comparison mercurial/match.py @ 14675:cfc89398f710

match: introduce basic fileset support

14674:1c151b963254

#
# This software may be used and distributed according to the terms of the
# GNU General Public License version 2 or any later version.

import re
import scmutil, util
from i18n import _

class match(object):
    def __init__(self, root, cwd, patterns, include=[], exclude=[],
                 default='glob', exact=False, auditor=None, ctx=None):
        """build an object to match a set of file patterns

        're:<regexp>' - a regular expression
        'path:<path>' - a path relative to canonroot
        'relglob:<glob>' - an unrooted glob (*.c matches C files in all dirs)
        'relpath:<path>' - a path relative to cwd
        'relre:<regexp>' - a regexp that needn't match the start of a name
        '<something>' - a pattern of the specified default type
        """

        self._root = root
        self._cwd = cwd
        self._files = []
        self._anypats = bool(include or exclude)
        self._ctx = ctx

        if include:
            pats = _normalize(include, 'glob', root, cwd, auditor)
            self.includepat, im = _buildmatch(pats, '(?:/|$)')
        if exclude:
            pats = _normalize(exclude, 'glob', root, cwd, auditor)
            self.excludepat, em = _buildmatch(pats, '(?:/|$)')
        if exact:
            self._files = patterns
            pm = self.exact
        elif patterns:
            pats = _normalize(patterns, default, root, cwd, auditor)
            self._files = _roots(pats)
            self._anypats = self._anypats or _anypats(pats)
            self.patternspat, pm = _buildmatch(pats, '$')

        if patterns or exact:
            if include:
                if exclude:
                    m = lambda f: im(f) and not em(f) and pm(f)

    """Split a string into an optional pattern kind prefix and the
    actual pattern."""
    if ':' in pat:
        kind, val = pat.split(':', 1)
        if kind in ('re', 'glob', 'path', 'relglob', 'relpath', 'relre',
                    'listfile', 'listfile0'):
            return kind, val
    return default, pat

def _globre(pat):
    "convert a glob pattern into a regexp"

        if name.startswith('^'):
            return name
        return '.*' + name
    return _globre(name) + tail

def _buildmatch(pats, tail):
    """build a matching function from a set of patterns"""
    try:
        pat = '(?:%s)' % '|'.join([_regex(k, p, tail) for (k, p) in pats])
        if len(pat) > 20000:
            raise OverflowError()

14675:cfc89398f710

#
# This software may be used and distributed according to the terms of the
# GNU General Public License version 2 or any later version.

import re
import scmutil, util, fileset
from i18n import _

def _expandsets(pats, ctx):
    '''convert set: patterns into a list of files in the given context'''
    fset = set()
    other = []

    for kind, expr in pats:
        if kind == 'set':
            if not ctx:
                raise util.Abort("fileset expression with no context")
            s = fileset.getfileset(ctx, expr)
            fset.update(s)
            continue
        other.append((kind, expr))
    return fset, other

class match(object):
    def __init__(self, root, cwd, patterns, include=[], exclude=[],
                 default='glob', exact=False, auditor=None, ctx=None):
        """build an object to match a set of file patterns

        're:<regexp>' - a regular expression
        'path:<path>' - a path relative to canonroot
        'relglob:<glob>' - an unrooted glob (*.c matches C files in all dirs)
        'relpath:<path>' - a path relative to cwd
        'relre:<regexp>' - a regexp that needn't match the start of a name
        'set:<fileset>' - a fileset expression
        '<something>' - a pattern of the specified default type
        """

        self._ctx = None
        self._root = root
        self._cwd = cwd
        self._files = []
        self._anypats = bool(include or exclude)
        self._ctx = ctx

        if include:
            pats = _normalize(include, 'glob', root, cwd, auditor)
            self.includepat, im = _buildmatch(ctx, pats, '(?:/|$)')
        if exclude:
            pats = _normalize(exclude, 'glob', root, cwd, auditor)
            self.excludepat, em = _buildmatch(ctx, pats, '(?:/|$)')
        if exact:
            self._files = patterns
            pm = self.exact
        elif patterns:
            pats = _normalize(patterns, default, root, cwd, auditor)
            self._files = _roots(pats)
            self._anypats = self._anypats or _anypats(pats)
            self.patternspat, pm = _buildmatch(ctx, pats, '$')

        if patterns or exact:
            if include:
                if exclude:
                    m = lambda f: im(f) and not em(f) and pm(f)

    """Split a string into an optional pattern kind prefix and the
    actual pattern."""
    if ':' in pat:
        kind, val = pat.split(':', 1)
        if kind in ('re', 'glob', 'path', 'relglob', 'relpath', 'relre',
                    'listfile', 'listfile0', 'set'):
            return kind, val
    return default, pat

def _globre(pat):
    "convert a glob pattern into a regexp"

        if name.startswith('^'):
            return name
        return '.*' + name
    return _globre(name) + tail

def _buildmatch(ctx, pats, tail):
    fset, pats = _expandsets(pats, ctx)
    if not pats:
        return "", fset.__contains__

    pat, mf = _buildregexmatch(pats, tail)
    if fset:
        return pat, lambda f: f in fset or mf(f)
    return pat, mf

def _buildregexmatch(pats, tail):
    """build a matching function from a set of patterns"""
    try:
        pat = '(?:%s)' % '|'.join([_regex(k, p, tail) for (k, p) in pats])
        if len(pat) > 20000:
            raise OverflowError()