Show More
@@ -1,613 +1,613 b'' | |||||
1 | # match.py - filename matching |
|
1 | # match.py - filename matching | |
2 | # |
|
2 | # | |
3 | # Copyright 2008, 2009 Matt Mackall <mpm@selenic.com> and others |
|
3 | # Copyright 2008, 2009 Matt Mackall <mpm@selenic.com> and others | |
4 | # |
|
4 | # | |
5 | # This software may be used and distributed according to the terms of the |
|
5 | # This software may be used and distributed according to the terms of the | |
6 | # GNU General Public License version 2 or any later version. |
|
6 | # GNU General Public License version 2 or any later version. | |
7 |
|
7 | |||
8 | import re |
|
8 | import re | |
9 | import util, pathutil |
|
9 | import util, pathutil | |
10 | from i18n import _ |
|
10 | from i18n import _ | |
11 |
|
11 | |||
12 | propertycache = util.propertycache |
|
12 | propertycache = util.propertycache | |
13 |
|
13 | |||
14 | def _rematcher(regex): |
|
14 | def _rematcher(regex): | |
15 | '''compile the regexp with the best available regexp engine and return a |
|
15 | '''compile the regexp with the best available regexp engine and return a | |
16 | matcher function''' |
|
16 | matcher function''' | |
17 | m = util.re.compile(regex) |
|
17 | m = util.re.compile(regex) | |
18 | try: |
|
18 | try: | |
19 | # slightly faster, provided by facebook's re2 bindings |
|
19 | # slightly faster, provided by facebook's re2 bindings | |
20 | return m.test_match |
|
20 | return m.test_match | |
21 | except AttributeError: |
|
21 | except AttributeError: | |
22 | return m.match |
|
22 | return m.match | |
23 |
|
23 | |||
24 | def _expandsets(kindpats, ctx, listsubrepos): |
|
24 | def _expandsets(kindpats, ctx, listsubrepos): | |
25 | '''Returns the kindpats list with the 'set' patterns expanded.''' |
|
25 | '''Returns the kindpats list with the 'set' patterns expanded.''' | |
26 | fset = set() |
|
26 | fset = set() | |
27 | other = [] |
|
27 | other = [] | |
28 |
|
28 | |||
29 | for kind, pat, source in kindpats: |
|
29 | for kind, pat, source in kindpats: | |
30 | if kind == 'set': |
|
30 | if kind == 'set': | |
31 | if not ctx: |
|
31 | if not ctx: | |
32 | raise util.Abort("fileset expression with no context") |
|
32 | raise util.Abort("fileset expression with no context") | |
33 | s = ctx.getfileset(pat) |
|
33 | s = ctx.getfileset(pat) | |
34 | fset.update(s) |
|
34 | fset.update(s) | |
35 |
|
35 | |||
36 | if listsubrepos: |
|
36 | if listsubrepos: | |
37 | for subpath in ctx.substate: |
|
37 | for subpath in ctx.substate: | |
38 | s = ctx.sub(subpath).getfileset(pat) |
|
38 | s = ctx.sub(subpath).getfileset(pat) | |
39 | fset.update(subpath + '/' + f for f in s) |
|
39 | fset.update(subpath + '/' + f for f in s) | |
40 |
|
40 | |||
41 | continue |
|
41 | continue | |
42 | other.append((kind, pat, source)) |
|
42 | other.append((kind, pat, source)) | |
43 | return fset, other |
|
43 | return fset, other | |
44 |
|
44 | |||
45 | def _kindpatsalwaysmatch(kindpats): |
|
45 | def _kindpatsalwaysmatch(kindpats): | |
46 | """"Checks whether the kindspats match everything, as e.g. |
|
46 | """"Checks whether the kindspats match everything, as e.g. | |
47 | 'relpath:.' does. |
|
47 | 'relpath:.' does. | |
48 | """ |
|
48 | """ | |
49 | for kind, pat, source in kindpats: |
|
49 | for kind, pat, source in kindpats: | |
50 | if pat != '' or kind not in ['relpath', 'glob']: |
|
50 | if pat != '' or kind not in ['relpath', 'glob']: | |
51 | return False |
|
51 | return False | |
52 | return True |
|
52 | return True | |
53 |
|
53 | |||
54 | class match(object): |
|
54 | class match(object): | |
55 | def __init__(self, root, cwd, patterns, include=[], exclude=[], |
|
55 | def __init__(self, root, cwd, patterns, include=[], exclude=[], | |
56 | default='glob', exact=False, auditor=None, ctx=None, |
|
56 | default='glob', exact=False, auditor=None, ctx=None, | |
57 | listsubrepos=False, warn=None): |
|
57 | listsubrepos=False, warn=None): | |
58 | """build an object to match a set of file patterns |
|
58 | """build an object to match a set of file patterns | |
59 |
|
59 | |||
60 | arguments: |
|
60 | arguments: | |
61 | root - the canonical root of the tree you're matching against |
|
61 | root - the canonical root of the tree you're matching against | |
62 | cwd - the current working directory, if relevant |
|
62 | cwd - the current working directory, if relevant | |
63 | patterns - patterns to find |
|
63 | patterns - patterns to find | |
64 | include - patterns to include (unless they are excluded) |
|
64 | include - patterns to include (unless they are excluded) | |
65 | exclude - patterns to exclude (even if they are included) |
|
65 | exclude - patterns to exclude (even if they are included) | |
66 | default - if a pattern in patterns has no explicit type, assume this one |
|
66 | default - if a pattern in patterns has no explicit type, assume this one | |
67 | exact - patterns are actually filenames (include/exclude still apply) |
|
67 | exact - patterns are actually filenames (include/exclude still apply) | |
68 | warn - optional function used for printing warnings |
|
68 | warn - optional function used for printing warnings | |
69 |
|
69 | |||
70 | a pattern is one of: |
|
70 | a pattern is one of: | |
71 | 'glob:<glob>' - a glob relative to cwd |
|
71 | 'glob:<glob>' - a glob relative to cwd | |
72 | 're:<regexp>' - a regular expression |
|
72 | 're:<regexp>' - a regular expression | |
73 | 'path:<path>' - a path relative to repository root |
|
73 | 'path:<path>' - a path relative to repository root | |
74 | 'relglob:<glob>' - an unrooted glob (*.c matches C files in all dirs) |
|
74 | 'relglob:<glob>' - an unrooted glob (*.c matches C files in all dirs) | |
75 | 'relpath:<path>' - a path relative to cwd |
|
75 | 'relpath:<path>' - a path relative to cwd | |
76 | 'relre:<regexp>' - a regexp that needn't match the start of a name |
|
76 | 'relre:<regexp>' - a regexp that needn't match the start of a name | |
77 | 'set:<fileset>' - a fileset expression |
|
77 | 'set:<fileset>' - a fileset expression | |
78 | 'include:<path>' - a file of patterns to read and include |
|
78 | 'include:<path>' - a file of patterns to read and include | |
79 | '<something>' - a pattern of the specified default type |
|
79 | '<something>' - a pattern of the specified default type | |
80 | """ |
|
80 | """ | |
81 |
|
81 | |||
82 | self._root = root |
|
82 | self._root = root | |
83 | self._cwd = cwd |
|
83 | self._cwd = cwd | |
84 | self._files = [] # exact files and roots of patterns |
|
84 | self._files = [] # exact files and roots of patterns | |
85 | self._anypats = bool(include or exclude) |
|
85 | self._anypats = bool(include or exclude) | |
86 | self._always = False |
|
86 | self._always = False | |
87 | self._pathrestricted = bool(include or exclude or patterns) |
|
87 | self._pathrestricted = bool(include or exclude or patterns) | |
88 | self._warn = warn |
|
88 | self._warn = warn | |
89 | self._includeroots = set() |
|
89 | self._includeroots = set() | |
90 | self._includedirs = set(['.']) |
|
90 | self._includedirs = set(['.']) | |
91 | self._excluderoots = set() |
|
91 | self._excluderoots = set() | |
92 |
|
92 | |||
93 | matchfns = [] |
|
93 | matchfns = [] | |
94 | if include: |
|
94 | if include: | |
95 | kindpats = self._normalize(include, 'glob', root, cwd, auditor) |
|
95 | kindpats = self._normalize(include, 'glob', root, cwd, auditor) | |
96 | self.includepat, im = _buildmatch(ctx, kindpats, '(?:/|$)', |
|
96 | self.includepat, im = _buildmatch(ctx, kindpats, '(?:/|$)', | |
97 | listsubrepos) |
|
97 | listsubrepos, root) | |
98 | self._includeroots.update(_roots(kindpats)) |
|
98 | self._includeroots.update(_roots(kindpats)) | |
99 | self._includeroots.discard('.') |
|
99 | self._includeroots.discard('.') | |
100 | self._includedirs.update(util.dirs(self._includeroots)) |
|
100 | self._includedirs.update(util.dirs(self._includeroots)) | |
101 | matchfns.append(im) |
|
101 | matchfns.append(im) | |
102 | if exclude: |
|
102 | if exclude: | |
103 | kindpats = self._normalize(exclude, 'glob', root, cwd, auditor) |
|
103 | kindpats = self._normalize(exclude, 'glob', root, cwd, auditor) | |
104 | self.excludepat, em = _buildmatch(ctx, kindpats, '(?:/|$)', |
|
104 | self.excludepat, em = _buildmatch(ctx, kindpats, '(?:/|$)', | |
105 | listsubrepos) |
|
105 | listsubrepos, root) | |
106 | self._excluderoots.update(_roots(kindpats)) |
|
106 | self._excluderoots.update(_roots(kindpats)) | |
107 | self._excluderoots.discard('.') |
|
107 | self._excluderoots.discard('.') | |
108 | matchfns.append(lambda f: not em(f)) |
|
108 | matchfns.append(lambda f: not em(f)) | |
109 | if exact: |
|
109 | if exact: | |
110 | if isinstance(patterns, list): |
|
110 | if isinstance(patterns, list): | |
111 | self._files = patterns |
|
111 | self._files = patterns | |
112 | else: |
|
112 | else: | |
113 | self._files = list(patterns) |
|
113 | self._files = list(patterns) | |
114 | matchfns.append(self.exact) |
|
114 | matchfns.append(self.exact) | |
115 | elif patterns: |
|
115 | elif patterns: | |
116 | kindpats = self._normalize(patterns, default, root, cwd, auditor) |
|
116 | kindpats = self._normalize(patterns, default, root, cwd, auditor) | |
117 | if not _kindpatsalwaysmatch(kindpats): |
|
117 | if not _kindpatsalwaysmatch(kindpats): | |
118 | self._files = _roots(kindpats) |
|
118 | self._files = _roots(kindpats) | |
119 | self._anypats = self._anypats or _anypats(kindpats) |
|
119 | self._anypats = self._anypats or _anypats(kindpats) | |
120 | self.patternspat, pm = _buildmatch(ctx, kindpats, '$', |
|
120 | self.patternspat, pm = _buildmatch(ctx, kindpats, '$', | |
121 | listsubrepos) |
|
121 | listsubrepos, root) | |
122 | matchfns.append(pm) |
|
122 | matchfns.append(pm) | |
123 |
|
123 | |||
124 | if not matchfns: |
|
124 | if not matchfns: | |
125 | m = util.always |
|
125 | m = util.always | |
126 | self._always = True |
|
126 | self._always = True | |
127 | elif len(matchfns) == 1: |
|
127 | elif len(matchfns) == 1: | |
128 | m = matchfns[0] |
|
128 | m = matchfns[0] | |
129 | else: |
|
129 | else: | |
130 | def m(f): |
|
130 | def m(f): | |
131 | for matchfn in matchfns: |
|
131 | for matchfn in matchfns: | |
132 | if not matchfn(f): |
|
132 | if not matchfn(f): | |
133 | return False |
|
133 | return False | |
134 | return True |
|
134 | return True | |
135 |
|
135 | |||
136 | self.matchfn = m |
|
136 | self.matchfn = m | |
137 | self._fileroots = set(self._files) |
|
137 | self._fileroots = set(self._files) | |
138 |
|
138 | |||
139 | def __call__(self, fn): |
|
139 | def __call__(self, fn): | |
140 | return self.matchfn(fn) |
|
140 | return self.matchfn(fn) | |
141 | def __iter__(self): |
|
141 | def __iter__(self): | |
142 | for f in self._files: |
|
142 | for f in self._files: | |
143 | yield f |
|
143 | yield f | |
144 |
|
144 | |||
145 | # Callbacks related to how the matcher is used by dirstate.walk. |
|
145 | # Callbacks related to how the matcher is used by dirstate.walk. | |
146 | # Subscribers to these events must monkeypatch the matcher object. |
|
146 | # Subscribers to these events must monkeypatch the matcher object. | |
147 | def bad(self, f, msg): |
|
147 | def bad(self, f, msg): | |
148 | '''Callback from dirstate.walk for each explicit file that can't be |
|
148 | '''Callback from dirstate.walk for each explicit file that can't be | |
149 | found/accessed, with an error message.''' |
|
149 | found/accessed, with an error message.''' | |
150 | pass |
|
150 | pass | |
151 |
|
151 | |||
152 | # If an explicitdir is set, it will be called when an explicitly listed |
|
152 | # If an explicitdir is set, it will be called when an explicitly listed | |
153 | # directory is visited. |
|
153 | # directory is visited. | |
154 | explicitdir = None |
|
154 | explicitdir = None | |
155 |
|
155 | |||
156 | # If an traversedir is set, it will be called when a directory discovered |
|
156 | # If an traversedir is set, it will be called when a directory discovered | |
157 | # by recursive traversal is visited. |
|
157 | # by recursive traversal is visited. | |
158 | traversedir = None |
|
158 | traversedir = None | |
159 |
|
159 | |||
160 | def abs(self, f): |
|
160 | def abs(self, f): | |
161 | '''Convert a repo path back to path that is relative to the root of the |
|
161 | '''Convert a repo path back to path that is relative to the root of the | |
162 | matcher.''' |
|
162 | matcher.''' | |
163 | return f |
|
163 | return f | |
164 |
|
164 | |||
165 | def rel(self, f): |
|
165 | def rel(self, f): | |
166 | '''Convert repo path back to path that is relative to cwd of matcher.''' |
|
166 | '''Convert repo path back to path that is relative to cwd of matcher.''' | |
167 | return util.pathto(self._root, self._cwd, f) |
|
167 | return util.pathto(self._root, self._cwd, f) | |
168 |
|
168 | |||
169 | def uipath(self, f): |
|
169 | def uipath(self, f): | |
170 | '''Convert repo path to a display path. If patterns or -I/-X were used |
|
170 | '''Convert repo path to a display path. If patterns or -I/-X were used | |
171 | to create this matcher, the display path will be relative to cwd. |
|
171 | to create this matcher, the display path will be relative to cwd. | |
172 | Otherwise it is relative to the root of the repo.''' |
|
172 | Otherwise it is relative to the root of the repo.''' | |
173 | return (self._pathrestricted and self.rel(f)) or self.abs(f) |
|
173 | return (self._pathrestricted and self.rel(f)) or self.abs(f) | |
174 |
|
174 | |||
175 | def files(self): |
|
175 | def files(self): | |
176 | '''Explicitly listed files or patterns or roots: |
|
176 | '''Explicitly listed files or patterns or roots: | |
177 | if no patterns or .always(): empty list, |
|
177 | if no patterns or .always(): empty list, | |
178 | if exact: list exact files, |
|
178 | if exact: list exact files, | |
179 | if not .anypats(): list all files and dirs, |
|
179 | if not .anypats(): list all files and dirs, | |
180 | else: optimal roots''' |
|
180 | else: optimal roots''' | |
181 | return self._files |
|
181 | return self._files | |
182 |
|
182 | |||
183 | @propertycache |
|
183 | @propertycache | |
184 | def _dirs(self): |
|
184 | def _dirs(self): | |
185 | return set(util.dirs(self._fileroots)) | set(['.']) |
|
185 | return set(util.dirs(self._fileroots)) | set(['.']) | |
186 |
|
186 | |||
187 | def visitdir(self, dir): |
|
187 | def visitdir(self, dir): | |
188 | '''Decides whether a directory should be visited based on whether it |
|
188 | '''Decides whether a directory should be visited based on whether it | |
189 | has potential matches in it or one of its subdirectories. This is |
|
189 | has potential matches in it or one of its subdirectories. This is | |
190 | based on the match's primary, included, and excluded patterns. |
|
190 | based on the match's primary, included, and excluded patterns. | |
191 |
|
191 | |||
192 | This function's behavior is undefined if it has returned False for |
|
192 | This function's behavior is undefined if it has returned False for | |
193 | one of the dir's parent directories. |
|
193 | one of the dir's parent directories. | |
194 | ''' |
|
194 | ''' | |
195 | if dir in self._excluderoots: |
|
195 | if dir in self._excluderoots: | |
196 | return False |
|
196 | return False | |
197 | parentdirs = None |
|
197 | parentdirs = None | |
198 | if (self._includeroots and dir not in self._includeroots and |
|
198 | if (self._includeroots and dir not in self._includeroots and | |
199 | dir not in self._includedirs): |
|
199 | dir not in self._includedirs): | |
200 | parentdirs = util.finddirs(dir) |
|
200 | parentdirs = util.finddirs(dir) | |
201 | if not any(parent in self._includeroots for parent in parentdirs): |
|
201 | if not any(parent in self._includeroots for parent in parentdirs): | |
202 | return False |
|
202 | return False | |
203 | return (not self._fileroots or '.' in self._fileroots or |
|
203 | return (not self._fileroots or '.' in self._fileroots or | |
204 | dir in self._fileroots or dir in self._dirs or |
|
204 | dir in self._fileroots or dir in self._dirs or | |
205 | any(parentdir in self._fileroots |
|
205 | any(parentdir in self._fileroots | |
206 | for parentdir in parentdirs or util.finddirs(dir))) |
|
206 | for parentdir in parentdirs or util.finddirs(dir))) | |
207 |
|
207 | |||
208 | def exact(self, f): |
|
208 | def exact(self, f): | |
209 | '''Returns True if f is in .files().''' |
|
209 | '''Returns True if f is in .files().''' | |
210 | return f in self._fileroots |
|
210 | return f in self._fileroots | |
211 |
|
211 | |||
212 | def anypats(self): |
|
212 | def anypats(self): | |
213 | '''Matcher uses patterns or include/exclude.''' |
|
213 | '''Matcher uses patterns or include/exclude.''' | |
214 | return self._anypats |
|
214 | return self._anypats | |
215 |
|
215 | |||
216 | def always(self): |
|
216 | def always(self): | |
217 | '''Matcher will match everything and .files() will be empty |
|
217 | '''Matcher will match everything and .files() will be empty | |
218 | - optimization might be possible and necessary.''' |
|
218 | - optimization might be possible and necessary.''' | |
219 | return self._always |
|
219 | return self._always | |
220 |
|
220 | |||
221 | def ispartial(self): |
|
221 | def ispartial(self): | |
222 | '''True if the matcher won't always match. |
|
222 | '''True if the matcher won't always match. | |
223 |
|
223 | |||
224 | Although it's just the inverse of _always in this implementation, |
|
224 | Although it's just the inverse of _always in this implementation, | |
225 | an extenion such as narrowhg might make it return something |
|
225 | an extenion such as narrowhg might make it return something | |
226 | slightly different.''' |
|
226 | slightly different.''' | |
227 | return not self._always |
|
227 | return not self._always | |
228 |
|
228 | |||
229 | def isexact(self): |
|
229 | def isexact(self): | |
230 | return self.matchfn == self.exact |
|
230 | return self.matchfn == self.exact | |
231 |
|
231 | |||
232 | def prefix(self): |
|
232 | def prefix(self): | |
233 | return not self.always() and not self.isexact() and not self.anypats() |
|
233 | return not self.always() and not self.isexact() and not self.anypats() | |
234 |
|
234 | |||
235 | def _normalize(self, patterns, default, root, cwd, auditor): |
|
235 | def _normalize(self, patterns, default, root, cwd, auditor): | |
236 | '''Convert 'kind:pat' from the patterns list to tuples with kind and |
|
236 | '''Convert 'kind:pat' from the patterns list to tuples with kind and | |
237 | normalized and rooted patterns and with listfiles expanded.''' |
|
237 | normalized and rooted patterns and with listfiles expanded.''' | |
238 | kindpats = [] |
|
238 | kindpats = [] | |
239 | for kind, pat in [_patsplit(p, default) for p in patterns]: |
|
239 | for kind, pat in [_patsplit(p, default) for p in patterns]: | |
240 | if kind in ('glob', 'relpath'): |
|
240 | if kind in ('glob', 'relpath'): | |
241 | pat = pathutil.canonpath(root, cwd, pat, auditor) |
|
241 | pat = pathutil.canonpath(root, cwd, pat, auditor) | |
242 | elif kind in ('relglob', 'path'): |
|
242 | elif kind in ('relglob', 'path'): | |
243 | pat = util.normpath(pat) |
|
243 | pat = util.normpath(pat) | |
244 | elif kind in ('listfile', 'listfile0'): |
|
244 | elif kind in ('listfile', 'listfile0'): | |
245 | try: |
|
245 | try: | |
246 | files = util.readfile(pat) |
|
246 | files = util.readfile(pat) | |
247 | if kind == 'listfile0': |
|
247 | if kind == 'listfile0': | |
248 | files = files.split('\0') |
|
248 | files = files.split('\0') | |
249 | else: |
|
249 | else: | |
250 | files = files.splitlines() |
|
250 | files = files.splitlines() | |
251 | files = [f for f in files if f] |
|
251 | files = [f for f in files if f] | |
252 | except EnvironmentError: |
|
252 | except EnvironmentError: | |
253 | raise util.Abort(_("unable to read file list (%s)") % pat) |
|
253 | raise util.Abort(_("unable to read file list (%s)") % pat) | |
254 | for k, p, source in self._normalize(files, default, root, cwd, |
|
254 | for k, p, source in self._normalize(files, default, root, cwd, | |
255 | auditor): |
|
255 | auditor): | |
256 | kindpats.append((k, p, pat)) |
|
256 | kindpats.append((k, p, pat)) | |
257 | continue |
|
257 | continue | |
258 | elif kind == 'include': |
|
258 | elif kind == 'include': | |
259 | try: |
|
259 | try: | |
260 | includepats = readpatternfile(pat, self._warn) |
|
260 | includepats = readpatternfile(pat, self._warn) | |
261 | for k, p, source in self._normalize(includepats, default, |
|
261 | for k, p, source in self._normalize(includepats, default, | |
262 | root, cwd, auditor): |
|
262 | root, cwd, auditor): | |
263 | kindpats.append((k, p, source or pat)) |
|
263 | kindpats.append((k, p, source or pat)) | |
264 | except util.Abort, inst: |
|
264 | except util.Abort, inst: | |
265 | raise util.Abort('%s: %s' % (pat, inst[0])) |
|
265 | raise util.Abort('%s: %s' % (pat, inst[0])) | |
266 | except IOError, inst: |
|
266 | except IOError, inst: | |
267 | if self._warn: |
|
267 | if self._warn: | |
268 | self._warn(_("skipping unreadable pattern file " |
|
268 | self._warn(_("skipping unreadable pattern file " | |
269 | "'%s': %s\n") % (pat, inst.strerror)) |
|
269 | "'%s': %s\n") % (pat, inst.strerror)) | |
270 | continue |
|
270 | continue | |
271 | # else: re or relre - which cannot be normalized |
|
271 | # else: re or relre - which cannot be normalized | |
272 | kindpats.append((kind, pat, '')) |
|
272 | kindpats.append((kind, pat, '')) | |
273 | return kindpats |
|
273 | return kindpats | |
274 |
|
274 | |||
275 | def exact(root, cwd, files): |
|
275 | def exact(root, cwd, files): | |
276 | return match(root, cwd, files, exact=True) |
|
276 | return match(root, cwd, files, exact=True) | |
277 |
|
277 | |||
278 | def always(root, cwd): |
|
278 | def always(root, cwd): | |
279 | return match(root, cwd, []) |
|
279 | return match(root, cwd, []) | |
280 |
|
280 | |||
281 | class narrowmatcher(match): |
|
281 | class narrowmatcher(match): | |
282 | """Adapt a matcher to work on a subdirectory only. |
|
282 | """Adapt a matcher to work on a subdirectory only. | |
283 |
|
283 | |||
284 | The paths are remapped to remove/insert the path as needed: |
|
284 | The paths are remapped to remove/insert the path as needed: | |
285 |
|
285 | |||
286 | >>> m1 = match('root', '', ['a.txt', 'sub/b.txt']) |
|
286 | >>> m1 = match('root', '', ['a.txt', 'sub/b.txt']) | |
287 | >>> m2 = narrowmatcher('sub', m1) |
|
287 | >>> m2 = narrowmatcher('sub', m1) | |
288 | >>> bool(m2('a.txt')) |
|
288 | >>> bool(m2('a.txt')) | |
289 | False |
|
289 | False | |
290 | >>> bool(m2('b.txt')) |
|
290 | >>> bool(m2('b.txt')) | |
291 | True |
|
291 | True | |
292 | >>> bool(m2.matchfn('a.txt')) |
|
292 | >>> bool(m2.matchfn('a.txt')) | |
293 | False |
|
293 | False | |
294 | >>> bool(m2.matchfn('b.txt')) |
|
294 | >>> bool(m2.matchfn('b.txt')) | |
295 | True |
|
295 | True | |
296 | >>> m2.files() |
|
296 | >>> m2.files() | |
297 | ['b.txt'] |
|
297 | ['b.txt'] | |
298 | >>> m2.exact('b.txt') |
|
298 | >>> m2.exact('b.txt') | |
299 | True |
|
299 | True | |
300 | >>> util.pconvert(m2.rel('b.txt')) |
|
300 | >>> util.pconvert(m2.rel('b.txt')) | |
301 | 'sub/b.txt' |
|
301 | 'sub/b.txt' | |
302 | >>> def bad(f, msg): |
|
302 | >>> def bad(f, msg): | |
303 | ... print "%s: %s" % (f, msg) |
|
303 | ... print "%s: %s" % (f, msg) | |
304 | >>> m1.bad = bad |
|
304 | >>> m1.bad = bad | |
305 | >>> m2.bad('x.txt', 'No such file') |
|
305 | >>> m2.bad('x.txt', 'No such file') | |
306 | sub/x.txt: No such file |
|
306 | sub/x.txt: No such file | |
307 | >>> m2.abs('c.txt') |
|
307 | >>> m2.abs('c.txt') | |
308 | 'sub/c.txt' |
|
308 | 'sub/c.txt' | |
309 | """ |
|
309 | """ | |
310 |
|
310 | |||
311 | def __init__(self, path, matcher): |
|
311 | def __init__(self, path, matcher): | |
312 | self._root = matcher._root |
|
312 | self._root = matcher._root | |
313 | self._cwd = matcher._cwd |
|
313 | self._cwd = matcher._cwd | |
314 | self._path = path |
|
314 | self._path = path | |
315 | self._matcher = matcher |
|
315 | self._matcher = matcher | |
316 | self._always = matcher._always |
|
316 | self._always = matcher._always | |
317 | self._pathrestricted = matcher._pathrestricted |
|
317 | self._pathrestricted = matcher._pathrestricted | |
318 |
|
318 | |||
319 | self._files = [f[len(path) + 1:] for f in matcher._files |
|
319 | self._files = [f[len(path) + 1:] for f in matcher._files | |
320 | if f.startswith(path + "/")] |
|
320 | if f.startswith(path + "/")] | |
321 |
|
321 | |||
322 | # If the parent repo had a path to this subrepo and no patterns are |
|
322 | # If the parent repo had a path to this subrepo and no patterns are | |
323 | # specified, this submatcher always matches. |
|
323 | # specified, this submatcher always matches. | |
324 | if not self._always and not matcher._anypats: |
|
324 | if not self._always and not matcher._anypats: | |
325 | self._always = any(f == path for f in matcher._files) |
|
325 | self._always = any(f == path for f in matcher._files) | |
326 |
|
326 | |||
327 | self._anypats = matcher._anypats |
|
327 | self._anypats = matcher._anypats | |
328 | self.matchfn = lambda fn: matcher.matchfn(self._path + "/" + fn) |
|
328 | self.matchfn = lambda fn: matcher.matchfn(self._path + "/" + fn) | |
329 | self._fileroots = set(self._files) |
|
329 | self._fileroots = set(self._files) | |
330 |
|
330 | |||
331 | def abs(self, f): |
|
331 | def abs(self, f): | |
332 | return self._matcher.abs(self._path + "/" + f) |
|
332 | return self._matcher.abs(self._path + "/" + f) | |
333 |
|
333 | |||
334 | def bad(self, f, msg): |
|
334 | def bad(self, f, msg): | |
335 | self._matcher.bad(self._path + "/" + f, msg) |
|
335 | self._matcher.bad(self._path + "/" + f, msg) | |
336 |
|
336 | |||
337 | def rel(self, f): |
|
337 | def rel(self, f): | |
338 | return self._matcher.rel(self._path + "/" + f) |
|
338 | return self._matcher.rel(self._path + "/" + f) | |
339 |
|
339 | |||
340 | class icasefsmatcher(match): |
|
340 | class icasefsmatcher(match): | |
341 | """A matcher for wdir on case insensitive filesystems, which normalizes the |
|
341 | """A matcher for wdir on case insensitive filesystems, which normalizes the | |
342 | given patterns to the case in the filesystem. |
|
342 | given patterns to the case in the filesystem. | |
343 | """ |
|
343 | """ | |
344 |
|
344 | |||
345 | def __init__(self, root, cwd, patterns, include, exclude, default, auditor, |
|
345 | def __init__(self, root, cwd, patterns, include, exclude, default, auditor, | |
346 | ctx, listsubrepos=False): |
|
346 | ctx, listsubrepos=False): | |
347 | init = super(icasefsmatcher, self).__init__ |
|
347 | init = super(icasefsmatcher, self).__init__ | |
348 | self._dsnormalize = ctx.repo().dirstate.normalize |
|
348 | self._dsnormalize = ctx.repo().dirstate.normalize | |
349 |
|
349 | |||
350 | init(root, cwd, patterns, include, exclude, default, auditor=auditor, |
|
350 | init(root, cwd, patterns, include, exclude, default, auditor=auditor, | |
351 | ctx=ctx, listsubrepos=listsubrepos) |
|
351 | ctx=ctx, listsubrepos=listsubrepos) | |
352 |
|
352 | |||
353 | # m.exact(file) must be based off of the actual user input, otherwise |
|
353 | # m.exact(file) must be based off of the actual user input, otherwise | |
354 | # inexact case matches are treated as exact, and not noted without -v. |
|
354 | # inexact case matches are treated as exact, and not noted without -v. | |
355 | if self._files: |
|
355 | if self._files: | |
356 | self._fileroots = set(_roots(self._kp)) |
|
356 | self._fileroots = set(_roots(self._kp)) | |
357 |
|
357 | |||
358 | def _normalize(self, patterns, default, root, cwd, auditor): |
|
358 | def _normalize(self, patterns, default, root, cwd, auditor): | |
359 | self._kp = super(icasefsmatcher, self)._normalize(patterns, default, |
|
359 | self._kp = super(icasefsmatcher, self)._normalize(patterns, default, | |
360 | root, cwd, auditor) |
|
360 | root, cwd, auditor) | |
361 | kindpats = [] |
|
361 | kindpats = [] | |
362 | for kind, pats, source in self._kp: |
|
362 | for kind, pats, source in self._kp: | |
363 | if kind not in ('re', 'relre'): # regex can't be normalized |
|
363 | if kind not in ('re', 'relre'): # regex can't be normalized | |
364 | pats = self._dsnormalize(pats) |
|
364 | pats = self._dsnormalize(pats) | |
365 | kindpats.append((kind, pats, source)) |
|
365 | kindpats.append((kind, pats, source)) | |
366 | return kindpats |
|
366 | return kindpats | |
367 |
|
367 | |||
368 | def patkind(pattern, default=None): |
|
368 | def patkind(pattern, default=None): | |
369 | '''If pattern is 'kind:pat' with a known kind, return kind.''' |
|
369 | '''If pattern is 'kind:pat' with a known kind, return kind.''' | |
370 | return _patsplit(pattern, default)[0] |
|
370 | return _patsplit(pattern, default)[0] | |
371 |
|
371 | |||
372 | def _patsplit(pattern, default): |
|
372 | def _patsplit(pattern, default): | |
373 | """Split a string into the optional pattern kind prefix and the actual |
|
373 | """Split a string into the optional pattern kind prefix and the actual | |
374 | pattern.""" |
|
374 | pattern.""" | |
375 | if ':' in pattern: |
|
375 | if ':' in pattern: | |
376 | kind, pat = pattern.split(':', 1) |
|
376 | kind, pat = pattern.split(':', 1) | |
377 | if kind in ('re', 'glob', 'path', 'relglob', 'relpath', 'relre', |
|
377 | if kind in ('re', 'glob', 'path', 'relglob', 'relpath', 'relre', | |
378 | 'listfile', 'listfile0', 'set', 'include'): |
|
378 | 'listfile', 'listfile0', 'set', 'include'): | |
379 | return kind, pat |
|
379 | return kind, pat | |
380 | return default, pattern |
|
380 | return default, pattern | |
381 |
|
381 | |||
382 | def _globre(pat): |
|
382 | def _globre(pat): | |
383 | r'''Convert an extended glob string to a regexp string. |
|
383 | r'''Convert an extended glob string to a regexp string. | |
384 |
|
384 | |||
385 | >>> print _globre(r'?') |
|
385 | >>> print _globre(r'?') | |
386 | . |
|
386 | . | |
387 | >>> print _globre(r'*') |
|
387 | >>> print _globre(r'*') | |
388 | [^/]* |
|
388 | [^/]* | |
389 | >>> print _globre(r'**') |
|
389 | >>> print _globre(r'**') | |
390 | .* |
|
390 | .* | |
391 | >>> print _globre(r'**/a') |
|
391 | >>> print _globre(r'**/a') | |
392 | (?:.*/)?a |
|
392 | (?:.*/)?a | |
393 | >>> print _globre(r'a/**/b') |
|
393 | >>> print _globre(r'a/**/b') | |
394 | a\/(?:.*/)?b |
|
394 | a\/(?:.*/)?b | |
395 | >>> print _globre(r'[a*?!^][^b][!c]') |
|
395 | >>> print _globre(r'[a*?!^][^b][!c]') | |
396 | [a*?!^][\^b][^c] |
|
396 | [a*?!^][\^b][^c] | |
397 | >>> print _globre(r'{a,b}') |
|
397 | >>> print _globre(r'{a,b}') | |
398 | (?:a|b) |
|
398 | (?:a|b) | |
399 | >>> print _globre(r'.\*\?') |
|
399 | >>> print _globre(r'.\*\?') | |
400 | \.\*\? |
|
400 | \.\*\? | |
401 | ''' |
|
401 | ''' | |
402 | i, n = 0, len(pat) |
|
402 | i, n = 0, len(pat) | |
403 | res = '' |
|
403 | res = '' | |
404 | group = 0 |
|
404 | group = 0 | |
405 | escape = util.re.escape |
|
405 | escape = util.re.escape | |
406 | def peek(): |
|
406 | def peek(): | |
407 | return i < n and pat[i] |
|
407 | return i < n and pat[i] | |
408 | while i < n: |
|
408 | while i < n: | |
409 | c = pat[i] |
|
409 | c = pat[i] | |
410 | i += 1 |
|
410 | i += 1 | |
411 | if c not in '*?[{},\\': |
|
411 | if c not in '*?[{},\\': | |
412 | res += escape(c) |
|
412 | res += escape(c) | |
413 | elif c == '*': |
|
413 | elif c == '*': | |
414 | if peek() == '*': |
|
414 | if peek() == '*': | |
415 | i += 1 |
|
415 | i += 1 | |
416 | if peek() == '/': |
|
416 | if peek() == '/': | |
417 | i += 1 |
|
417 | i += 1 | |
418 | res += '(?:.*/)?' |
|
418 | res += '(?:.*/)?' | |
419 | else: |
|
419 | else: | |
420 | res += '.*' |
|
420 | res += '.*' | |
421 | else: |
|
421 | else: | |
422 | res += '[^/]*' |
|
422 | res += '[^/]*' | |
423 | elif c == '?': |
|
423 | elif c == '?': | |
424 | res += '.' |
|
424 | res += '.' | |
425 | elif c == '[': |
|
425 | elif c == '[': | |
426 | j = i |
|
426 | j = i | |
427 | if j < n and pat[j] in '!]': |
|
427 | if j < n and pat[j] in '!]': | |
428 | j += 1 |
|
428 | j += 1 | |
429 | while j < n and pat[j] != ']': |
|
429 | while j < n and pat[j] != ']': | |
430 | j += 1 |
|
430 | j += 1 | |
431 | if j >= n: |
|
431 | if j >= n: | |
432 | res += '\\[' |
|
432 | res += '\\[' | |
433 | else: |
|
433 | else: | |
434 | stuff = pat[i:j].replace('\\','\\\\') |
|
434 | stuff = pat[i:j].replace('\\','\\\\') | |
435 | i = j + 1 |
|
435 | i = j + 1 | |
436 | if stuff[0] == '!': |
|
436 | if stuff[0] == '!': | |
437 | stuff = '^' + stuff[1:] |
|
437 | stuff = '^' + stuff[1:] | |
438 | elif stuff[0] == '^': |
|
438 | elif stuff[0] == '^': | |
439 | stuff = '\\' + stuff |
|
439 | stuff = '\\' + stuff | |
440 | res = '%s[%s]' % (res, stuff) |
|
440 | res = '%s[%s]' % (res, stuff) | |
441 | elif c == '{': |
|
441 | elif c == '{': | |
442 | group += 1 |
|
442 | group += 1 | |
443 | res += '(?:' |
|
443 | res += '(?:' | |
444 | elif c == '}' and group: |
|
444 | elif c == '}' and group: | |
445 | res += ')' |
|
445 | res += ')' | |
446 | group -= 1 |
|
446 | group -= 1 | |
447 | elif c == ',' and group: |
|
447 | elif c == ',' and group: | |
448 | res += '|' |
|
448 | res += '|' | |
449 | elif c == '\\': |
|
449 | elif c == '\\': | |
450 | p = peek() |
|
450 | p = peek() | |
451 | if p: |
|
451 | if p: | |
452 | i += 1 |
|
452 | i += 1 | |
453 | res += escape(p) |
|
453 | res += escape(p) | |
454 | else: |
|
454 | else: | |
455 | res += escape(c) |
|
455 | res += escape(c) | |
456 | else: |
|
456 | else: | |
457 | res += escape(c) |
|
457 | res += escape(c) | |
458 | return res |
|
458 | return res | |
459 |
|
459 | |||
460 | def _regex(kind, pat, globsuffix): |
|
460 | def _regex(kind, pat, globsuffix): | |
461 | '''Convert a (normalized) pattern of any kind into a regular expression. |
|
461 | '''Convert a (normalized) pattern of any kind into a regular expression. | |
462 | globsuffix is appended to the regexp of globs.''' |
|
462 | globsuffix is appended to the regexp of globs.''' | |
463 | if not pat: |
|
463 | if not pat: | |
464 | return '' |
|
464 | return '' | |
465 | if kind == 're': |
|
465 | if kind == 're': | |
466 | return pat |
|
466 | return pat | |
467 | if kind == 'path': |
|
467 | if kind == 'path': | |
468 | return '^' + util.re.escape(pat) + '(?:/|$)' |
|
468 | return '^' + util.re.escape(pat) + '(?:/|$)' | |
469 | if kind == 'relglob': |
|
469 | if kind == 'relglob': | |
470 | return '(?:|.*/)' + _globre(pat) + globsuffix |
|
470 | return '(?:|.*/)' + _globre(pat) + globsuffix | |
471 | if kind == 'relpath': |
|
471 | if kind == 'relpath': | |
472 | return util.re.escape(pat) + '(?:/|$)' |
|
472 | return util.re.escape(pat) + '(?:/|$)' | |
473 | if kind == 'relre': |
|
473 | if kind == 'relre': | |
474 | if pat.startswith('^'): |
|
474 | if pat.startswith('^'): | |
475 | return pat |
|
475 | return pat | |
476 | return '.*' + pat |
|
476 | return '.*' + pat | |
477 | return _globre(pat) + globsuffix |
|
477 | return _globre(pat) + globsuffix | |
478 |
|
478 | |||
479 | def _buildmatch(ctx, kindpats, globsuffix, listsubrepos): |
|
479 | def _buildmatch(ctx, kindpats, globsuffix, listsubrepos, root): | |
480 | '''Return regexp string and a matcher function for kindpats. |
|
480 | '''Return regexp string and a matcher function for kindpats. | |
481 | globsuffix is appended to the regexp of globs.''' |
|
481 | globsuffix is appended to the regexp of globs.''' | |
482 | fset, kindpats = _expandsets(kindpats, ctx, listsubrepos) |
|
482 | fset, kindpats = _expandsets(kindpats, ctx, listsubrepos) | |
483 | if not kindpats: |
|
483 | if not kindpats: | |
484 | return "", fset.__contains__ |
|
484 | return "", fset.__contains__ | |
485 |
|
485 | |||
486 | regex, mf = _buildregexmatch(kindpats, globsuffix) |
|
486 | regex, mf = _buildregexmatch(kindpats, globsuffix) | |
487 | if fset: |
|
487 | if fset: | |
488 | return regex, lambda f: f in fset or mf(f) |
|
488 | return regex, lambda f: f in fset or mf(f) | |
489 | return regex, mf |
|
489 | return regex, mf | |
490 |
|
490 | |||
491 | def _buildregexmatch(kindpats, globsuffix): |
|
491 | def _buildregexmatch(kindpats, globsuffix): | |
492 | """Build a match function from a list of kinds and kindpats, |
|
492 | """Build a match function from a list of kinds and kindpats, | |
493 | return regexp string and a matcher function.""" |
|
493 | return regexp string and a matcher function.""" | |
494 | try: |
|
494 | try: | |
495 | regex = '(?:%s)' % '|'.join([_regex(k, p, globsuffix) |
|
495 | regex = '(?:%s)' % '|'.join([_regex(k, p, globsuffix) | |
496 | for (k, p, s) in kindpats]) |
|
496 | for (k, p, s) in kindpats]) | |
497 | if len(regex) > 20000: |
|
497 | if len(regex) > 20000: | |
498 | raise OverflowError |
|
498 | raise OverflowError | |
499 | return regex, _rematcher(regex) |
|
499 | return regex, _rematcher(regex) | |
500 | except OverflowError: |
|
500 | except OverflowError: | |
501 | # We're using a Python with a tiny regex engine and we |
|
501 | # We're using a Python with a tiny regex engine and we | |
502 | # made it explode, so we'll divide the pattern list in two |
|
502 | # made it explode, so we'll divide the pattern list in two | |
503 | # until it works |
|
503 | # until it works | |
504 | l = len(kindpats) |
|
504 | l = len(kindpats) | |
505 | if l < 2: |
|
505 | if l < 2: | |
506 | raise |
|
506 | raise | |
507 | regexa, a = _buildregexmatch(kindpats[:l//2], globsuffix) |
|
507 | regexa, a = _buildregexmatch(kindpats[:l//2], globsuffix) | |
508 | regexb, b = _buildregexmatch(kindpats[l//2:], globsuffix) |
|
508 | regexb, b = _buildregexmatch(kindpats[l//2:], globsuffix) | |
509 | return regex, lambda s: a(s) or b(s) |
|
509 | return regex, lambda s: a(s) or b(s) | |
510 | except re.error: |
|
510 | except re.error: | |
511 | for k, p, s in kindpats: |
|
511 | for k, p, s in kindpats: | |
512 | try: |
|
512 | try: | |
513 | _rematcher('(?:%s)' % _regex(k, p, globsuffix)) |
|
513 | _rematcher('(?:%s)' % _regex(k, p, globsuffix)) | |
514 | except re.error: |
|
514 | except re.error: | |
515 | if s: |
|
515 | if s: | |
516 | raise util.Abort(_("%s: invalid pattern (%s): %s") % |
|
516 | raise util.Abort(_("%s: invalid pattern (%s): %s") % | |
517 | (s, k, p)) |
|
517 | (s, k, p)) | |
518 | else: |
|
518 | else: | |
519 | raise util.Abort(_("invalid pattern (%s): %s") % (k, p)) |
|
519 | raise util.Abort(_("invalid pattern (%s): %s") % (k, p)) | |
520 | raise util.Abort(_("invalid pattern")) |
|
520 | raise util.Abort(_("invalid pattern")) | |
521 |
|
521 | |||
522 | def _roots(kindpats): |
|
522 | def _roots(kindpats): | |
523 | '''return roots and exact explicitly listed files from patterns |
|
523 | '''return roots and exact explicitly listed files from patterns | |
524 |
|
524 | |||
525 | >>> _roots([('glob', 'g/*', ''), ('glob', 'g', ''), ('glob', 'g*', '')]) |
|
525 | >>> _roots([('glob', 'g/*', ''), ('glob', 'g', ''), ('glob', 'g*', '')]) | |
526 | ['g', 'g', '.'] |
|
526 | ['g', 'g', '.'] | |
527 | >>> _roots([('relpath', 'r', ''), ('path', 'p/p', ''), ('path', '', '')]) |
|
527 | >>> _roots([('relpath', 'r', ''), ('path', 'p/p', ''), ('path', '', '')]) | |
528 | ['r', 'p/p', '.'] |
|
528 | ['r', 'p/p', '.'] | |
529 | >>> _roots([('relglob', 'rg*', ''), ('re', 're/', ''), ('relre', 'rr', '')]) |
|
529 | >>> _roots([('relglob', 'rg*', ''), ('re', 're/', ''), ('relre', 'rr', '')]) | |
530 | ['.', '.', '.'] |
|
530 | ['.', '.', '.'] | |
531 | ''' |
|
531 | ''' | |
532 | r = [] |
|
532 | r = [] | |
533 | for kind, pat, source in kindpats: |
|
533 | for kind, pat, source in kindpats: | |
534 | if kind == 'glob': # find the non-glob prefix |
|
534 | if kind == 'glob': # find the non-glob prefix | |
535 | root = [] |
|
535 | root = [] | |
536 | for p in pat.split('/'): |
|
536 | for p in pat.split('/'): | |
537 | if '[' in p or '{' in p or '*' in p or '?' in p: |
|
537 | if '[' in p or '{' in p or '*' in p or '?' in p: | |
538 | break |
|
538 | break | |
539 | root.append(p) |
|
539 | root.append(p) | |
540 | r.append('/'.join(root) or '.') |
|
540 | r.append('/'.join(root) or '.') | |
541 | elif kind in ('relpath', 'path'): |
|
541 | elif kind in ('relpath', 'path'): | |
542 | r.append(pat or '.') |
|
542 | r.append(pat or '.') | |
543 | else: # relglob, re, relre |
|
543 | else: # relglob, re, relre | |
544 | r.append('.') |
|
544 | r.append('.') | |
545 | return r |
|
545 | return r | |
546 |
|
546 | |||
547 | def _anypats(kindpats): |
|
547 | def _anypats(kindpats): | |
548 | for kind, pat, source in kindpats: |
|
548 | for kind, pat, source in kindpats: | |
549 | if kind in ('glob', 're', 'relglob', 'relre', 'set'): |
|
549 | if kind in ('glob', 're', 'relglob', 'relre', 'set'): | |
550 | return True |
|
550 | return True | |
551 |
|
551 | |||
552 | _commentre = None |
|
552 | _commentre = None | |
553 |
|
553 | |||
554 | def readpatternfile(filepath, warn): |
|
554 | def readpatternfile(filepath, warn): | |
555 | '''parse a pattern file, returning a list of |
|
555 | '''parse a pattern file, returning a list of | |
556 | patterns. These patterns should be given to compile() |
|
556 | patterns. These patterns should be given to compile() | |
557 | to be validated and converted into a match function. |
|
557 | to be validated and converted into a match function. | |
558 |
|
558 | |||
559 | trailing white space is dropped. |
|
559 | trailing white space is dropped. | |
560 | the escape character is backslash. |
|
560 | the escape character is backslash. | |
561 | comments start with #. |
|
561 | comments start with #. | |
562 | empty lines are skipped. |
|
562 | empty lines are skipped. | |
563 |
|
563 | |||
564 | lines can be of the following formats: |
|
564 | lines can be of the following formats: | |
565 |
|
565 | |||
566 | syntax: regexp # defaults following lines to non-rooted regexps |
|
566 | syntax: regexp # defaults following lines to non-rooted regexps | |
567 | syntax: glob # defaults following lines to non-rooted globs |
|
567 | syntax: glob # defaults following lines to non-rooted globs | |
568 | re:pattern # non-rooted regular expression |
|
568 | re:pattern # non-rooted regular expression | |
569 | glob:pattern # non-rooted glob |
|
569 | glob:pattern # non-rooted glob | |
570 | pattern # pattern of the current default type''' |
|
570 | pattern # pattern of the current default type''' | |
571 |
|
571 | |||
572 | syntaxes = {'re': 'relre:', 'regexp': 'relre:', 'glob': 'relglob:', |
|
572 | syntaxes = {'re': 'relre:', 'regexp': 'relre:', 'glob': 'relglob:', | |
573 | 'include': 'include'} |
|
573 | 'include': 'include'} | |
574 | syntax = 'relre:' |
|
574 | syntax = 'relre:' | |
575 | patterns = [] |
|
575 | patterns = [] | |
576 |
|
576 | |||
577 | fp = open(filepath) |
|
577 | fp = open(filepath) | |
578 | for line in fp: |
|
578 | for line in fp: | |
579 | if "#" in line: |
|
579 | if "#" in line: | |
580 | global _commentre |
|
580 | global _commentre | |
581 | if not _commentre: |
|
581 | if not _commentre: | |
582 | _commentre = re.compile(r'((^|[^\\])(\\\\)*)#.*') |
|
582 | _commentre = re.compile(r'((^|[^\\])(\\\\)*)#.*') | |
583 | # remove comments prefixed by an even number of escapes |
|
583 | # remove comments prefixed by an even number of escapes | |
584 | line = _commentre.sub(r'\1', line) |
|
584 | line = _commentre.sub(r'\1', line) | |
585 | # fixup properly escaped comments that survived the above |
|
585 | # fixup properly escaped comments that survived the above | |
586 | line = line.replace("\\#", "#") |
|
586 | line = line.replace("\\#", "#") | |
587 | line = line.rstrip() |
|
587 | line = line.rstrip() | |
588 | if not line: |
|
588 | if not line: | |
589 | continue |
|
589 | continue | |
590 |
|
590 | |||
591 | if line.startswith('syntax:'): |
|
591 | if line.startswith('syntax:'): | |
592 | s = line[7:].strip() |
|
592 | s = line[7:].strip() | |
593 | try: |
|
593 | try: | |
594 | syntax = syntaxes[s] |
|
594 | syntax = syntaxes[s] | |
595 | except KeyError: |
|
595 | except KeyError: | |
596 | if warn: |
|
596 | if warn: | |
597 | warn(_("%s: ignoring invalid syntax '%s'\n") % |
|
597 | warn(_("%s: ignoring invalid syntax '%s'\n") % | |
598 | (filepath, s)) |
|
598 | (filepath, s)) | |
599 | continue |
|
599 | continue | |
600 |
|
600 | |||
601 | linesyntax = syntax |
|
601 | linesyntax = syntax | |
602 | for s, rels in syntaxes.iteritems(): |
|
602 | for s, rels in syntaxes.iteritems(): | |
603 | if line.startswith(rels): |
|
603 | if line.startswith(rels): | |
604 | linesyntax = rels |
|
604 | linesyntax = rels | |
605 | line = line[len(rels):] |
|
605 | line = line[len(rels):] | |
606 | break |
|
606 | break | |
607 | elif line.startswith(s+':'): |
|
607 | elif line.startswith(s+':'): | |
608 | linesyntax = rels |
|
608 | linesyntax = rels | |
609 | line = line[len(s) + 1:] |
|
609 | line = line[len(s) + 1:] | |
610 | break |
|
610 | break | |
611 | patterns.append(linesyntax + line) |
|
611 | patterns.append(linesyntax + line) | |
612 | fp.close() |
|
612 | fp.close() | |
613 | return patterns |
|
613 | return patterns |
General Comments 0
You need to be logged in to leave comments.
Login now