Show More
@@ -1,837 +1,841 b'' | |||
|
1 | 1 | # -*- coding: utf-8 -*- |
|
2 | 2 | |
|
3 | 3 | # Copyright (C) 2014-2019 RhodeCode GmbH |
|
4 | 4 | # |
|
5 | 5 | # This program is free software: you can redistribute it and/or modify |
|
6 | 6 | # it under the terms of the GNU Affero General Public License, version 3 |
|
7 | 7 | # (only), as published by the Free Software Foundation. |
|
8 | 8 | # |
|
9 | 9 | # This program is distributed in the hope that it will be useful, |
|
10 | 10 | # but WITHOUT ANY WARRANTY; without even the implied warranty of |
|
11 | 11 | # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the |
|
12 | 12 | # GNU General Public License for more details. |
|
13 | 13 | # |
|
14 | 14 | # You should have received a copy of the GNU Affero General Public License |
|
15 | 15 | # along with this program. If not, see <http://www.gnu.org/licenses/>. |
|
16 | 16 | # |
|
17 | 17 | # This program is dual-licensed. If you wish to learn more about the |
|
18 | 18 | # RhodeCode Enterprise Edition, including its added features, Support services, |
|
19 | 19 | # and proprietary license terms, please see https://rhodecode.com/licenses/ |
|
20 | 20 | |
|
21 | 21 | """ |
|
22 | 22 | Module holding everything related to vcs nodes, with vcs2 architecture. |
|
23 | 23 | """ |
|
24 | 24 | |
|
25 | 25 | import os |
|
26 | 26 | import stat |
|
27 | 27 | |
|
28 | 28 | from zope.cachedescriptors.property import Lazy as LazyProperty |
|
29 | 29 | |
|
30 | 30 | from rhodecode.config.conf import LANGUAGES_EXTENSIONS_MAP |
|
31 | 31 | from rhodecode.lib.utils import safe_unicode, safe_str |
|
32 | 32 | from rhodecode.lib.utils2 import md5 |
|
33 | 33 | from rhodecode.lib.vcs import path as vcspath |
|
34 | 34 | from rhodecode.lib.vcs.backends.base import EmptyCommit, FILEMODE_DEFAULT |
|
35 | 35 | from rhodecode.lib.vcs.conf.mtypes import get_mimetypes_db |
|
36 | 36 | from rhodecode.lib.vcs.exceptions import NodeError, RemovedFileNodeError |
|
37 | 37 | |
|
38 | 38 | LARGEFILE_PREFIX = '.hglf' |
|
39 | 39 | |
|
40 | 40 | |
|
41 | 41 | class NodeKind: |
|
42 | 42 | SUBMODULE = -1 |
|
43 | 43 | DIR = 1 |
|
44 | 44 | FILE = 2 |
|
45 | 45 | LARGEFILE = 3 |
|
46 | 46 | |
|
47 | 47 | |
|
48 | 48 | class NodeState: |
|
49 | 49 | ADDED = u'added' |
|
50 | 50 | CHANGED = u'changed' |
|
51 | 51 | NOT_CHANGED = u'not changed' |
|
52 | 52 | REMOVED = u'removed' |
|
53 | 53 | |
|
54 | 54 | |
|
55 | 55 | class NodeGeneratorBase(object): |
|
56 | 56 | """ |
|
57 | 57 | Base class for removed added and changed filenodes, it's a lazy generator |
|
58 | 58 | class that will create filenodes only on iteration or call |
|
59 | 59 | |
|
60 | 60 | The len method doesn't need to create filenodes at all |
|
61 | 61 | """ |
|
62 | 62 | |
|
63 | 63 | def __init__(self, current_paths, cs): |
|
64 | 64 | self.cs = cs |
|
65 | 65 | self.current_paths = current_paths |
|
66 | 66 | |
|
67 | 67 | def __call__(self): |
|
68 | 68 | return [n for n in self] |
|
69 | 69 | |
|
70 | 70 | def __getslice__(self, i, j): |
|
71 | 71 | for p in self.current_paths[i:j]: |
|
72 | 72 | yield self.cs.get_node(p) |
|
73 | 73 | |
|
74 | 74 | def __len__(self): |
|
75 | 75 | return len(self.current_paths) |
|
76 | 76 | |
|
77 | 77 | def __iter__(self): |
|
78 | 78 | for p in self.current_paths: |
|
79 | 79 | yield self.cs.get_node(p) |
|
80 | 80 | |
|
81 | 81 | |
|
82 | 82 | class AddedFileNodesGenerator(NodeGeneratorBase): |
|
83 | 83 | """ |
|
84 | 84 | Class holding added files for current commit |
|
85 | 85 | """ |
|
86 | 86 | |
|
87 | 87 | |
|
88 | 88 | class ChangedFileNodesGenerator(NodeGeneratorBase): |
|
89 | 89 | """ |
|
90 | 90 | Class holding changed files for current commit |
|
91 | 91 | """ |
|
92 | 92 | |
|
93 | 93 | |
|
94 | 94 | class RemovedFileNodesGenerator(NodeGeneratorBase): |
|
95 | 95 | """ |
|
96 | 96 | Class holding removed files for current commit |
|
97 | 97 | """ |
|
98 | 98 | def __iter__(self): |
|
99 | 99 | for p in self.current_paths: |
|
100 | 100 | yield RemovedFileNode(path=p) |
|
101 | 101 | |
|
102 | 102 | def __getslice__(self, i, j): |
|
103 | 103 | for p in self.current_paths[i:j]: |
|
104 | 104 | yield RemovedFileNode(path=p) |
|
105 | 105 | |
|
106 | 106 | |
|
107 | 107 | class Node(object): |
|
108 | 108 | """ |
|
109 | 109 | Simplest class representing file or directory on repository. SCM backends |
|
110 | 110 | should use ``FileNode`` and ``DirNode`` subclasses rather than ``Node`` |
|
111 | 111 | directly. |
|
112 | 112 | |
|
113 | 113 | Node's ``path`` cannot start with slash as we operate on *relative* paths |
|
114 | 114 | only. Moreover, every single node is identified by the ``path`` attribute, |
|
115 | 115 | so it cannot end with slash, too. Otherwise, path could lead to mistakes. |
|
116 | 116 | """ |
|
117 | 117 | RTLO_MARKER = u"\u202E" # RTLO marker allows swapping text, and certain |
|
118 | 118 | # security attacks could be used with this |
|
119 | 119 | commit = None |
|
120 | 120 | |
|
121 | 121 | def __init__(self, path, kind): |
|
122 | 122 | self._validate_path(path) # can throw exception if path is invalid |
|
123 | 123 | self.path = safe_str(path.rstrip('/')) # we store paths as str |
|
124 | 124 | if path == '' and kind != NodeKind.DIR: |
|
125 | 125 | raise NodeError("Only DirNode and its subclasses may be " |
|
126 | 126 | "initialized with empty path") |
|
127 | 127 | self.kind = kind |
|
128 | 128 | |
|
129 | 129 | if self.is_root() and not self.is_dir(): |
|
130 | 130 | raise NodeError("Root node cannot be FILE kind") |
|
131 | 131 | |
|
132 | 132 | def _validate_path(self, path): |
|
133 | 133 | if path.startswith('/'): |
|
134 | 134 | raise NodeError( |
|
135 | 135 | "Cannot initialize Node objects with slash at " |
|
136 | 136 | "the beginning as only relative paths are supported. " |
|
137 | 137 | "Got %s" % (path,)) |
|
138 | 138 | |
|
139 | 139 | @LazyProperty |
|
140 | 140 | def parent(self): |
|
141 | 141 | parent_path = self.get_parent_path() |
|
142 | 142 | if parent_path: |
|
143 | 143 | if self.commit: |
|
144 | 144 | return self.commit.get_node(parent_path) |
|
145 | 145 | return DirNode(parent_path) |
|
146 | 146 | return None |
|
147 | 147 | |
|
148 | 148 | @LazyProperty |
|
149 | 149 | def unicode_path(self): |
|
150 | 150 | return safe_unicode(self.path) |
|
151 | 151 | |
|
152 | 152 | @LazyProperty |
|
153 | 153 | def has_rtlo(self): |
|
154 | 154 | """Detects if a path has right-to-left-override marker""" |
|
155 | 155 | return self.RTLO_MARKER in self.unicode_path |
|
156 | 156 | |
|
157 | 157 | @LazyProperty |
|
158 | 158 | def unicode_path_safe(self): |
|
159 | 159 | """ |
|
160 | 160 | Special SAFE representation of path without the right-to-left-override. |
|
161 | 161 | This should be only used for "showing" the file, cannot be used for any |
|
162 | 162 | urls etc. |
|
163 | 163 | """ |
|
164 | 164 | return safe_unicode(self.path).replace(self.RTLO_MARKER, '') |
|
165 | 165 | |
|
166 | 166 | @LazyProperty |
|
167 | 167 | def dir_path(self): |
|
168 | 168 | """ |
|
169 | 169 | Returns name of the directory from full path of this vcs node. Empty |
|
170 | 170 | string is returned if there's no directory in the path |
|
171 | 171 | """ |
|
172 | 172 | _parts = self.path.rstrip('/').rsplit('/', 1) |
|
173 | 173 | if len(_parts) == 2: |
|
174 | 174 | return safe_unicode(_parts[0]) |
|
175 | 175 | return u'' |
|
176 | 176 | |
|
177 | 177 | @LazyProperty |
|
178 | 178 | def name(self): |
|
179 | 179 | """ |
|
180 | 180 | Returns name of the node so if its path |
|
181 | 181 | then only last part is returned. |
|
182 | 182 | """ |
|
183 | 183 | return safe_unicode(self.path.rstrip('/').split('/')[-1]) |
|
184 | 184 | |
|
185 | 185 | @property |
|
186 | 186 | def kind(self): |
|
187 | 187 | return self._kind |
|
188 | 188 | |
|
189 | 189 | @kind.setter |
|
190 | 190 | def kind(self, kind): |
|
191 | 191 | if hasattr(self, '_kind'): |
|
192 | 192 | raise NodeError("Cannot change node's kind") |
|
193 | 193 | else: |
|
194 | 194 | self._kind = kind |
|
195 | 195 | # Post setter check (path's trailing slash) |
|
196 | 196 | if self.path.endswith('/'): |
|
197 | 197 | raise NodeError("Node's path cannot end with slash") |
|
198 | 198 | |
|
199 | 199 | def __cmp__(self, other): |
|
200 | 200 | """ |
|
201 | 201 | Comparator using name of the node, needed for quick list sorting. |
|
202 | 202 | """ |
|
203 | ||
|
203 | 204 | kind_cmp = cmp(self.kind, other.kind) |
|
204 | 205 | if kind_cmp: |
|
206 | if isinstance(self, SubModuleNode): | |
|
207 | # we make submodules equal to dirnode for "sorting" purposes | |
|
208 | return NodeKind.DIR | |
|
205 | 209 | return kind_cmp |
|
206 | 210 | return cmp(self.name, other.name) |
|
207 | 211 | |
|
208 | 212 | def __eq__(self, other): |
|
209 | 213 | for attr in ['name', 'path', 'kind']: |
|
210 | 214 | if getattr(self, attr) != getattr(other, attr): |
|
211 | 215 | return False |
|
212 | 216 | if self.is_file(): |
|
213 | 217 | if self.content != other.content: |
|
214 | 218 | return False |
|
215 | 219 | else: |
|
216 | 220 | # For DirNode's check without entering each dir |
|
217 | 221 | self_nodes_paths = list(sorted(n.path for n in self.nodes)) |
|
218 | 222 | other_nodes_paths = list(sorted(n.path for n in self.nodes)) |
|
219 | 223 | if self_nodes_paths != other_nodes_paths: |
|
220 | 224 | return False |
|
221 | 225 | return True |
|
222 | 226 | |
|
223 | 227 | def __ne__(self, other): |
|
224 | 228 | return not self.__eq__(other) |
|
225 | 229 | |
|
226 | 230 | def __repr__(self): |
|
227 | 231 | return '<%s %r>' % (self.__class__.__name__, self.path) |
|
228 | 232 | |
|
229 | 233 | def __str__(self): |
|
230 | 234 | return self.__repr__() |
|
231 | 235 | |
|
232 | 236 | def __unicode__(self): |
|
233 | 237 | return self.name |
|
234 | 238 | |
|
235 | 239 | def get_parent_path(self): |
|
236 | 240 | """ |
|
237 | 241 | Returns node's parent path or empty string if node is root. |
|
238 | 242 | """ |
|
239 | 243 | if self.is_root(): |
|
240 | 244 | return '' |
|
241 | 245 | return vcspath.dirname(self.path.rstrip('/')) + '/' |
|
242 | 246 | |
|
243 | 247 | def is_file(self): |
|
244 | 248 | """ |
|
245 | 249 | Returns ``True`` if node's kind is ``NodeKind.FILE``, ``False`` |
|
246 | 250 | otherwise. |
|
247 | 251 | """ |
|
248 | 252 | return self.kind == NodeKind.FILE |
|
249 | 253 | |
|
250 | 254 | def is_dir(self): |
|
251 | 255 | """ |
|
252 | 256 | Returns ``True`` if node's kind is ``NodeKind.DIR``, ``False`` |
|
253 | 257 | otherwise. |
|
254 | 258 | """ |
|
255 | 259 | return self.kind == NodeKind.DIR |
|
256 | 260 | |
|
257 | 261 | def is_root(self): |
|
258 | 262 | """ |
|
259 | 263 | Returns ``True`` if node is a root node and ``False`` otherwise. |
|
260 | 264 | """ |
|
261 | 265 | return self.kind == NodeKind.DIR and self.path == '' |
|
262 | 266 | |
|
263 | 267 | def is_submodule(self): |
|
264 | 268 | """ |
|
265 | 269 | Returns ``True`` if node's kind is ``NodeKind.SUBMODULE``, ``False`` |
|
266 | 270 | otherwise. |
|
267 | 271 | """ |
|
268 | 272 | return self.kind == NodeKind.SUBMODULE |
|
269 | 273 | |
|
270 | 274 | def is_largefile(self): |
|
271 | 275 | """ |
|
272 | 276 | Returns ``True`` if node's kind is ``NodeKind.LARGEFILE``, ``False`` |
|
273 | 277 | otherwise |
|
274 | 278 | """ |
|
275 | 279 | return self.kind == NodeKind.LARGEFILE |
|
276 | 280 | |
|
277 | 281 | def is_link(self): |
|
278 | 282 | if self.commit: |
|
279 | 283 | return self.commit.is_link(self.path) |
|
280 | 284 | return False |
|
281 | 285 | |
|
282 | 286 | @LazyProperty |
|
283 | 287 | def added(self): |
|
284 | 288 | return self.state is NodeState.ADDED |
|
285 | 289 | |
|
286 | 290 | @LazyProperty |
|
287 | 291 | def changed(self): |
|
288 | 292 | return self.state is NodeState.CHANGED |
|
289 | 293 | |
|
290 | 294 | @LazyProperty |
|
291 | 295 | def not_changed(self): |
|
292 | 296 | return self.state is NodeState.NOT_CHANGED |
|
293 | 297 | |
|
294 | 298 | @LazyProperty |
|
295 | 299 | def removed(self): |
|
296 | 300 | return self.state is NodeState.REMOVED |
|
297 | 301 | |
|
298 | 302 | |
|
299 | 303 | class FileNode(Node): |
|
300 | 304 | """ |
|
301 | 305 | Class representing file nodes. |
|
302 | 306 | |
|
303 | 307 | :attribute: path: path to the node, relative to repository's root |
|
304 | 308 | :attribute: content: if given arbitrary sets content of the file |
|
305 | 309 | :attribute: commit: if given, first time content is accessed, callback |
|
306 | 310 | :attribute: mode: stat mode for a node. Default is `FILEMODE_DEFAULT`. |
|
307 | 311 | """ |
|
308 | 312 | _filter_pre_load = [] |
|
309 | 313 | |
|
310 | 314 | def __init__(self, path, content=None, commit=None, mode=None, pre_load=None): |
|
311 | 315 | """ |
|
312 | 316 | Only one of ``content`` and ``commit`` may be given. Passing both |
|
313 | 317 | would raise ``NodeError`` exception. |
|
314 | 318 | |
|
315 | 319 | :param path: relative path to the node |
|
316 | 320 | :param content: content may be passed to constructor |
|
317 | 321 | :param commit: if given, will use it to lazily fetch content |
|
318 | 322 | :param mode: ST_MODE (i.e. 0100644) |
|
319 | 323 | """ |
|
320 | 324 | if content and commit: |
|
321 | 325 | raise NodeError("Cannot use both content and commit") |
|
322 | 326 | super(FileNode, self).__init__(path, kind=NodeKind.FILE) |
|
323 | 327 | self.commit = commit |
|
324 | 328 | self._content = content |
|
325 | 329 | self._mode = mode or FILEMODE_DEFAULT |
|
326 | 330 | |
|
327 | 331 | self._set_bulk_properties(pre_load) |
|
328 | 332 | |
|
329 | 333 | def _set_bulk_properties(self, pre_load): |
|
330 | 334 | if not pre_load: |
|
331 | 335 | return |
|
332 | 336 | pre_load = [entry for entry in pre_load |
|
333 | 337 | if entry not in self._filter_pre_load] |
|
334 | 338 | if not pre_load: |
|
335 | 339 | return |
|
336 | 340 | |
|
337 | 341 | for attr_name in pre_load: |
|
338 | 342 | result = getattr(self, attr_name) |
|
339 | 343 | if callable(result): |
|
340 | 344 | result = result() |
|
341 | 345 | self.__dict__[attr_name] = result |
|
342 | 346 | |
|
343 | 347 | @LazyProperty |
|
344 | 348 | def mode(self): |
|
345 | 349 | """ |
|
346 | 350 | Returns lazily mode of the FileNode. If `commit` is not set, would |
|
347 | 351 | use value given at initialization or `FILEMODE_DEFAULT` (default). |
|
348 | 352 | """ |
|
349 | 353 | if self.commit: |
|
350 | 354 | mode = self.commit.get_file_mode(self.path) |
|
351 | 355 | else: |
|
352 | 356 | mode = self._mode |
|
353 | 357 | return mode |
|
354 | 358 | |
|
355 | 359 | @LazyProperty |
|
356 | 360 | def raw_bytes(self): |
|
357 | 361 | """ |
|
358 | 362 | Returns lazily the raw bytes of the FileNode. |
|
359 | 363 | """ |
|
360 | 364 | if self.commit: |
|
361 | 365 | if self._content is None: |
|
362 | 366 | self._content = self.commit.get_file_content(self.path) |
|
363 | 367 | content = self._content |
|
364 | 368 | else: |
|
365 | 369 | content = self._content |
|
366 | 370 | return content |
|
367 | 371 | |
|
368 | 372 | @LazyProperty |
|
369 | 373 | def md5(self): |
|
370 | 374 | """ |
|
371 | 375 | Returns md5 of the file node. |
|
372 | 376 | """ |
|
373 | 377 | return md5(self.raw_bytes) |
|
374 | 378 | |
|
375 | 379 | def metadata_uncached(self): |
|
376 | 380 | """ |
|
377 | 381 | Returns md5, binary flag of the file node, without any cache usage. |
|
378 | 382 | """ |
|
379 | 383 | |
|
380 | 384 | if self.commit: |
|
381 | 385 | content = self.commit.get_file_content(self.path) |
|
382 | 386 | else: |
|
383 | 387 | content = self._content |
|
384 | 388 | |
|
385 | 389 | is_binary = content and '\0' in content |
|
386 | 390 | size = 0 |
|
387 | 391 | if content: |
|
388 | 392 | size = len(content) |
|
389 | 393 | return is_binary, md5(content), size |
|
390 | 394 | |
|
391 | 395 | @LazyProperty |
|
392 | 396 | def content(self): |
|
393 | 397 | """ |
|
394 | 398 | Returns lazily content of the FileNode. If possible, would try to |
|
395 | 399 | decode content from UTF-8. |
|
396 | 400 | """ |
|
397 | 401 | content = self.raw_bytes |
|
398 | 402 | |
|
399 | 403 | if self.is_binary: |
|
400 | 404 | return content |
|
401 | 405 | return safe_unicode(content) |
|
402 | 406 | |
|
403 | 407 | @LazyProperty |
|
404 | 408 | def size(self): |
|
405 | 409 | if self.commit: |
|
406 | 410 | return self.commit.get_file_size(self.path) |
|
407 | 411 | raise NodeError( |
|
408 | 412 | "Cannot retrieve size of the file without related " |
|
409 | 413 | "commit attribute") |
|
410 | 414 | |
|
411 | 415 | @LazyProperty |
|
412 | 416 | def message(self): |
|
413 | 417 | if self.commit: |
|
414 | 418 | return self.last_commit.message |
|
415 | 419 | raise NodeError( |
|
416 | 420 | "Cannot retrieve message of the file without related " |
|
417 | 421 | "commit attribute") |
|
418 | 422 | |
|
419 | 423 | @LazyProperty |
|
420 | 424 | def last_commit(self): |
|
421 | 425 | if self.commit: |
|
422 | 426 | pre_load = ["author", "date", "message"] |
|
423 | 427 | return self.commit.get_path_commit(self.path, pre_load=pre_load) |
|
424 | 428 | raise NodeError( |
|
425 | 429 | "Cannot retrieve last commit of the file without " |
|
426 | 430 | "related commit attribute") |
|
427 | 431 | |
|
428 | 432 | def get_mimetype(self): |
|
429 | 433 | """ |
|
430 | 434 | Mimetype is calculated based on the file's content. If ``_mimetype`` |
|
431 | 435 | attribute is available, it will be returned (backends which store |
|
432 | 436 | mimetypes or can easily recognize them, should set this private |
|
433 | 437 | attribute to indicate that type should *NOT* be calculated). |
|
434 | 438 | """ |
|
435 | 439 | |
|
436 | 440 | if hasattr(self, '_mimetype'): |
|
437 | 441 | if (isinstance(self._mimetype, (tuple, list,)) and |
|
438 | 442 | len(self._mimetype) == 2): |
|
439 | 443 | return self._mimetype |
|
440 | 444 | else: |
|
441 | 445 | raise NodeError('given _mimetype attribute must be an 2 ' |
|
442 | 446 | 'element list or tuple') |
|
443 | 447 | |
|
444 | 448 | db = get_mimetypes_db() |
|
445 | 449 | mtype, encoding = db.guess_type(self.name) |
|
446 | 450 | |
|
447 | 451 | if mtype is None: |
|
448 | 452 | if self.is_binary: |
|
449 | 453 | mtype = 'application/octet-stream' |
|
450 | 454 | encoding = None |
|
451 | 455 | else: |
|
452 | 456 | mtype = 'text/plain' |
|
453 | 457 | encoding = None |
|
454 | 458 | |
|
455 | 459 | # try with pygments |
|
456 | 460 | try: |
|
457 | 461 | from pygments.lexers import get_lexer_for_filename |
|
458 | 462 | mt = get_lexer_for_filename(self.name).mimetypes |
|
459 | 463 | except Exception: |
|
460 | 464 | mt = None |
|
461 | 465 | |
|
462 | 466 | if mt: |
|
463 | 467 | mtype = mt[0] |
|
464 | 468 | |
|
465 | 469 | return mtype, encoding |
|
466 | 470 | |
|
467 | 471 | @LazyProperty |
|
468 | 472 | def mimetype(self): |
|
469 | 473 | """ |
|
470 | 474 | Wrapper around full mimetype info. It returns only type of fetched |
|
471 | 475 | mimetype without the encoding part. use get_mimetype function to fetch |
|
472 | 476 | full set of (type,encoding) |
|
473 | 477 | """ |
|
474 | 478 | return self.get_mimetype()[0] |
|
475 | 479 | |
|
476 | 480 | @LazyProperty |
|
477 | 481 | def mimetype_main(self): |
|
478 | 482 | return self.mimetype.split('/')[0] |
|
479 | 483 | |
|
480 | 484 | @classmethod |
|
481 | 485 | def get_lexer(cls, filename, content=None): |
|
482 | 486 | from pygments import lexers |
|
483 | 487 | |
|
484 | 488 | extension = filename.split('.')[-1] |
|
485 | 489 | lexer = None |
|
486 | 490 | |
|
487 | 491 | try: |
|
488 | 492 | lexer = lexers.guess_lexer_for_filename( |
|
489 | 493 | filename, content, stripnl=False) |
|
490 | 494 | except lexers.ClassNotFound: |
|
491 | 495 | lexer = None |
|
492 | 496 | |
|
493 | 497 | # try our EXTENSION_MAP |
|
494 | 498 | if not lexer: |
|
495 | 499 | try: |
|
496 | 500 | lexer_class = LANGUAGES_EXTENSIONS_MAP.get(extension) |
|
497 | 501 | if lexer_class: |
|
498 | 502 | lexer = lexers.get_lexer_by_name(lexer_class[0]) |
|
499 | 503 | except lexers.ClassNotFound: |
|
500 | 504 | lexer = None |
|
501 | 505 | |
|
502 | 506 | if not lexer: |
|
503 | 507 | lexer = lexers.TextLexer(stripnl=False) |
|
504 | 508 | |
|
505 | 509 | return lexer |
|
506 | 510 | |
|
507 | 511 | @LazyProperty |
|
508 | 512 | def lexer(self): |
|
509 | 513 | """ |
|
510 | 514 | Returns pygment's lexer class. Would try to guess lexer taking file's |
|
511 | 515 | content, name and mimetype. |
|
512 | 516 | """ |
|
513 | 517 | return self.get_lexer(self.name, self.content) |
|
514 | 518 | |
|
515 | 519 | @LazyProperty |
|
516 | 520 | def lexer_alias(self): |
|
517 | 521 | """ |
|
518 | 522 | Returns first alias of the lexer guessed for this file. |
|
519 | 523 | """ |
|
520 | 524 | return self.lexer.aliases[0] |
|
521 | 525 | |
|
522 | 526 | @LazyProperty |
|
523 | 527 | def history(self): |
|
524 | 528 | """ |
|
525 | 529 | Returns a list of commit for this file in which the file was changed |
|
526 | 530 | """ |
|
527 | 531 | if self.commit is None: |
|
528 | 532 | raise NodeError('Unable to get commit for this FileNode') |
|
529 | 533 | return self.commit.get_path_history(self.path) |
|
530 | 534 | |
|
531 | 535 | @LazyProperty |
|
532 | 536 | def annotate(self): |
|
533 | 537 | """ |
|
534 | 538 | Returns a list of three element tuples with lineno, commit and line |
|
535 | 539 | """ |
|
536 | 540 | if self.commit is None: |
|
537 | 541 | raise NodeError('Unable to get commit for this FileNode') |
|
538 | 542 | pre_load = ["author", "date", "message"] |
|
539 | 543 | return self.commit.get_file_annotate(self.path, pre_load=pre_load) |
|
540 | 544 | |
|
541 | 545 | @LazyProperty |
|
542 | 546 | def state(self): |
|
543 | 547 | if not self.commit: |
|
544 | 548 | raise NodeError( |
|
545 | 549 | "Cannot check state of the node if it's not " |
|
546 | 550 | "linked with commit") |
|
547 | 551 | elif self.path in (node.path for node in self.commit.added): |
|
548 | 552 | return NodeState.ADDED |
|
549 | 553 | elif self.path in (node.path for node in self.commit.changed): |
|
550 | 554 | return NodeState.CHANGED |
|
551 | 555 | else: |
|
552 | 556 | return NodeState.NOT_CHANGED |
|
553 | 557 | |
|
554 | 558 | @LazyProperty |
|
555 | 559 | def is_binary(self): |
|
556 | 560 | """ |
|
557 | 561 | Returns True if file has binary content. |
|
558 | 562 | """ |
|
559 | 563 | _bin = self.raw_bytes and '\0' in self.raw_bytes |
|
560 | 564 | return _bin |
|
561 | 565 | |
|
562 | 566 | @LazyProperty |
|
563 | 567 | def extension(self): |
|
564 | 568 | """Returns filenode extension""" |
|
565 | 569 | return self.name.split('.')[-1] |
|
566 | 570 | |
|
567 | 571 | @property |
|
568 | 572 | def is_executable(self): |
|
569 | 573 | """ |
|
570 | 574 | Returns ``True`` if file has executable flag turned on. |
|
571 | 575 | """ |
|
572 | 576 | return bool(self.mode & stat.S_IXUSR) |
|
573 | 577 | |
|
574 | 578 | def get_largefile_node(self): |
|
575 | 579 | """ |
|
576 | 580 | Try to return a Mercurial FileNode from this node. It does internal |
|
577 | 581 | checks inside largefile store, if that file exist there it will |
|
578 | 582 | create special instance of LargeFileNode which can get content from |
|
579 | 583 | LF store. |
|
580 | 584 | """ |
|
581 | 585 | if self.commit: |
|
582 | 586 | return self.commit.get_largefile_node(self.path) |
|
583 | 587 | |
|
584 | 588 | def lines(self, count_empty=False): |
|
585 | 589 | all_lines, empty_lines = 0, 0 |
|
586 | 590 | |
|
587 | 591 | if not self.is_binary: |
|
588 | 592 | content = self.content |
|
589 | 593 | if count_empty: |
|
590 | 594 | all_lines = 0 |
|
591 | 595 | empty_lines = 0 |
|
592 | 596 | for line in content.splitlines(True): |
|
593 | 597 | if line == '\n': |
|
594 | 598 | empty_lines += 1 |
|
595 | 599 | all_lines += 1 |
|
596 | 600 | |
|
597 | 601 | return all_lines, all_lines - empty_lines |
|
598 | 602 | else: |
|
599 | 603 | # fast method |
|
600 | 604 | empty_lines = all_lines = content.count('\n') |
|
601 | 605 | if all_lines == 0 and content: |
|
602 | 606 | # one-line without a newline |
|
603 | 607 | empty_lines = all_lines = 1 |
|
604 | 608 | |
|
605 | 609 | return all_lines, empty_lines |
|
606 | 610 | |
|
607 | 611 | def __repr__(self): |
|
608 | 612 | return '<%s %r @ %s>' % (self.__class__.__name__, self.path, |
|
609 | 613 | getattr(self.commit, 'short_id', '')) |
|
610 | 614 | |
|
611 | 615 | |
|
612 | 616 | class RemovedFileNode(FileNode): |
|
613 | 617 | """ |
|
614 | 618 | Dummy FileNode class - trying to access any public attribute except path, |
|
615 | 619 | name, kind or state (or methods/attributes checking those two) would raise |
|
616 | 620 | RemovedFileNodeError. |
|
617 | 621 | """ |
|
618 | 622 | ALLOWED_ATTRIBUTES = [ |
|
619 | 623 | 'name', 'path', 'state', 'is_root', 'is_file', 'is_dir', 'kind', |
|
620 | 624 | 'added', 'changed', 'not_changed', 'removed' |
|
621 | 625 | ] |
|
622 | 626 | |
|
623 | 627 | def __init__(self, path): |
|
624 | 628 | """ |
|
625 | 629 | :param path: relative path to the node |
|
626 | 630 | """ |
|
627 | 631 | super(RemovedFileNode, self).__init__(path=path) |
|
628 | 632 | |
|
629 | 633 | def __getattribute__(self, attr): |
|
630 | 634 | if attr.startswith('_') or attr in RemovedFileNode.ALLOWED_ATTRIBUTES: |
|
631 | 635 | return super(RemovedFileNode, self).__getattribute__(attr) |
|
632 | 636 | raise RemovedFileNodeError( |
|
633 | 637 | "Cannot access attribute %s on RemovedFileNode" % attr) |
|
634 | 638 | |
|
635 | 639 | @LazyProperty |
|
636 | 640 | def state(self): |
|
637 | 641 | return NodeState.REMOVED |
|
638 | 642 | |
|
639 | 643 | |
|
640 | 644 | class DirNode(Node): |
|
641 | 645 | """ |
|
642 | 646 | DirNode stores list of files and directories within this node. |
|
643 | 647 | Nodes may be used standalone but within repository context they |
|
644 | 648 | lazily fetch data within same repositorty's commit. |
|
645 | 649 | """ |
|
646 | 650 | |
|
647 | 651 | def __init__(self, path, nodes=(), commit=None): |
|
648 | 652 | """ |
|
649 | 653 | Only one of ``nodes`` and ``commit`` may be given. Passing both |
|
650 | 654 | would raise ``NodeError`` exception. |
|
651 | 655 | |
|
652 | 656 | :param path: relative path to the node |
|
653 | 657 | :param nodes: content may be passed to constructor |
|
654 | 658 | :param commit: if given, will use it to lazily fetch content |
|
655 | 659 | """ |
|
656 | 660 | if nodes and commit: |
|
657 | 661 | raise NodeError("Cannot use both nodes and commit") |
|
658 | 662 | super(DirNode, self).__init__(path, NodeKind.DIR) |
|
659 | 663 | self.commit = commit |
|
660 | 664 | self._nodes = nodes |
|
661 | 665 | |
|
662 | 666 | @LazyProperty |
|
663 | 667 | def content(self): |
|
664 | 668 | raise NodeError( |
|
665 | 669 | "%s represents a dir and has no `content` attribute" % self) |
|
666 | 670 | |
|
667 | 671 | @LazyProperty |
|
668 | 672 | def nodes(self): |
|
669 | 673 | if self.commit: |
|
670 | 674 | nodes = self.commit.get_nodes(self.path) |
|
671 | 675 | else: |
|
672 | 676 | nodes = self._nodes |
|
673 | 677 | self._nodes_dict = dict((node.path, node) for node in nodes) |
|
674 | 678 | return sorted(nodes) |
|
675 | 679 | |
|
676 | 680 | @LazyProperty |
|
677 | 681 | def files(self): |
|
678 | 682 | return sorted((node for node in self.nodes if node.is_file())) |
|
679 | 683 | |
|
680 | 684 | @LazyProperty |
|
681 | 685 | def dirs(self): |
|
682 | 686 | return sorted((node for node in self.nodes if node.is_dir())) |
|
683 | 687 | |
|
684 | 688 | def __iter__(self): |
|
685 | 689 | for node in self.nodes: |
|
686 | 690 | yield node |
|
687 | 691 | |
|
688 | 692 | def get_node(self, path): |
|
689 | 693 | """ |
|
690 | 694 | Returns node from within this particular ``DirNode``, so it is now |
|
691 | 695 | allowed to fetch, i.e. node located at 'docs/api/index.rst' from node |
|
692 | 696 | 'docs'. In order to access deeper nodes one must fetch nodes between |
|
693 | 697 | them first - this would work:: |
|
694 | 698 | |
|
695 | 699 | docs = root.get_node('docs') |
|
696 | 700 | docs.get_node('api').get_node('index.rst') |
|
697 | 701 | |
|
698 | 702 | :param: path - relative to the current node |
|
699 | 703 | |
|
700 | 704 | .. note:: |
|
701 | 705 | To access lazily (as in example above) node have to be initialized |
|
702 | 706 | with related commit object - without it node is out of |
|
703 | 707 | context and may know nothing about anything else than nearest |
|
704 | 708 | (located at same level) nodes. |
|
705 | 709 | """ |
|
706 | 710 | try: |
|
707 | 711 | path = path.rstrip('/') |
|
708 | 712 | if path == '': |
|
709 | 713 | raise NodeError("Cannot retrieve node without path") |
|
710 | 714 | self.nodes # access nodes first in order to set _nodes_dict |
|
711 | 715 | paths = path.split('/') |
|
712 | 716 | if len(paths) == 1: |
|
713 | 717 | if not self.is_root(): |
|
714 | 718 | path = '/'.join((self.path, paths[0])) |
|
715 | 719 | else: |
|
716 | 720 | path = paths[0] |
|
717 | 721 | return self._nodes_dict[path] |
|
718 | 722 | elif len(paths) > 1: |
|
719 | 723 | if self.commit is None: |
|
720 | 724 | raise NodeError( |
|
721 | 725 | "Cannot access deeper nodes without commit") |
|
722 | 726 | else: |
|
723 | 727 | path1, path2 = paths[0], '/'.join(paths[1:]) |
|
724 | 728 | return self.get_node(path1).get_node(path2) |
|
725 | 729 | else: |
|
726 | 730 | raise KeyError |
|
727 | 731 | except KeyError: |
|
728 | 732 | raise NodeError("Node does not exist at %s" % path) |
|
729 | 733 | |
|
730 | 734 | @LazyProperty |
|
731 | 735 | def state(self): |
|
732 | 736 | raise NodeError("Cannot access state of DirNode") |
|
733 | 737 | |
|
734 | 738 | @LazyProperty |
|
735 | 739 | def size(self): |
|
736 | 740 | size = 0 |
|
737 | 741 | for root, dirs, files in self.commit.walk(self.path): |
|
738 | 742 | for f in files: |
|
739 | 743 | size += f.size |
|
740 | 744 | |
|
741 | 745 | return size |
|
742 | 746 | |
|
743 | 747 | @LazyProperty |
|
744 | 748 | def last_commit(self): |
|
745 | 749 | if self.commit: |
|
746 | 750 | pre_load = ["author", "date", "message"] |
|
747 | 751 | return self.commit.get_path_commit(self.path, pre_load=pre_load) |
|
748 | 752 | raise NodeError( |
|
749 | 753 | "Cannot retrieve last commit of the file without " |
|
750 | 754 | "related commit attribute") |
|
751 | 755 | |
|
752 | 756 | def __repr__(self): |
|
753 | 757 | return '<%s %r @ %s>' % (self.__class__.__name__, self.path, |
|
754 | 758 | getattr(self.commit, 'short_id', '')) |
|
755 | 759 | |
|
756 | 760 | |
|
757 | 761 | class RootNode(DirNode): |
|
758 | 762 | """ |
|
759 | 763 | DirNode being the root node of the repository. |
|
760 | 764 | """ |
|
761 | 765 | |
|
762 | 766 | def __init__(self, nodes=(), commit=None): |
|
763 | 767 | super(RootNode, self).__init__(path='', nodes=nodes, commit=commit) |
|
764 | 768 | |
|
765 | 769 | def __repr__(self): |
|
766 | 770 | return '<%s>' % self.__class__.__name__ |
|
767 | 771 | |
|
768 | 772 | |
|
769 | 773 | class SubModuleNode(Node): |
|
770 | 774 | """ |
|
771 | 775 | represents a SubModule of Git or SubRepo of Mercurial |
|
772 | 776 | """ |
|
773 | 777 | is_binary = False |
|
774 | 778 | size = 0 |
|
775 | 779 | |
|
776 | 780 | def __init__(self, name, url=None, commit=None, alias=None): |
|
777 | 781 | self.path = name |
|
778 | 782 | self.kind = NodeKind.SUBMODULE |
|
779 | 783 | self.alias = alias |
|
780 | 784 | |
|
781 | 785 | # we have to use EmptyCommit here since this can point to svn/git/hg |
|
782 | 786 | # submodules we cannot get from repository |
|
783 | 787 | self.commit = EmptyCommit(str(commit), alias=alias) |
|
784 | 788 | self.url = url or self._extract_submodule_url() |
|
785 | 789 | |
|
786 | 790 | def __repr__(self): |
|
787 | 791 | return '<%s %r @ %s>' % (self.__class__.__name__, self.path, |
|
788 | 792 | getattr(self.commit, 'short_id', '')) |
|
789 | 793 | |
|
790 | 794 | def _extract_submodule_url(self): |
|
791 | 795 | # TODO: find a way to parse gits submodule file and extract the |
|
792 | 796 | # linking URL |
|
793 | 797 | return self.path |
|
794 | 798 | |
|
795 | 799 | @LazyProperty |
|
796 | 800 | def name(self): |
|
797 | 801 | """ |
|
798 | 802 | Returns name of the node so if its path |
|
799 | 803 | then only last part is returned. |
|
800 | 804 | """ |
|
801 | 805 | org = safe_unicode(self.path.rstrip('/').split('/')[-1]) |
|
802 | 806 | return u'%s @ %s' % (org, self.commit.short_id) |
|
803 | 807 | |
|
804 | 808 | |
|
805 | 809 | class LargeFileNode(FileNode): |
|
806 | 810 | |
|
807 | 811 | def __init__(self, path, url=None, commit=None, alias=None, org_path=None): |
|
808 | 812 | self.path = path |
|
809 | 813 | self.org_path = org_path |
|
810 | 814 | self.kind = NodeKind.LARGEFILE |
|
811 | 815 | self.alias = alias |
|
812 | 816 | |
|
813 | 817 | def _validate_path(self, path): |
|
814 | 818 | """ |
|
815 | 819 | we override check since the LargeFileNode path is system absolute |
|
816 | 820 | """ |
|
817 | 821 | pass |
|
818 | 822 | |
|
819 | 823 | def __repr__(self): |
|
820 | 824 | return '<%s %r>' % (self.__class__.__name__, self.path) |
|
821 | 825 | |
|
822 | 826 | @LazyProperty |
|
823 | 827 | def size(self): |
|
824 | 828 | return os.stat(self.path).st_size |
|
825 | 829 | |
|
826 | 830 | @LazyProperty |
|
827 | 831 | def raw_bytes(self): |
|
828 | 832 | with open(self.path, 'rb') as f: |
|
829 | 833 | content = f.read() |
|
830 | 834 | return content |
|
831 | 835 | |
|
832 | 836 | @LazyProperty |
|
833 | 837 | def name(self): |
|
834 | 838 | """ |
|
835 | 839 | Overwrites name to be the org lf path |
|
836 | 840 | """ |
|
837 | 841 | return self.org_path |
General Comments 0
You need to be logged in to leave comments.
Login now