[PATCH 6 of 6] fileset: extract language processing part to new module (API)

Yuya Nishihara yuya at tcha.org
Thu Aug 2 09:26:06 EDT 2018


# HG changeset patch
# User Yuya Nishihara <yuya at tcha.org>
# Date 1532242245 -32400
#      Sun Jul 22 15:50:45 2018 +0900
# Node ID e6e56b590c20c3a026a304f569e4e1edfd3b1b6a
# Parent  7b744a18768440cdca7662db7dc6179c1abd6613
fileset: extract language processing part to new module (API)

I'll add a couple more functions that work on parsed tree.

  % wc -l mercurial/fileset*.py
  559 mercurial/fileset.py
  135 mercurial/filesetlang.py
  694 total

diff --git a/hgext/lfs/__init__.py b/hgext/lfs/__init__.py
--- a/hgext/lfs/__init__.py
+++ b/hgext/lfs/__init__.py
@@ -136,7 +136,7 @@ from mercurial import (
     exchange,
     extensions,
     filelog,
-    fileset,
+    filesetlang,
     hg,
     localrepo,
     minifileset,
@@ -261,7 +261,7 @@ def _trackedmatcher(repo):
         # deprecated config: lfs.threshold
         threshold = repo.ui.configbytes('lfs', 'threshold')
         if threshold:
-            fileset.parse(trackspec)  # make sure syntax errors are confined
+            filesetlang.parse(trackspec)  # make sure syntax errors are confined
             trackspec = "(%s) | size('>%d')" % (trackspec, threshold)
 
         return minifileset.compile(trackspec)
@@ -361,7 +361,7 @@ def extsetup(ui):
 def lfsfileset(mctx, x):
     """File that uses LFS storage."""
     # i18n: "lfs" is a keyword
-    fileset.getargs(x, 0, 0, _("lfs takes no arguments"))
+    filesetlang.getargs(x, 0, 0, _("lfs takes no arguments"))
     ctx = mctx.ctx
     def lfsfilep(f):
         return wrapper.pointerfromctx(ctx, f, removed=True) is not None
diff --git a/mercurial/debugcommands.py b/mercurial/debugcommands.py
--- a/mercurial/debugcommands.py
+++ b/mercurial/debugcommands.py
@@ -48,7 +48,7 @@ from . import (
     exchange,
     extensions,
     filemerge,
-    fileset,
+    filesetlang,
     formatter,
     hg,
     httppeer,
@@ -916,13 +916,13 @@ def debugfileset(ui, repo, expr, **opts)
                 raise error.Abort(_('invalid stage name: %s') % n)
         showalways.update(opts['show_stage'])
 
-    tree = fileset.parse(expr)
+    tree = filesetlang.parse(expr)
     for n, f in stages:
         tree = f(tree)
         if n in showalways:
             if opts['show_stage'] or n != 'parsed':
                 ui.write(("* %s:\n") % n)
-            ui.write(fileset.prettyformat(tree), "\n")
+            ui.write(filesetlang.prettyformat(tree), "\n")
 
     files = set()
     if opts['all_files']:
diff --git a/mercurial/fileset.py b/mercurial/fileset.py
--- a/mercurial/fileset.py
+++ b/mercurial/fileset.py
@@ -13,9 +13,9 @@ import re
 from .i18n import _
 from . import (
     error,
+    filesetlang,
     match as matchmod,
     merge,
-    parser,
     pycompat,
     registrar,
     scmutil,
@@ -25,120 +25,12 @@ from .utils import (
     stringutil,
 )
 
-elements = {
-    # token-type: binding-strength, primary, prefix, infix, suffix
-    "(": (20, None, ("group", 1, ")"), ("func", 1, ")"), None),
-    ":": (15, None, None, ("kindpat", 15), None),
-    "-": (5, None, ("negate", 19), ("minus", 5), None),
-    "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),
-    "or": (4, None, None, ("or", 4), None),
-    "|": (4, None, None, ("or", 4), None),
-    "+": (4, None, None, ("or", 4), None),
-    ",": (2, None, None, ("list", 2), 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'}
-
-globchars = ".*{}[]?/\\_"
-
-def tokenize(program):
-    pos, l = 0, len(program)
-    program = pycompat.bytestr(program)
-    while pos < l:
-        c = program[pos]
-        if c.isspace(): # skip inter-token whitespace
-            pass
-        elif c in "(),-:|&+!": # handle simple operators
-            yield (c, None, pos)
-        elif (c in '"\'' 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)
-        elif c.isalnum() or c in globchars or ord(c) > 127:
-            # gather up a symbol/keyword
-            s = pos
-            pos += 1
-            while pos < l: # find end of symbol
-                d = program[pos]
-                if not (d.isalnum() or d in globchars or ord(d) > 127):
-                    break
-                pos += 1
-            sym = program[s:pos]
-            if sym in keywords: # operator keywords
-                yield (sym, None, s)
-            else:
-                yield ('symbol', sym, s)
-            pos -= 1
-        else:
-            raise error.ParseError(_("syntax error"), pos)
-        pos += 1
-    yield ('end', None, pos)
-
-def parse(expr):
-    p = parser.parser(elements)
-    tree, pos = p.parse(tokenize(expr))
-    if pos != len(expr):
-        raise error.ParseError(_("invalid token"), pos)
-    return parser.simplifyinfixops(tree, {'list', 'or'})
-
-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 _getkindpat(x, y, allkinds, err):
-    kind = getsymbol(x)
-    pat = getstring(y, err)
-    if kind not in allkinds:
-        raise error.ParseError(_("invalid pattern kind: %s") % kind)
-    return '%s:%s' % (kind, pat)
-
-def getpattern(x, allkinds, err):
-    if x and x[0] == 'kindpat':
-        return _getkindpat(x[1], x[2], allkinds, err)
-    return getstring(x, err)
-
-def getlist(x):
-    if not x:
-        return []
-    if x[0] == 'list':
-        return list(x[1:])
-    return [x]
-
-def getargs(x, min, max, err):
-    l = getlist(x)
-    if len(l) < min or len(l) > max:
-        raise error.ParseError(err)
-    return l
+# helpers for processing parsed tree
+getsymbol = filesetlang.getsymbol
+getstring = filesetlang.getstring
+_getkindpat = filesetlang.getkindpat
+getpattern = filesetlang.getpattern
+getargs = filesetlang.getargs
 
 def getmatch(mctx, x):
     if not x:
@@ -192,7 +84,7 @@ def func(mctx, a, b):
 # with:
 #  mctx - current matchctx instance
 #  x - argument in tree form
-symbols = {}
+symbols = filesetlang.symbols
 
 # filesets using matchctx.status()
 _statuscallers = set()
@@ -635,7 +527,7 @@ def _intree(funcs, tree):
 
 def match(ctx, expr, badfn=None):
     """Create a matcher for a single fileset expression"""
-    tree = parse(expr)
+    tree = filesetlang.parse(expr)
     mctx = matchctx(ctx, _buildstatus(ctx, tree), badfn=badfn)
     return getmatch(mctx, tree)
 
@@ -653,9 +545,6 @@ def _buildstatus(ctx, tree, basectx=None
     else:
         return None
 
-def prettyformat(tree):
-    return parser.prettyformat(tree, ('string', 'symbol'))
-
 def loadpredicate(ui, extname, registrarobj):
     """Load fileset predicates from specified registrarobj
     """
diff --git a/mercurial/fileset.py b/mercurial/filesetlang.py
copy from mercurial/fileset.py
copy to mercurial/filesetlang.py
--- a/mercurial/fileset.py
+++ b/mercurial/filesetlang.py
@@ -1,4 +1,4 @@
-# fileset.py - file set queries for mercurial
+# filesetlang.py - parser, tokenizer and utility for file set language
 #
 # Copyright 2010 Matt Mackall <mpm at selenic.com>
 #
@@ -7,22 +7,11 @@
 
 from __future__ import absolute_import
 
-import errno
-import re
-
 from .i18n import _
 from . import (
     error,
-    match as matchmod,
-    merge,
     parser,
     pycompat,
-    registrar,
-    scmutil,
-    util,
-)
-from .utils import (
-    stringutil,
 )
 
 elements = {
@@ -46,6 +35,8 @@ elements = {
 
 keywords = {'and', 'or', 'not'}
 
+symbols = {}
+
 globchars = ".*{}[]?/\\_"
 
 def tokenize(program):
@@ -115,7 +106,7 @@ def getstring(x, err):
         return x[1]
     raise error.ParseError(err)
 
-def _getkindpat(x, y, allkinds, err):
+def getkindpat(x, y, allkinds, err):
     kind = getsymbol(x)
     pat = getstring(y, err)
     if kind not in allkinds:
@@ -124,7 +115,7 @@ def _getkindpat(x, y, allkinds, err):
 
 def getpattern(x, allkinds, err):
     if x and x[0] == 'kindpat':
-        return _getkindpat(x[1], x[2], allkinds, err)
+        return getkindpat(x[1], x[2], allkinds, err)
     return getstring(x, err)
 
 def getlist(x):
@@ -140,532 +131,5 @@ def getargs(x, min, max, err):
         raise error.ParseError(err)
     return l
 
-def getmatch(mctx, x):
-    if not x:
-        raise error.ParseError(_("missing argument"))
-    return methods[x[0]](mctx, *x[1:])
-
-def stringmatch(mctx, x):
-    return mctx.matcher([x])
-
-def kindpatmatch(mctx, x, y):
-    return stringmatch(mctx, _getkindpat(x, y, matchmod.allpatternkinds,
-                                         _("pattern must be a string")))
-
-def andmatch(mctx, x, y):
-    xm = getmatch(mctx, x)
-    ym = getmatch(mctx, y)
-    return matchmod.intersectmatchers(xm, ym)
-
-def ormatch(mctx, *xs):
-    ms = [getmatch(mctx, x) for x in xs]
-    return matchmod.unionmatcher(ms)
-
-def notmatch(mctx, x):
-    m = getmatch(mctx, x)
-    return mctx.predicate(lambda f: not m(f), predrepr=('<not %r>', m))
-
-def minusmatch(mctx, x, y):
-    xm = getmatch(mctx, x)
-    ym = getmatch(mctx, y)
-    return matchmod.differencematcher(xm, ym)
-
-def negatematch(mctx, x):
-    raise error.ParseError(_("can't use negate operator in this context"))
-
-def listmatch(mctx, *xs):
-    raise error.ParseError(_("can't use a list in this context"),
-                           hint=_('see hg help "filesets.x or y"'))
-
-def func(mctx, a, b):
-    funcname = getsymbol(a)
-    if funcname in symbols:
-        return symbols[funcname](mctx, b)
-
-    keep = lambda fn: getattr(fn, '__doc__', None) is not None
-
-    syms = [s for (s, fn) in symbols.items() if keep(fn)]
-    raise error.UnknownIdentifier(funcname, syms)
-
-# symbols are callable like:
-#  fun(mctx, x)
-# with:
-#  mctx - current matchctx instance
-#  x - argument in tree form
-symbols = {}
-
-# filesets using matchctx.status()
-_statuscallers = set()
-
-predicate = registrar.filesetpredicate()
-
- at predicate('modified()', callstatus=True)
-def modified(mctx, x):
-    """File that is modified according to :hg:`status`.
-    """
-    # i18n: "modified" is a keyword
-    getargs(x, 0, 0, _("modified takes no arguments"))
-    s = set(mctx.status().modified)
-    return mctx.predicate(s.__contains__, predrepr='modified')
-
- at predicate('added()', callstatus=True)
-def added(mctx, x):
-    """File that is added according to :hg:`status`.
-    """
-    # i18n: "added" is a keyword
-    getargs(x, 0, 0, _("added takes no arguments"))
-    s = set(mctx.status().added)
-    return mctx.predicate(s.__contains__, predrepr='added')
-
- at predicate('removed()', callstatus=True)
-def removed(mctx, x):
-    """File that is removed according to :hg:`status`.
-    """
-    # i18n: "removed" is a keyword
-    getargs(x, 0, 0, _("removed takes no arguments"))
-    s = set(mctx.status().removed)
-    return mctx.predicate(s.__contains__, predrepr='removed')
-
- at predicate('deleted()', callstatus=True)
-def deleted(mctx, x):
-    """Alias for ``missing()``.
-    """
-    # i18n: "deleted" is a keyword
-    getargs(x, 0, 0, _("deleted takes no arguments"))
-    s = set(mctx.status().deleted)
-    return mctx.predicate(s.__contains__, predrepr='deleted')
-
- at predicate('missing()', callstatus=True)
-def missing(mctx, x):
-    """File that is missing according to :hg:`status`.
-    """
-    # i18n: "missing" is a keyword
-    getargs(x, 0, 0, _("missing takes no arguments"))
-    s = set(mctx.status().deleted)
-    return mctx.predicate(s.__contains__, predrepr='deleted')
-
- at predicate('unknown()', callstatus=True)
-def unknown(mctx, x):
-    """File that is unknown according to :hg:`status`."""
-    # i18n: "unknown" is a keyword
-    getargs(x, 0, 0, _("unknown takes no arguments"))
-    s = set(mctx.status().unknown)
-    return mctx.predicate(s.__contains__, predrepr='unknown')
-
- at predicate('ignored()', callstatus=True)
-def ignored(mctx, x):
-    """File that is ignored according to :hg:`status`."""
-    # i18n: "ignored" is a keyword
-    getargs(x, 0, 0, _("ignored takes no arguments"))
-    s = set(mctx.status().ignored)
-    return mctx.predicate(s.__contains__, predrepr='ignored')
-
- at predicate('clean()', callstatus=True)
-def clean(mctx, x):
-    """File that is clean according to :hg:`status`.
-    """
-    # i18n: "clean" is a keyword
-    getargs(x, 0, 0, _("clean takes no arguments"))
-    s = set(mctx.status().clean)
-    return mctx.predicate(s.__contains__, predrepr='clean')
-
- at predicate('tracked()')
-def tracked(mctx, x):
-    """File that is under Mercurial control."""
-    # i18n: "tracked" is a keyword
-    getargs(x, 0, 0, _("tracked takes no arguments"))
-    return mctx.predicate(mctx.ctx.__contains__, predrepr='tracked')
-
- at predicate('binary()')
-def binary(mctx, x):
-    """File that appears to be binary (contains NUL bytes).
-    """
-    # i18n: "binary" is a keyword
-    getargs(x, 0, 0, _("binary takes no arguments"))
-    return mctx.fpredicate(lambda fctx: fctx.isbinary(),
-                           predrepr='binary', cache=True)
-
- at predicate('exec()')
-def exec_(mctx, x):
-    """File that is marked as executable.
-    """
-    # i18n: "exec" is a keyword
-    getargs(x, 0, 0, _("exec takes no arguments"))
-    ctx = mctx.ctx
-    return mctx.predicate(lambda f: ctx.flags(f) == 'x', predrepr='exec')
-
- at predicate('symlink()')
-def symlink(mctx, x):
-    """File that is marked as a symlink.
-    """
-    # i18n: "symlink" is a keyword
-    getargs(x, 0, 0, _("symlink takes no arguments"))
-    ctx = mctx.ctx
-    return mctx.predicate(lambda f: ctx.flags(f) == 'l', predrepr='symlink')
-
- at predicate('resolved()')
-def resolved(mctx, x):
-    """File that is marked resolved according to :hg:`resolve -l`.
-    """
-    # i18n: "resolved" is a keyword
-    getargs(x, 0, 0, _("resolved takes no arguments"))
-    if mctx.ctx.rev() is not None:
-        return mctx.never()
-    ms = merge.mergestate.read(mctx.ctx.repo())
-    return mctx.predicate(lambda f: f in ms and ms[f] == 'r',
-                          predrepr='resolved')
-
- at predicate('unresolved()')
-def unresolved(mctx, x):
-    """File that is marked unresolved according to :hg:`resolve -l`.
-    """
-    # i18n: "unresolved" is a keyword
-    getargs(x, 0, 0, _("unresolved takes no arguments"))
-    if mctx.ctx.rev() is not None:
-        return mctx.never()
-    ms = merge.mergestate.read(mctx.ctx.repo())
-    return mctx.predicate(lambda f: f in ms and ms[f] == 'u',
-                          predrepr='unresolved')
-
- at predicate('hgignore()')
-def hgignore(mctx, x):
-    """File that matches the active .hgignore pattern.
-    """
-    # i18n: "hgignore" is a keyword
-    getargs(x, 0, 0, _("hgignore takes no arguments"))
-    return mctx.ctx.repo().dirstate._ignore
-
- at predicate('portable()')
-def portable(mctx, x):
-    """File that has a portable name. (This doesn't include filenames with case
-    collisions.)
-    """
-    # i18n: "portable" is a keyword
-    getargs(x, 0, 0, _("portable takes no arguments"))
-    return mctx.predicate(lambda f: util.checkwinfilename(f) is None,
-                          predrepr='portable')
-
- at predicate('grep(regex)')
-def grep(mctx, x):
-    """File contains the given regular expression.
-    """
-    try:
-        # i18n: "grep" is a keyword
-        r = re.compile(getstring(x, _("grep requires a pattern")))
-    except re.error as e:
-        raise error.ParseError(_('invalid match pattern: %s') %
-                               stringutil.forcebytestr(e))
-    return mctx.fpredicate(lambda fctx: r.search(fctx.data()),
-                           predrepr=('grep(%r)', r.pattern), cache=True)
-
-def _sizetomax(s):
-    try:
-        s = s.strip().lower()
-        for k, v in util._sizeunits:
-            if s.endswith(k):
-                # max(4k) = 5k - 1, max(4.5k) = 4.6k - 1
-                n = s[:-len(k)]
-                inc = 1.0
-                if "." in n:
-                    inc /= 10 ** len(n.split(".")[1])
-                return int((float(n) + inc) * v) - 1
-        # no extension, this is a precise value
-        return int(s)
-    except ValueError:
-        raise error.ParseError(_("couldn't parse size: %s") % s)
-
-def sizematcher(expr):
-    """Return a function(size) -> bool from the ``size()`` expression"""
-    expr = expr.strip()
-    if '-' in expr: # do we have a range?
-        a, b = expr.split('-', 1)
-        a = util.sizetoint(a)
-        b = util.sizetoint(b)
-        return lambda x: x >= a and x <= b
-    elif expr.startswith("<="):
-        a = util.sizetoint(expr[2:])
-        return lambda x: x <= a
-    elif expr.startswith("<"):
-        a = util.sizetoint(expr[1:])
-        return lambda x: x < a
-    elif expr.startswith(">="):
-        a = util.sizetoint(expr[2:])
-        return lambda x: x >= a
-    elif expr.startswith(">"):
-        a = util.sizetoint(expr[1:])
-        return lambda x: x > a
-    else:
-        a = util.sizetoint(expr)
-        b = _sizetomax(expr)
-        return lambda x: x >= a and x <= b
-
- at predicate('size(expression)')
-def size(mctx, x):
-    """File size matches the given expression. Examples:
-
-    - size('1k') - files from 1024 to 2047 bytes
-    - size('< 20k') - files less than 20480 bytes
-    - size('>= .5MB') - files at least 524288 bytes
-    - size('4k - 1MB') - files from 4096 bytes to 1048576 bytes
-    """
-    # i18n: "size" is a keyword
-    expr = getstring(x, _("size requires an expression"))
-    m = sizematcher(expr)
-    return mctx.fpredicate(lambda fctx: m(fctx.size()),
-                           predrepr=('size(%r)', expr), cache=True)
-
- at predicate('encoding(name)')
-def encoding(mctx, x):
-    """File can be successfully decoded with the given character
-    encoding. May not be useful for encodings other than ASCII and
-    UTF-8.
-    """
-
-    # i18n: "encoding" is a keyword
-    enc = getstring(x, _("encoding requires an encoding name"))
-
-    def encp(fctx):
-        d = fctx.data()
-        try:
-            d.decode(pycompat.sysstr(enc))
-            return True
-        except LookupError:
-            raise error.Abort(_("unknown encoding '%s'") % enc)
-        except UnicodeDecodeError:
-            return False
-
-    return mctx.fpredicate(encp, predrepr=('encoding(%r)', enc), cache=True)
-
- at predicate('eol(style)')
-def eol(mctx, x):
-    """File contains newlines of the given style (dos, unix, mac). Binary
-    files are excluded, files with mixed line endings match multiple
-    styles.
-    """
-
-    # i18n: "eol" is a keyword
-    enc = getstring(x, _("eol requires a style name"))
-
-    def eolp(fctx):
-        if fctx.isbinary():
-            return False
-        d = fctx.data()
-        if (enc == 'dos' or enc == 'win') and '\r\n' in d:
-            return True
-        elif enc == 'unix' and re.search('(?<!\r)\n', d):
-            return True
-        elif enc == 'mac' and re.search('\r(?!\n)', d):
-            return True
-        return False
-    return mctx.fpredicate(eolp, predrepr=('eol(%r)', enc), cache=True)
-
- at predicate('copied()')
-def copied(mctx, x):
-    """File that is recorded as being copied.
-    """
-    # i18n: "copied" is a keyword
-    getargs(x, 0, 0, _("copied takes no arguments"))
-    def copiedp(fctx):
-        p = fctx.parents()
-        return p and p[0].path() != fctx.path()
-    return mctx.fpredicate(copiedp, predrepr='copied', cache=True)
-
- at predicate('revs(revs, pattern)')
-def revs(mctx, x):
-    """Evaluate set in the specified revisions. If the revset match multiple
-    revs, this will return file matching pattern in any of the revision.
-    """
-    # i18n: "revs" is a keyword
-    r, x = getargs(x, 2, 2, _("revs takes two arguments"))
-    # i18n: "revs" is a keyword
-    revspec = getstring(r, _("first argument to revs must be a revision"))
-    repo = mctx.ctx.repo()
-    revs = scmutil.revrange(repo, [revspec])
-
-    matchers = []
-    for r in revs:
-        ctx = repo[r]
-        matchers.append(getmatch(mctx.switch(ctx, _buildstatus(ctx, x)), x))
-    if not matchers:
-        return mctx.never()
-    if len(matchers) == 1:
-        return matchers[0]
-    return matchmod.unionmatcher(matchers)
-
- at predicate('status(base, rev, pattern)')
-def status(mctx, x):
-    """Evaluate predicate using status change between ``base`` and
-    ``rev``. Examples:
-
-    - ``status(3, 7, added())`` - matches files added from "3" to "7"
-    """
-    repo = mctx.ctx.repo()
-    # i18n: "status" is a keyword
-    b, r, x = getargs(x, 3, 3, _("status takes three arguments"))
-    # i18n: "status" is a keyword
-    baseerr = _("first argument to status must be a revision")
-    baserevspec = getstring(b, baseerr)
-    if not baserevspec:
-        raise error.ParseError(baseerr)
-    reverr = _("second argument to status must be a revision")
-    revspec = getstring(r, reverr)
-    if not revspec:
-        raise error.ParseError(reverr)
-    basectx, ctx = scmutil.revpair(repo, [baserevspec, revspec])
-    return getmatch(mctx.switch(ctx, _buildstatus(ctx, x, basectx=basectx)), x)
-
- at predicate('subrepo([pattern])')
-def subrepo(mctx, x):
-    """Subrepositories whose paths match the given pattern.
-    """
-    # i18n: "subrepo" is a keyword
-    getargs(x, 0, 1, _("subrepo takes at most one argument"))
-    ctx = mctx.ctx
-    sstate = ctx.substate
-    if x:
-        pat = getpattern(x, matchmod.allpatternkinds,
-                         # i18n: "subrepo" is a keyword
-                         _("subrepo requires a pattern or no arguments"))
-        fast = not matchmod.patkind(pat)
-        if fast:
-            def m(s):
-                return (s == pat)
-        else:
-            m = matchmod.match(ctx.repo().root, '', [pat], ctx=ctx)
-        return mctx.predicate(lambda f: f in sstate and m(f),
-                              predrepr=('subrepo(%r)', pat))
-    else:
-        return mctx.predicate(sstate.__contains__, predrepr='subrepo')
-
-methods = {
-    'string': stringmatch,
-    'symbol': stringmatch,
-    'kindpat': kindpatmatch,
-    'and': andmatch,
-    'or': ormatch,
-    'minus': minusmatch,
-    'negate': negatematch,
-    'list': listmatch,
-    'group': getmatch,
-    'not': notmatch,
-    'func': func,
-}
-
-class matchctx(object):
-    def __init__(self, ctx, status=None, badfn=None):
-        self.ctx = ctx
-        self._status = status
-        self._badfn = badfn
-
-    def status(self):
-        return self._status
-
-    def matcher(self, patterns):
-        return self.ctx.match(patterns, badfn=self._badfn)
-
-    def predicate(self, predfn, predrepr=None, cache=False):
-        """Create a matcher to select files by predfn(filename)"""
-        if cache:
-            predfn = util.cachefunc(predfn)
-        repo = self.ctx.repo()
-        return matchmod.predicatematcher(repo.root, repo.getcwd(), predfn,
-                                         predrepr=predrepr, badfn=self._badfn)
-
-    def fpredicate(self, predfn, predrepr=None, cache=False):
-        """Create a matcher to select files by predfn(fctx) at the current
-        revision
-
-        Missing files are ignored.
-        """
-        ctx = self.ctx
-        if ctx.rev() is None:
-            def fctxpredfn(f):
-                try:
-                    fctx = ctx[f]
-                except error.LookupError:
-                    return False
-                try:
-                    fctx.audit()
-                except error.Abort:
-                    return False
-                try:
-                    return predfn(fctx)
-                except (IOError, OSError) as e:
-                    # open()-ing a directory fails with EACCES on Windows
-                    if e.errno in (errno.ENOENT, errno.EACCES, errno.ENOTDIR,
-                                   errno.EISDIR):
-                        return False
-                    raise
-        else:
-            def fctxpredfn(f):
-                try:
-                    fctx = ctx[f]
-                except error.LookupError:
-                    return False
-                return predfn(fctx)
-        return self.predicate(fctxpredfn, predrepr=predrepr, cache=cache)
-
-    def never(self):
-        """Create a matcher to select nothing"""
-        repo = self.ctx.repo()
-        return matchmod.nevermatcher(repo.root, repo.getcwd(),
-                                     badfn=self._badfn)
-
-    def switch(self, ctx, status=None):
-        return matchctx(ctx, status, self._badfn)
-
-# filesets using matchctx.switch()
-_switchcallers = [
-    'revs',
-    'status',
-]
-
-def _intree(funcs, tree):
-    if isinstance(tree, tuple):
-        if tree[0] == 'func' and tree[1][0] == 'symbol':
-            if tree[1][1] in funcs:
-                return True
-            if tree[1][1] in _switchcallers:
-                # arguments won't be evaluated in the current context
-                return False
-        for s in tree[1:]:
-            if _intree(funcs, s):
-                return True
-    return False
-
-def match(ctx, expr, badfn=None):
-    """Create a matcher for a single fileset expression"""
-    tree = parse(expr)
-    mctx = matchctx(ctx, _buildstatus(ctx, tree), badfn=badfn)
-    return getmatch(mctx, tree)
-
-def _buildstatus(ctx, tree, basectx=None):
-    # do we need status info?
-
-    if _intree(_statuscallers, tree):
-        unknown = _intree(['unknown'], tree)
-        ignored = _intree(['ignored'], tree)
-
-        if basectx is None:
-            basectx = ctx.p1()
-        return basectx.status(ctx, listunknown=unknown, listignored=ignored,
-                              listclean=True)
-    else:
-        return None
-
 def prettyformat(tree):
     return parser.prettyformat(tree, ('string', 'symbol'))
-
-def loadpredicate(ui, extname, registrarobj):
-    """Load fileset predicates from specified registrarobj
-    """
-    for name, func in registrarobj._table.iteritems():
-        symbols[name] = func
-        if func._callstatus:
-            _statuscallers.add(name)
-
-# load built-in predicates explicitly to setup _statuscallers
-loadpredicate(None, None, predicate)
-
-# tell hggettext to extract docstrings from these functions:
-i18nfunctions = symbols.values()
diff --git a/mercurial/minifileset.py b/mercurial/minifileset.py
--- a/mercurial/minifileset.py
+++ b/mercurial/minifileset.py
@@ -11,12 +11,13 @@ from .i18n import _
 from . import (
     error,
     fileset,
+    filesetlang,
     pycompat,
 )
 
 def _sizep(x):
     # i18n: "size" is a keyword
-    expr = fileset.getstring(x, _("size requires an expression"))
+    expr = filesetlang.getstring(x, _("size requires an expression"))
     return fileset.sizematcher(expr)
 
 def _compile(tree):
@@ -24,7 +25,7 @@ def _compile(tree):
         raise error.ParseError(_("missing argument"))
     op = tree[0]
     if op in {'symbol', 'string', 'kindpat'}:
-        name = fileset.getpattern(tree, {'path'}, _('invalid file pattern'))
+        name = filesetlang.getpattern(tree, {'path'}, _('invalid file pattern'))
         if name.startswith('**'): # file extension test, ex. "**.tar.gz"
             ext = name[2:]
             for c in pycompat.bytestr(ext):
@@ -57,7 +58,7 @@ def _compile(tree):
             'size': lambda n, s: _sizep(tree[2])(s),
         }
 
-        name = fileset.getsymbol(tree[1])
+        name = filesetlang.getsymbol(tree[1])
         if name in symbols:
             return symbols[name]
 
@@ -87,5 +88,5 @@ def compile(text):
     files whose name ends with ".zip", and all files under "bin" in the repo
     root except for "bin/README".
     """
-    tree = fileset.parse(text)
+    tree = filesetlang.parse(text)
     return _compile(tree)


More information about the Mercurial-devel mailing list