isinstance(tree, tuple) or tree[0] in ('string', 'symbol'): return set() else: funcs = set() for s in tree[1:]: funcs |= funcsused(s) if tree[0] == 'func': funcs.add(tree[1][1]) return funcs _hashre = util.re.compile('[0-9a-fA-F]{1,40}$') def _ishashlikesymbol(symbol): """returns true if the symbol looks like a hash""" return _hashre.match(symbol) def gethashlikesymbols(tree): """returns the list of symbols of the tree that look like hashes >>> gethashlikesymbols(parse(b'3::abe3ff')) ['3', 'abe3ff'] >>> gethashlikesymbols(parse(b'precursors(.)')) [] >>> gethashlikesymbols(parse(b'precursors(34)')) ['34'] >>> gethashlikesymbols(parse(b'abe3ffZ')) [] """ if not tree: return [] if tree[0] == "symbol": if _ishashlikesymbol(tree[1]): return [tree[1]] elif len(tree) >= 3: results = [] for subtree in tree[1:]: results += gethashlikesymbols(subtree) return results return []