Show More
@@ -1,393 +1,416 b'' | |||
|
1 | 1 | # fileset.py - file set queries for mercurial |
|
2 | 2 | # |
|
3 | 3 | # Copyright 2010 Matt Mackall <mpm@selenic.com> |
|
4 | 4 | # |
|
5 | 5 | # This software may be used and distributed according to the terms of the |
|
6 | 6 | # GNU General Public License version 2 or any later version. |
|
7 | 7 | |
|
8 | 8 | import parser, error, util, merge, re |
|
9 | 9 | from i18n import _ |
|
10 | 10 | |
|
11 | 11 | elements = { |
|
12 | 12 | "(": (20, ("group", 1, ")"), ("func", 1, ")")), |
|
13 | 13 | "-": (5, ("negate", 19), ("minus", 5)), |
|
14 | 14 | "not": (10, ("not", 10)), |
|
15 | 15 | "!": (10, ("not", 10)), |
|
16 | 16 | "and": (5, None, ("and", 5)), |
|
17 | 17 | "&": (5, None, ("and", 5)), |
|
18 | 18 | "or": (4, None, ("or", 4)), |
|
19 | 19 | "|": (4, None, ("or", 4)), |
|
20 | 20 | "+": (4, None, ("or", 4)), |
|
21 | 21 | ",": (2, None, ("list", 2)), |
|
22 | 22 | ")": (0, None, None), |
|
23 | 23 | "symbol": (0, ("symbol",), None), |
|
24 | 24 | "string": (0, ("string",), None), |
|
25 | 25 | "end": (0, None, None), |
|
26 | 26 | } |
|
27 | 27 | |
|
28 | 28 | keywords = set(['and', 'or', 'not']) |
|
29 | 29 | |
|
30 | 30 | globchars = ".*{}[]?/\\" |
|
31 | 31 | |
|
32 | 32 | def tokenize(program): |
|
33 | 33 | pos, l = 0, len(program) |
|
34 | 34 | while pos < l: |
|
35 | 35 | c = program[pos] |
|
36 | 36 | if c.isspace(): # skip inter-token whitespace |
|
37 | 37 | pass |
|
38 | 38 | elif c in "(),-|&+!": # handle simple operators |
|
39 | 39 | yield (c, None, pos) |
|
40 | 40 | elif (c in '"\'' or c == 'r' and |
|
41 | 41 | program[pos:pos + 2] in ("r'", 'r"')): # handle quoted strings |
|
42 | 42 | if c == 'r': |
|
43 | 43 | pos += 1 |
|
44 | 44 | c = program[pos] |
|
45 | 45 | decode = lambda x: x |
|
46 | 46 | else: |
|
47 | 47 | decode = lambda x: x.decode('string-escape') |
|
48 | 48 | pos += 1 |
|
49 | 49 | s = pos |
|
50 | 50 | while pos < l: # find closing quote |
|
51 | 51 | d = program[pos] |
|
52 | 52 | if d == '\\': # skip over escaped characters |
|
53 | 53 | pos += 2 |
|
54 | 54 | continue |
|
55 | 55 | if d == c: |
|
56 | 56 | yield ('string', decode(program[s:pos]), s) |
|
57 | 57 | break |
|
58 | 58 | pos += 1 |
|
59 | 59 | else: |
|
60 | 60 | raise error.ParseError(_("unterminated string"), s) |
|
61 | 61 | elif c.isalnum() or c in globchars or ord(c) > 127: |
|
62 | 62 | # gather up a symbol/keyword |
|
63 | 63 | s = pos |
|
64 | 64 | pos += 1 |
|
65 | 65 | while pos < l: # find end of symbol |
|
66 | 66 | d = program[pos] |
|
67 | 67 | if not (d.isalnum() or d in globchars or ord(d) > 127): |
|
68 | 68 | break |
|
69 | 69 | pos += 1 |
|
70 | 70 | sym = program[s:pos] |
|
71 | 71 | if sym in keywords: # operator keywords |
|
72 | 72 | yield (sym, None, s) |
|
73 | 73 | else: |
|
74 | 74 | yield ('symbol', sym, s) |
|
75 | 75 | pos -= 1 |
|
76 | 76 | else: |
|
77 | 77 | raise error.ParseError(_("syntax error"), pos) |
|
78 | 78 | pos += 1 |
|
79 | 79 | yield ('end', None, pos) |
|
80 | 80 | |
|
81 | 81 | parse = parser.parser(tokenize, elements).parse |
|
82 | 82 | |
|
83 | 83 | def getstring(x, err): |
|
84 | 84 | if x and (x[0] == 'string' or x[0] == 'symbol'): |
|
85 | 85 | return x[1] |
|
86 | 86 | raise error.ParseError(err) |
|
87 | 87 | |
|
88 | 88 | def getset(mctx, x): |
|
89 | 89 | if not x: |
|
90 | 90 | raise error.ParseError(_("missing argument")) |
|
91 | 91 | return methods[x[0]](mctx, *x[1:]) |
|
92 | 92 | |
|
93 | 93 | def stringset(mctx, x): |
|
94 | 94 | m = mctx.matcher([x]) |
|
95 | 95 | return [f for f in mctx.subset if m(f)] |
|
96 | 96 | |
|
97 | 97 | def andset(mctx, x, y): |
|
98 | 98 | return getset(mctx.narrow(getset(mctx, x)), y) |
|
99 | 99 | |
|
100 | 100 | def orset(mctx, x, y): |
|
101 | 101 | # needs optimizing |
|
102 | 102 | xl = getset(mctx, x) |
|
103 | 103 | yl = getset(mctx, y) |
|
104 | 104 | return xl + [f for f in yl if f not in xl] |
|
105 | 105 | |
|
106 | 106 | def notset(mctx, x): |
|
107 | 107 | s = set(getset(mctx, x)) |
|
108 | 108 | return [r for r in mctx.subset if r not in s] |
|
109 | 109 | |
|
110 | 110 | def listset(mctx, a, b): |
|
111 | 111 | raise error.ParseError(_("can't use a list in this context")) |
|
112 | 112 | |
|
113 | 113 | def modified(mctx, x): |
|
114 | 114 | """``modified()`` |
|
115 | 115 | File that is modified according to status. |
|
116 | 116 | """ |
|
117 | 117 | getargs(x, 0, 0, _("modified takes no arguments")) |
|
118 | 118 | s = mctx.status()[0] |
|
119 | 119 | return [f for f in mctx.subset if f in s] |
|
120 | 120 | |
|
121 | 121 | def added(mctx, x): |
|
122 | 122 | """``added()`` |
|
123 | 123 | File that is added according to status. |
|
124 | 124 | """ |
|
125 | 125 | getargs(x, 0, 0, _("added takes no arguments")) |
|
126 | 126 | s = mctx.status()[1] |
|
127 | 127 | return [f for f in mctx.subset if f in s] |
|
128 | 128 | |
|
129 | 129 | def removed(mctx, x): |
|
130 | 130 | """``removed()`` |
|
131 | 131 | File that is removed according to status. |
|
132 | 132 | """ |
|
133 | 133 | getargs(x, 0, 0, _("removed takes no arguments")) |
|
134 | 134 | s = mctx.status()[2] |
|
135 | 135 | return [f for f in mctx.subset if f in s] |
|
136 | 136 | |
|
137 | 137 | def deleted(mctx, x): |
|
138 | 138 | """``deleted()`` |
|
139 | 139 | File that is deleted according to status. |
|
140 | 140 | """ |
|
141 | 141 | getargs(x, 0, 0, _("deleted takes no arguments")) |
|
142 | 142 | s = mctx.status()[3] |
|
143 | 143 | return [f for f in mctx.subset if f in s] |
|
144 | 144 | |
|
145 | 145 | def unknown(mctx, x): |
|
146 | 146 | """``unknown()`` |
|
147 | 147 | File that is unknown according to status. These files will only be |
|
148 | 148 | considered if this predicate is used. |
|
149 | 149 | """ |
|
150 | 150 | getargs(x, 0, 0, _("unknown takes no arguments")) |
|
151 | 151 | s = mctx.status()[4] |
|
152 | 152 | return [f for f in mctx.subset if f in s] |
|
153 | 153 | |
|
154 | 154 | def ignored(mctx, x): |
|
155 | 155 | """``ignored()`` |
|
156 | 156 | File that is ignored according to status. These files will only be |
|
157 | 157 | considered if this predicate is used. |
|
158 | 158 | """ |
|
159 | 159 | getargs(x, 0, 0, _("ignored takes no arguments")) |
|
160 | 160 | s = mctx.status()[5] |
|
161 | 161 | return [f for f in mctx.subset if f in s] |
|
162 | 162 | |
|
163 | 163 | def clean(mctx, x): |
|
164 | 164 | """``clean()`` |
|
165 | 165 | File that is clean according to status. |
|
166 | 166 | """ |
|
167 | 167 | getargs(x, 0, 0, _("clean takes no arguments")) |
|
168 | 168 | s = mctx.status()[6] |
|
169 | 169 | return [f for f in mctx.subset if f in s] |
|
170 | 170 | |
|
171 | 171 | def func(mctx, a, b): |
|
172 | 172 | if a[0] == 'symbol' and a[1] in symbols: |
|
173 | 173 | return symbols[a[1]](mctx, b) |
|
174 | 174 | raise error.ParseError(_("not a function: %s") % a[1]) |
|
175 | 175 | |
|
176 | 176 | def getlist(x): |
|
177 | 177 | if not x: |
|
178 | 178 | return [] |
|
179 | 179 | if x[0] == 'list': |
|
180 | 180 | return getlist(x[1]) + [x[2]] |
|
181 | 181 | return [x] |
|
182 | 182 | |
|
183 | 183 | def getargs(x, min, max, err): |
|
184 | 184 | l = getlist(x) |
|
185 | 185 | if len(l) < min or len(l) > max: |
|
186 | 186 | raise error.ParseError(err) |
|
187 | 187 | return l |
|
188 | 188 | |
|
189 | 189 | def binary(mctx, x): |
|
190 | 190 | """``binary()`` |
|
191 | 191 | File that appears to be binary (contails NUL bytes). |
|
192 | 192 | """ |
|
193 | 193 | getargs(x, 0, 0, _("binary takes no arguments")) |
|
194 | 194 | return [f for f in mctx.subset if util.binary(mctx.ctx[f].data())] |
|
195 | 195 | |
|
196 | 196 | def exec_(mctx, x): |
|
197 | 197 | """``exec()`` |
|
198 | 198 | File that is marked as executable. |
|
199 | 199 | """ |
|
200 | 200 | getargs(x, 0, 0, _("exec takes no arguments")) |
|
201 | 201 | return [f for f in mctx.subset if mctx.ctx.flags(f) == 'x'] |
|
202 | 202 | |
|
203 | 203 | def symlink(mctx, x): |
|
204 | 204 | """``symlink()`` |
|
205 | 205 | File that is marked as a symlink. |
|
206 | 206 | """ |
|
207 | 207 | getargs(x, 0, 0, _("symlink takes no arguments")) |
|
208 | 208 | return [f for f in mctx.subset if mctx.ctx.flags(f) == 'l'] |
|
209 | 209 | |
|
210 | 210 | def resolved(mctx, x): |
|
211 | 211 | """``resolved()`` |
|
212 | 212 | File that is marked resolved according to the resolve state. |
|
213 | 213 | """ |
|
214 | 214 | getargs(x, 0, 0, _("resolved takes no arguments")) |
|
215 | 215 | if mctx.ctx.rev() is not None: |
|
216 | 216 | return [] |
|
217 | 217 | ms = merge.mergestate(mctx.ctx._repo) |
|
218 | 218 | return [f for f in mctx.subset if f in ms and ms[f] == 'r'] |
|
219 | 219 | |
|
220 | 220 | def unresolved(mctx, x): |
|
221 | 221 | """``unresolved()`` |
|
222 | 222 | File that is marked unresolved according to the resolve state. |
|
223 | 223 | """ |
|
224 | 224 | getargs(x, 0, 0, _("unresolved takes no arguments")) |
|
225 | 225 | if mctx.ctx.rev() is not None: |
|
226 | 226 | return [] |
|
227 | 227 | ms = merge.mergestate(mctx.ctx._repo) |
|
228 | 228 | return [f for f in mctx.subset if f in ms and ms[f] == 'u'] |
|
229 | 229 | |
|
230 | 230 | def hgignore(mctx, x): |
|
231 | 231 | """``resolved()`` |
|
232 | 232 | File that matches the active .hgignore pattern. |
|
233 | 233 | """ |
|
234 | 234 | getargs(x, 0, 0, _("hgignore takes no arguments")) |
|
235 | 235 | ignore = mctx.ctx._repo.dirstate._ignore |
|
236 | 236 | return [f for f in mctx.subset if ignore(f)] |
|
237 | 237 | |
|
238 | 238 | def grep(mctx, x): |
|
239 | 239 | """``grep(regex)`` |
|
240 | 240 | File contains the given regular expression. |
|
241 | 241 | """ |
|
242 | 242 | pat = getstring(x, _("grep requires a pattern")) |
|
243 | 243 | r = re.compile(pat) |
|
244 | 244 | return [f for f in mctx.subset if r.search(mctx.ctx[f].data())] |
|
245 | 245 | |
|
246 | 246 | _units = dict(k=2**10, K=2**10, kB=2**10, KB=2**10, |
|
247 | 247 | M=2**20, MB=2**20, G=2**30, GB=2**30, |
|
248 | 248 | kiB=10**3, MiB=10**6, GiB=10**9) |
|
249 | 249 | |
|
250 | 250 | def _sizetoint(s): |
|
251 | 251 | try: |
|
252 | 252 | s = s.strip() |
|
253 | 253 | for k, v in _units.items(): |
|
254 | 254 | if s.endswith(k): |
|
255 | 255 | return int(float(s[:-len(k)]) * v) |
|
256 | 256 | return int(s) |
|
257 | 257 | except ValueError: |
|
258 | 258 | raise |
|
259 | 259 | raise error.ParseError(_("couldn't parse size"), s) |
|
260 | 260 | |
|
261 | 261 | def _sizetomax(s): |
|
262 | 262 | try: |
|
263 | 263 | s = s.strip() |
|
264 | 264 | for k, v in _units.items(): |
|
265 | 265 | if s.endswith(k): |
|
266 | 266 | # max(4k) = 5k - 1, max(4.5k) = 4.6k - 1 |
|
267 | 267 | n = s[:-len(k)] |
|
268 | 268 | inc = 1.0 |
|
269 | 269 | if "." in n: |
|
270 | 270 | inc /= 10 ** len(n.split(".")[1]) |
|
271 | 271 | return int((float(n) + inc) * v) - 1 |
|
272 | 272 | # no extension, this is a precise value |
|
273 | 273 | return int(s) |
|
274 | 274 | except ValueError: |
|
275 | 275 | raise |
|
276 | 276 | raise error.ParseError(_("couldn't parse size"), s) |
|
277 | 277 | |
|
278 | 278 | def size(mctx, x): |
|
279 | 279 | """``size(expression)`` |
|
280 | 280 | File size matches the given expression. Examples: |
|
281 | 281 | |
|
282 | 282 | - 1k (files from 1024 to 2047 bytes) |
|
283 | 283 | - 1.0kiB (files from 1000 to 1100 bytes) |
|
284 | 284 | - < 20k (files less than 20480 bytes) |
|
285 | 285 | - >= .5MiB (files at least 500000 bytes) |
|
286 | 286 | - 4k - 1MB (files from 4096 bytes to 1048576 bytes) |
|
287 | 287 | """ |
|
288 | 288 | |
|
289 | 289 | expr = getstring(x, _("grep requires a pattern")).strip() |
|
290 | 290 | if '-' in expr: # do we have a range? |
|
291 | 291 | a, b = expr.split('-', 1) |
|
292 | 292 | a = _sizetoint(a) |
|
293 | 293 | b = _sizetoint(b) |
|
294 | 294 | m = lambda x: x >= a and x <= b |
|
295 | 295 | elif expr.startswith("<="): |
|
296 | 296 | a = _sizetoint(expr[2:]) |
|
297 | 297 | m = lambda x: x <= a |
|
298 | 298 | elif expr.startswith("<"): |
|
299 | 299 | a = _sizetoint(expr[1:]) |
|
300 | 300 | m = lambda x: x < a |
|
301 | 301 | elif expr.startswith(">="): |
|
302 | 302 | a = _sizetoint(expr[2:]) |
|
303 | 303 | m = lambda x: x >= a |
|
304 | 304 | elif expr.startswith(">"): |
|
305 | 305 | a = _sizetoint(expr[1:]) |
|
306 | 306 | m = lambda x: x > a |
|
307 | 307 | elif expr[0].isdigit or expr[0] == '.': |
|
308 | 308 | a = _sizetoint(expr) |
|
309 | 309 | b = _sizetomax(expr) |
|
310 | 310 | m = lambda x: x >=a and x <= b |
|
311 | 311 | else: |
|
312 | 312 | raise error.ParseError(_("couldn't parse size"), expr) |
|
313 | 313 | |
|
314 | 314 | return [f for f in mctx.subset if m(mctx.ctx[f].size())] |
|
315 | 315 | |
|
316 | def encoding(mctx, x): | |
|
317 | """``encoding(name)`` | |
|
318 | File can be successfully decoded with the given character | |
|
319 | encoding. May not be useful for encodings other than ASCII and | |
|
320 | UTF-8. | |
|
321 | """ | |
|
322 | ||
|
323 | enc = getstring(x, _("encoding requires an encoding name")) | |
|
324 | ||
|
325 | s = [] | |
|
326 | for f in mctx.subset: | |
|
327 | d = mctx.ctx[f].data() | |
|
328 | try: | |
|
329 | d.decode(enc) | |
|
330 | except LookupError: | |
|
331 | raise util.Abort(_("unknown encoding '%s'") % enc) | |
|
332 | except UnicodeDecodeError: | |
|
333 | continue | |
|
334 | s.append(f) | |
|
335 | ||
|
336 | return s | |
|
337 | ||
|
316 | 338 | symbols = { |
|
317 | 339 | 'added': added, |
|
318 | 340 | 'binary': binary, |
|
319 | 341 | 'clean': clean, |
|
320 | 342 | 'deleted': deleted, |
|
343 | 'encoding': encoding, | |
|
321 | 344 | 'exec': exec_, |
|
322 | 345 | 'grep': grep, |
|
323 | 346 | 'ignored': ignored, |
|
324 | 347 | 'hgignore': hgignore, |
|
325 | 348 | 'modified': modified, |
|
326 | 349 | 'removed': removed, |
|
327 | 350 | 'resolved': resolved, |
|
328 | 351 | 'size': size, |
|
329 | 352 | 'symlink': symlink, |
|
330 | 353 | 'unknown': unknown, |
|
331 | 354 | 'unresolved': unresolved, |
|
332 | 355 | } |
|
333 | 356 | |
|
334 | 357 | methods = { |
|
335 | 358 | 'string': stringset, |
|
336 | 359 | 'symbol': stringset, |
|
337 | 360 | 'and': andset, |
|
338 | 361 | 'or': orset, |
|
339 | 362 | 'list': listset, |
|
340 | 363 | 'group': getset, |
|
341 | 364 | 'not': notset, |
|
342 | 365 | 'func': func, |
|
343 | 366 | } |
|
344 | 367 | |
|
345 | 368 | class matchctx(object): |
|
346 | 369 | def __init__(self, ctx, subset=None, status=None): |
|
347 | 370 | self.ctx = ctx |
|
348 | 371 | self.subset = subset |
|
349 | 372 | self._status = status |
|
350 | 373 | def status(self): |
|
351 | 374 | return self._status |
|
352 | 375 | def matcher(self, patterns): |
|
353 | 376 | return self.ctx.match(patterns) |
|
354 | 377 | def filter(self, files): |
|
355 | 378 | return [f for f in files if f in self.subset] |
|
356 | 379 | def narrow(self, files): |
|
357 | 380 | return matchctx(self.ctx, self.filter(files), self._status) |
|
358 | 381 | |
|
359 | 382 | def _intree(funcs, tree): |
|
360 | 383 | if isinstance(tree, tuple): |
|
361 | 384 | if tree[0] == 'func' and tree[1][0] == 'symbol': |
|
362 | 385 | if tree[1][1] in funcs: |
|
363 | 386 | return True |
|
364 | 387 | for s in tree[1:]: |
|
365 | 388 | if _intree(funcs, s): |
|
366 | 389 | return True |
|
367 | 390 | return False |
|
368 | 391 | |
|
369 | 392 | def getfileset(ctx, expr): |
|
370 | 393 | tree, pos = parse(expr) |
|
371 | 394 | if (pos != len(expr)): |
|
372 | 395 | raise error.ParseError("invalid token", pos) |
|
373 | 396 | |
|
374 | 397 | # do we need status info? |
|
375 | 398 | if _intree(['modified', 'added', 'removed', 'deleted', |
|
376 | 399 | 'unknown', 'ignored', 'clean'], tree): |
|
377 | 400 | unknown = _intree(['unknown'], tree) |
|
378 | 401 | ignored = _intree(['ignored'], tree) |
|
379 | 402 | |
|
380 | 403 | r = ctx._repo |
|
381 | 404 | status = r.status(ctx.p1(), ctx, |
|
382 | 405 | unknown=unknown, ignored=ignored, clean=True) |
|
383 | 406 | subset = [] |
|
384 | 407 | for c in status: |
|
385 | 408 | subset.extend(c) |
|
386 | 409 | else: |
|
387 | 410 | status = None |
|
388 | 411 | subset = ctx.walk(ctx.match([])) |
|
389 | 412 | |
|
390 | 413 | return getset(matchctx(ctx, subset, status), tree) |
|
391 | 414 | |
|
392 | 415 | # tell hggettext to extract docstrings from these functions: |
|
393 | 416 | i18nfunctions = symbols.values() |
General Comments 0
You need to be logged in to leave comments.
Login now