13
13
# You should have received a copy of the GNU General Public License
14
14
# along with this program; if not, write to the Free Software
15
# Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
15
# Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
17
17
"""Tools for converting globs to regular expressions.
53
53
The pattern must not contain capturing groups.
54
54
The replacement might be either a string template in which \& will be
55
replaced with the match, or a function that will get the matching text
56
as argument. It does not get match object, because capturing is
55
replaced with the match, or a function that will get the matching text
56
as argument. It does not get match object, because capturing is
111
def _trailing_backslashes_regex(m):
112
"""Check trailing backslashes.
114
Does a head count on trailing backslashes to ensure there isn't an odd
115
one on the end that would escape the brackets we wrap the RE in.
117
if (len(m) % 2) != 0:
118
warning(u"Regular expressions cannot end with an odd number of '\\'. "
119
"Dropping the final '\\'.")
124
111
_sub_re = Replacer()
125
112
_sub_re.add(u'^RE:', u'')
126
113
_sub_re.add(u'\((?!\?)', u'(?:')
127
114
_sub_re.add(u'\(\?P<.*>', _invalid_regex(u'(?:'))
128
115
_sub_re.add(u'\(\?P=[^)]*\)', _invalid_regex(u''))
129
_sub_re.add(ur'\\+$', _trailing_backslashes_regex)
132
118
_sub_fullpath = Replacer()
161
147
Patterns are translated to regular expressions to expidite matching.
163
The regular expressions for multiple patterns are aggregated into
164
a super-regex containing groups of up to 99 patterns.
149
The regular expressions for multiple patterns are aggregated into
150
a super-regex containing groups of up to 99 patterns.
165
151
The 99 limitation is due to the grouping limit of the Python re module.
166
152
The resulting super-regex and associated patterns are stored as a list of
167
153
(regex,[patterns]) in _regex_patterns.
169
155
For performance reasons the patterns are categorised as extension patterns
170
156
(those that match against a file extension), basename patterns
171
157
(those that match against the basename of the filename),
172
158
and fullpath patterns (those that match against the full path).
173
The translations used for extensions and basenames are relatively simpler
159
The translations used for extensions and basenames are relatively simpler
174
160
and therefore faster to perform than the fullpath patterns.
176
Also, the extension patterns are more likely to find a match and
162
Also, the extension patterns are more likely to find a match and
177
163
so are matched first, then the basename patterns, then the fullpath
192
177
base_patterns.append(pat)
193
178
self._add_patterns(ext_patterns,_sub_extension,
194
179
prefix=r'(?:.*/)?(?!.*/)(?:.*\.)')
195
self._add_patterns(base_patterns,_sub_basename,
180
self._add_patterns(base_patterns,_sub_basename,
196
181
prefix=r'(?:.*/)?(?!.*/)')
197
self._add_patterns(path_patterns,_sub_fullpath)
182
self._add_patterns(path_patterns,_sub_fullpath)
199
184
def _add_patterns(self, patterns, translator, prefix=''):
201
186
grouped_rules = ['(%s)' % translator(pat) for pat in patterns[:99]]
202
187
joined_rule = '%s(?:%s)$' % (prefix, '|'.join(grouped_rules))
203
self._regex_patterns.append((re.compile(joined_rule, re.UNICODE),
188
self._regex_patterns.append((re.compile(joined_rule, re.UNICODE),
205
190
patterns = patterns[99:]
207
192
def match(self, filename):
208
193
"""Searches for a pattern that matches the given filename.
210
195
:return A matching pattern or None if there is no matching pattern.
212
197
for regex, patterns in self._regex_patterns:
215
200
return patterns[match.lastindex -1]
218
class ExceptionGlobster(object):
219
"""A Globster that supports exception patterns.
221
Exceptions are ignore patterns prefixed with '!'. Exception
222
patterns take precedence over regular patterns and cause a
223
matching filename to return None from the match() function.
224
Patterns using a '!!' prefix are highest precedence, and act
225
as regular ignores. '!!' patterns are useful to establish ignores
226
that apply under paths specified by '!' exception patterns.
229
def __init__(self,patterns):
230
ignores = [[], [], []]
232
if p.startswith(u'!!'):
233
ignores[2].append(p[2:])
234
elif p.startswith(u'!'):
235
ignores[1].append(p[1:])
238
self._ignores = [Globster(i) for i in ignores]
240
def match(self, filename):
241
"""Searches for a pattern that matches the given filename.
243
:return A matching pattern or None if there is no matching pattern.
245
double_neg = self._ignores[2].match(filename)
247
return "!!%s" % double_neg
248
elif self._ignores[1].match(filename):
251
return self._ignores[0].match(filename)
253
class _OrderedGlobster(Globster):
254
"""A Globster that keeps pattern order."""
256
def __init__(self, patterns):
259
:param patterns: sequence of glob patterns
261
# Note: This could be smarter by running like sequences together
262
self._regex_patterns = []
264
pat = normalize_pattern(pat)
265
if pat.startswith(u'RE:') or u'/' in pat:
266
self._add_patterns([pat], _sub_fullpath)
267
elif pat.startswith(u'*.'):
268
self._add_patterns([pat], _sub_extension,
269
prefix=r'(?:.*/)?(?!.*/)(?:.*\.)')
271
self._add_patterns([pat], _sub_basename,
272
prefix=r'(?:.*/)?(?!.*/)')
275
_slashes = re.compile(r'[\\/]+')
276
def normalize_pattern(pattern):
277
"""Converts backslashes in path patterns to forward slashes.
279
Doesn't normalize regular expressions - they may contain escapes.
281
if not (pattern.startswith('RE:') or pattern.startswith('!RE:')):
282
pattern = _slashes.sub('/', pattern)
284
pattern = pattern.rstrip('/')