]> Raphaƫl G. Git Repositories - youtubedl/blob - youtube_dl/utils.py
f6204692a81002cdfc44b02d183126e755283bd9
[youtubedl] / youtube_dl / utils.py
1 #!/usr/bin/env python
2 # coding: utf-8
3
4 from __future__ import unicode_literals
5
6 import base64
7 import binascii
8 import calendar
9 import codecs
10 import contextlib
11 import ctypes
12 import datetime
13 import email.utils
14 import email.header
15 import errno
16 import functools
17 import gzip
18 import io
19 import itertools
20 import json
21 import locale
22 import math
23 import operator
24 import os
25 import platform
26 import random
27 import re
28 import socket
29 import ssl
30 import subprocess
31 import sys
32 import tempfile
33 import traceback
34 import xml.etree.ElementTree
35 import zlib
36
37 from .compat import (
38 compat_HTMLParseError,
39 compat_HTMLParser,
40 compat_basestring,
41 compat_chr,
42 compat_cookiejar,
43 compat_ctypes_WINFUNCTYPE,
44 compat_etree_fromstring,
45 compat_expanduser,
46 compat_html_entities,
47 compat_html_entities_html5,
48 compat_http_client,
49 compat_integer_types,
50 compat_kwargs,
51 compat_os_name,
52 compat_parse_qs,
53 compat_shlex_quote,
54 compat_str,
55 compat_struct_pack,
56 compat_struct_unpack,
57 compat_urllib_error,
58 compat_urllib_parse,
59 compat_urllib_parse_urlencode,
60 compat_urllib_parse_urlparse,
61 compat_urllib_parse_unquote_plus,
62 compat_urllib_request,
63 compat_urlparse,
64 compat_xpath,
65 )
66
67 from .socks import (
68 ProxyType,
69 sockssocket,
70 )
71
72
73 def register_socks_protocols():
74 # "Register" SOCKS protocols
75 # In Python < 2.6.5, urlsplit() suffers from bug https://bugs.python.org/issue7904
76 # URLs with protocols not in urlparse.uses_netloc are not handled correctly
77 for scheme in ('socks', 'socks4', 'socks4a', 'socks5'):
78 if scheme not in compat_urlparse.uses_netloc:
79 compat_urlparse.uses_netloc.append(scheme)
80
81
82 # This is not clearly defined otherwise
83 compiled_regex_type = type(re.compile(''))
84
85
86 def random_user_agent():
87 _USER_AGENT_TPL = 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/%s Safari/537.36'
88 _CHROME_VERSIONS = (
89 '74.0.3729.129',
90 '76.0.3780.3',
91 '76.0.3780.2',
92 '74.0.3729.128',
93 '76.0.3780.1',
94 '76.0.3780.0',
95 '75.0.3770.15',
96 '74.0.3729.127',
97 '74.0.3729.126',
98 '76.0.3779.1',
99 '76.0.3779.0',
100 '75.0.3770.14',
101 '74.0.3729.125',
102 '76.0.3778.1',
103 '76.0.3778.0',
104 '75.0.3770.13',
105 '74.0.3729.124',
106 '74.0.3729.123',
107 '73.0.3683.121',
108 '76.0.3777.1',
109 '76.0.3777.0',
110 '75.0.3770.12',
111 '74.0.3729.122',
112 '76.0.3776.4',
113 '75.0.3770.11',
114 '74.0.3729.121',
115 '76.0.3776.3',
116 '76.0.3776.2',
117 '73.0.3683.120',
118 '74.0.3729.120',
119 '74.0.3729.119',
120 '74.0.3729.118',
121 '76.0.3776.1',
122 '76.0.3776.0',
123 '76.0.3775.5',
124 '75.0.3770.10',
125 '74.0.3729.117',
126 '76.0.3775.4',
127 '76.0.3775.3',
128 '74.0.3729.116',
129 '75.0.3770.9',
130 '76.0.3775.2',
131 '76.0.3775.1',
132 '76.0.3775.0',
133 '75.0.3770.8',
134 '74.0.3729.115',
135 '74.0.3729.114',
136 '76.0.3774.1',
137 '76.0.3774.0',
138 '75.0.3770.7',
139 '74.0.3729.113',
140 '74.0.3729.112',
141 '74.0.3729.111',
142 '76.0.3773.1',
143 '76.0.3773.0',
144 '75.0.3770.6',
145 '74.0.3729.110',
146 '74.0.3729.109',
147 '76.0.3772.1',
148 '76.0.3772.0',
149 '75.0.3770.5',
150 '74.0.3729.108',
151 '74.0.3729.107',
152 '76.0.3771.1',
153 '76.0.3771.0',
154 '75.0.3770.4',
155 '74.0.3729.106',
156 '74.0.3729.105',
157 '75.0.3770.3',
158 '74.0.3729.104',
159 '74.0.3729.103',
160 '74.0.3729.102',
161 '75.0.3770.2',
162 '74.0.3729.101',
163 '75.0.3770.1',
164 '75.0.3770.0',
165 '74.0.3729.100',
166 '75.0.3769.5',
167 '75.0.3769.4',
168 '74.0.3729.99',
169 '75.0.3769.3',
170 '75.0.3769.2',
171 '75.0.3768.6',
172 '74.0.3729.98',
173 '75.0.3769.1',
174 '75.0.3769.0',
175 '74.0.3729.97',
176 '73.0.3683.119',
177 '73.0.3683.118',
178 '74.0.3729.96',
179 '75.0.3768.5',
180 '75.0.3768.4',
181 '75.0.3768.3',
182 '75.0.3768.2',
183 '74.0.3729.95',
184 '74.0.3729.94',
185 '75.0.3768.1',
186 '75.0.3768.0',
187 '74.0.3729.93',
188 '74.0.3729.92',
189 '73.0.3683.117',
190 '74.0.3729.91',
191 '75.0.3766.3',
192 '74.0.3729.90',
193 '75.0.3767.2',
194 '75.0.3767.1',
195 '75.0.3767.0',
196 '74.0.3729.89',
197 '73.0.3683.116',
198 '75.0.3766.2',
199 '74.0.3729.88',
200 '75.0.3766.1',
201 '75.0.3766.0',
202 '74.0.3729.87',
203 '73.0.3683.115',
204 '74.0.3729.86',
205 '75.0.3765.1',
206 '75.0.3765.0',
207 '74.0.3729.85',
208 '73.0.3683.114',
209 '74.0.3729.84',
210 '75.0.3764.1',
211 '75.0.3764.0',
212 '74.0.3729.83',
213 '73.0.3683.113',
214 '75.0.3763.2',
215 '75.0.3761.4',
216 '74.0.3729.82',
217 '75.0.3763.1',
218 '75.0.3763.0',
219 '74.0.3729.81',
220 '73.0.3683.112',
221 '75.0.3762.1',
222 '75.0.3762.0',
223 '74.0.3729.80',
224 '75.0.3761.3',
225 '74.0.3729.79',
226 '73.0.3683.111',
227 '75.0.3761.2',
228 '74.0.3729.78',
229 '74.0.3729.77',
230 '75.0.3761.1',
231 '75.0.3761.0',
232 '73.0.3683.110',
233 '74.0.3729.76',
234 '74.0.3729.75',
235 '75.0.3760.0',
236 '74.0.3729.74',
237 '75.0.3759.8',
238 '75.0.3759.7',
239 '75.0.3759.6',
240 '74.0.3729.73',
241 '75.0.3759.5',
242 '74.0.3729.72',
243 '73.0.3683.109',
244 '75.0.3759.4',
245 '75.0.3759.3',
246 '74.0.3729.71',
247 '75.0.3759.2',
248 '74.0.3729.70',
249 '73.0.3683.108',
250 '74.0.3729.69',
251 '75.0.3759.1',
252 '75.0.3759.0',
253 '74.0.3729.68',
254 '73.0.3683.107',
255 '74.0.3729.67',
256 '75.0.3758.1',
257 '75.0.3758.0',
258 '74.0.3729.66',
259 '73.0.3683.106',
260 '74.0.3729.65',
261 '75.0.3757.1',
262 '75.0.3757.0',
263 '74.0.3729.64',
264 '73.0.3683.105',
265 '74.0.3729.63',
266 '75.0.3756.1',
267 '75.0.3756.0',
268 '74.0.3729.62',
269 '73.0.3683.104',
270 '75.0.3755.3',
271 '75.0.3755.2',
272 '73.0.3683.103',
273 '75.0.3755.1',
274 '75.0.3755.0',
275 '74.0.3729.61',
276 '73.0.3683.102',
277 '74.0.3729.60',
278 '75.0.3754.2',
279 '74.0.3729.59',
280 '75.0.3753.4',
281 '74.0.3729.58',
282 '75.0.3754.1',
283 '75.0.3754.0',
284 '74.0.3729.57',
285 '73.0.3683.101',
286 '75.0.3753.3',
287 '75.0.3752.2',
288 '75.0.3753.2',
289 '74.0.3729.56',
290 '75.0.3753.1',
291 '75.0.3753.0',
292 '74.0.3729.55',
293 '73.0.3683.100',
294 '74.0.3729.54',
295 '75.0.3752.1',
296 '75.0.3752.0',
297 '74.0.3729.53',
298 '73.0.3683.99',
299 '74.0.3729.52',
300 '75.0.3751.1',
301 '75.0.3751.0',
302 '74.0.3729.51',
303 '73.0.3683.98',
304 '74.0.3729.50',
305 '75.0.3750.0',
306 '74.0.3729.49',
307 '74.0.3729.48',
308 '74.0.3729.47',
309 '75.0.3749.3',
310 '74.0.3729.46',
311 '73.0.3683.97',
312 '75.0.3749.2',
313 '74.0.3729.45',
314 '75.0.3749.1',
315 '75.0.3749.0',
316 '74.0.3729.44',
317 '73.0.3683.96',
318 '74.0.3729.43',
319 '74.0.3729.42',
320 '75.0.3748.1',
321 '75.0.3748.0',
322 '74.0.3729.41',
323 '75.0.3747.1',
324 '73.0.3683.95',
325 '75.0.3746.4',
326 '74.0.3729.40',
327 '74.0.3729.39',
328 '75.0.3747.0',
329 '75.0.3746.3',
330 '75.0.3746.2',
331 '74.0.3729.38',
332 '75.0.3746.1',
333 '75.0.3746.0',
334 '74.0.3729.37',
335 '73.0.3683.94',
336 '75.0.3745.5',
337 '75.0.3745.4',
338 '75.0.3745.3',
339 '75.0.3745.2',
340 '74.0.3729.36',
341 '75.0.3745.1',
342 '75.0.3745.0',
343 '75.0.3744.2',
344 '74.0.3729.35',
345 '73.0.3683.93',
346 '74.0.3729.34',
347 '75.0.3744.1',
348 '75.0.3744.0',
349 '74.0.3729.33',
350 '73.0.3683.92',
351 '74.0.3729.32',
352 '74.0.3729.31',
353 '73.0.3683.91',
354 '75.0.3741.2',
355 '75.0.3740.5',
356 '74.0.3729.30',
357 '75.0.3741.1',
358 '75.0.3741.0',
359 '74.0.3729.29',
360 '75.0.3740.4',
361 '73.0.3683.90',
362 '74.0.3729.28',
363 '75.0.3740.3',
364 '73.0.3683.89',
365 '75.0.3740.2',
366 '74.0.3729.27',
367 '75.0.3740.1',
368 '75.0.3740.0',
369 '74.0.3729.26',
370 '73.0.3683.88',
371 '73.0.3683.87',
372 '74.0.3729.25',
373 '75.0.3739.1',
374 '75.0.3739.0',
375 '73.0.3683.86',
376 '74.0.3729.24',
377 '73.0.3683.85',
378 '75.0.3738.4',
379 '75.0.3738.3',
380 '75.0.3738.2',
381 '75.0.3738.1',
382 '75.0.3738.0',
383 '74.0.3729.23',
384 '73.0.3683.84',
385 '74.0.3729.22',
386 '74.0.3729.21',
387 '75.0.3737.1',
388 '75.0.3737.0',
389 '74.0.3729.20',
390 '73.0.3683.83',
391 '74.0.3729.19',
392 '75.0.3736.1',
393 '75.0.3736.0',
394 '74.0.3729.18',
395 '73.0.3683.82',
396 '74.0.3729.17',
397 '75.0.3735.1',
398 '75.0.3735.0',
399 '74.0.3729.16',
400 '73.0.3683.81',
401 '75.0.3734.1',
402 '75.0.3734.0',
403 '74.0.3729.15',
404 '73.0.3683.80',
405 '74.0.3729.14',
406 '75.0.3733.1',
407 '75.0.3733.0',
408 '75.0.3732.1',
409 '74.0.3729.13',
410 '74.0.3729.12',
411 '73.0.3683.79',
412 '74.0.3729.11',
413 '75.0.3732.0',
414 '74.0.3729.10',
415 '73.0.3683.78',
416 '74.0.3729.9',
417 '74.0.3729.8',
418 '74.0.3729.7',
419 '75.0.3731.3',
420 '75.0.3731.2',
421 '75.0.3731.0',
422 '74.0.3729.6',
423 '73.0.3683.77',
424 '73.0.3683.76',
425 '75.0.3730.5',
426 '75.0.3730.4',
427 '73.0.3683.75',
428 '74.0.3729.5',
429 '73.0.3683.74',
430 '75.0.3730.3',
431 '75.0.3730.2',
432 '74.0.3729.4',
433 '73.0.3683.73',
434 '73.0.3683.72',
435 '75.0.3730.1',
436 '75.0.3730.0',
437 '74.0.3729.3',
438 '73.0.3683.71',
439 '74.0.3729.2',
440 '73.0.3683.70',
441 '74.0.3729.1',
442 '74.0.3729.0',
443 '74.0.3726.4',
444 '73.0.3683.69',
445 '74.0.3726.3',
446 '74.0.3728.0',
447 '74.0.3726.2',
448 '73.0.3683.68',
449 '74.0.3726.1',
450 '74.0.3726.0',
451 '74.0.3725.4',
452 '73.0.3683.67',
453 '73.0.3683.66',
454 '74.0.3725.3',
455 '74.0.3725.2',
456 '74.0.3725.1',
457 '74.0.3724.8',
458 '74.0.3725.0',
459 '73.0.3683.65',
460 '74.0.3724.7',
461 '74.0.3724.6',
462 '74.0.3724.5',
463 '74.0.3724.4',
464 '74.0.3724.3',
465 '74.0.3724.2',
466 '74.0.3724.1',
467 '74.0.3724.0',
468 '73.0.3683.64',
469 '74.0.3723.1',
470 '74.0.3723.0',
471 '73.0.3683.63',
472 '74.0.3722.1',
473 '74.0.3722.0',
474 '73.0.3683.62',
475 '74.0.3718.9',
476 '74.0.3702.3',
477 '74.0.3721.3',
478 '74.0.3721.2',
479 '74.0.3721.1',
480 '74.0.3721.0',
481 '74.0.3720.6',
482 '73.0.3683.61',
483 '72.0.3626.122',
484 '73.0.3683.60',
485 '74.0.3720.5',
486 '72.0.3626.121',
487 '74.0.3718.8',
488 '74.0.3720.4',
489 '74.0.3720.3',
490 '74.0.3718.7',
491 '74.0.3720.2',
492 '74.0.3720.1',
493 '74.0.3720.0',
494 '74.0.3718.6',
495 '74.0.3719.5',
496 '73.0.3683.59',
497 '74.0.3718.5',
498 '74.0.3718.4',
499 '74.0.3719.4',
500 '74.0.3719.3',
501 '74.0.3719.2',
502 '74.0.3719.1',
503 '73.0.3683.58',
504 '74.0.3719.0',
505 '73.0.3683.57',
506 '73.0.3683.56',
507 '74.0.3718.3',
508 '73.0.3683.55',
509 '74.0.3718.2',
510 '74.0.3718.1',
511 '74.0.3718.0',
512 '73.0.3683.54',
513 '74.0.3717.2',
514 '73.0.3683.53',
515 '74.0.3717.1',
516 '74.0.3717.0',
517 '73.0.3683.52',
518 '74.0.3716.1',
519 '74.0.3716.0',
520 '73.0.3683.51',
521 '74.0.3715.1',
522 '74.0.3715.0',
523 '73.0.3683.50',
524 '74.0.3711.2',
525 '74.0.3714.2',
526 '74.0.3713.3',
527 '74.0.3714.1',
528 '74.0.3714.0',
529 '73.0.3683.49',
530 '74.0.3713.1',
531 '74.0.3713.0',
532 '72.0.3626.120',
533 '73.0.3683.48',
534 '74.0.3712.2',
535 '74.0.3712.1',
536 '74.0.3712.0',
537 '73.0.3683.47',
538 '72.0.3626.119',
539 '73.0.3683.46',
540 '74.0.3710.2',
541 '72.0.3626.118',
542 '74.0.3711.1',
543 '74.0.3711.0',
544 '73.0.3683.45',
545 '72.0.3626.117',
546 '74.0.3710.1',
547 '74.0.3710.0',
548 '73.0.3683.44',
549 '72.0.3626.116',
550 '74.0.3709.1',
551 '74.0.3709.0',
552 '74.0.3704.9',
553 '73.0.3683.43',
554 '72.0.3626.115',
555 '74.0.3704.8',
556 '74.0.3704.7',
557 '74.0.3708.0',
558 '74.0.3706.7',
559 '74.0.3704.6',
560 '73.0.3683.42',
561 '72.0.3626.114',
562 '74.0.3706.6',
563 '72.0.3626.113',
564 '74.0.3704.5',
565 '74.0.3706.5',
566 '74.0.3706.4',
567 '74.0.3706.3',
568 '74.0.3706.2',
569 '74.0.3706.1',
570 '74.0.3706.0',
571 '73.0.3683.41',
572 '72.0.3626.112',
573 '74.0.3705.1',
574 '74.0.3705.0',
575 '73.0.3683.40',
576 '72.0.3626.111',
577 '73.0.3683.39',
578 '74.0.3704.4',
579 '73.0.3683.38',
580 '74.0.3704.3',
581 '74.0.3704.2',
582 '74.0.3704.1',
583 '74.0.3704.0',
584 '73.0.3683.37',
585 '72.0.3626.110',
586 '72.0.3626.109',
587 '74.0.3703.3',
588 '74.0.3703.2',
589 '73.0.3683.36',
590 '74.0.3703.1',
591 '74.0.3703.0',
592 '73.0.3683.35',
593 '72.0.3626.108',
594 '74.0.3702.2',
595 '74.0.3699.3',
596 '74.0.3702.1',
597 '74.0.3702.0',
598 '73.0.3683.34',
599 '72.0.3626.107',
600 '73.0.3683.33',
601 '74.0.3701.1',
602 '74.0.3701.0',
603 '73.0.3683.32',
604 '73.0.3683.31',
605 '72.0.3626.105',
606 '74.0.3700.1',
607 '74.0.3700.0',
608 '73.0.3683.29',
609 '72.0.3626.103',
610 '74.0.3699.2',
611 '74.0.3699.1',
612 '74.0.3699.0',
613 '73.0.3683.28',
614 '72.0.3626.102',
615 '73.0.3683.27',
616 '73.0.3683.26',
617 '74.0.3698.0',
618 '74.0.3696.2',
619 '72.0.3626.101',
620 '73.0.3683.25',
621 '74.0.3696.1',
622 '74.0.3696.0',
623 '74.0.3694.8',
624 '72.0.3626.100',
625 '74.0.3694.7',
626 '74.0.3694.6',
627 '74.0.3694.5',
628 '74.0.3694.4',
629 '72.0.3626.99',
630 '72.0.3626.98',
631 '74.0.3694.3',
632 '73.0.3683.24',
633 '72.0.3626.97',
634 '72.0.3626.96',
635 '72.0.3626.95',
636 '73.0.3683.23',
637 '72.0.3626.94',
638 '73.0.3683.22',
639 '73.0.3683.21',
640 '72.0.3626.93',
641 '74.0.3694.2',
642 '72.0.3626.92',
643 '74.0.3694.1',
644 '74.0.3694.0',
645 '74.0.3693.6',
646 '73.0.3683.20',
647 '72.0.3626.91',
648 '74.0.3693.5',
649 '74.0.3693.4',
650 '74.0.3693.3',
651 '74.0.3693.2',
652 '73.0.3683.19',
653 '74.0.3693.1',
654 '74.0.3693.0',
655 '73.0.3683.18',
656 '72.0.3626.90',
657 '74.0.3692.1',
658 '74.0.3692.0',
659 '73.0.3683.17',
660 '72.0.3626.89',
661 '74.0.3687.3',
662 '74.0.3691.1',
663 '74.0.3691.0',
664 '73.0.3683.16',
665 '72.0.3626.88',
666 '72.0.3626.87',
667 '73.0.3683.15',
668 '74.0.3690.1',
669 '74.0.3690.0',
670 '73.0.3683.14',
671 '72.0.3626.86',
672 '73.0.3683.13',
673 '73.0.3683.12',
674 '74.0.3689.1',
675 '74.0.3689.0',
676 '73.0.3683.11',
677 '72.0.3626.85',
678 '73.0.3683.10',
679 '72.0.3626.84',
680 '73.0.3683.9',
681 '74.0.3688.1',
682 '74.0.3688.0',
683 '73.0.3683.8',
684 '72.0.3626.83',
685 '74.0.3687.2',
686 '74.0.3687.1',
687 '74.0.3687.0',
688 '73.0.3683.7',
689 '72.0.3626.82',
690 '74.0.3686.4',
691 '72.0.3626.81',
692 '74.0.3686.3',
693 '74.0.3686.2',
694 '74.0.3686.1',
695 '74.0.3686.0',
696 '73.0.3683.6',
697 '72.0.3626.80',
698 '74.0.3685.1',
699 '74.0.3685.0',
700 '73.0.3683.5',
701 '72.0.3626.79',
702 '74.0.3684.1',
703 '74.0.3684.0',
704 '73.0.3683.4',
705 '72.0.3626.78',
706 '72.0.3626.77',
707 '73.0.3683.3',
708 '73.0.3683.2',
709 '72.0.3626.76',
710 '73.0.3683.1',
711 '73.0.3683.0',
712 '72.0.3626.75',
713 '71.0.3578.141',
714 '73.0.3682.1',
715 '73.0.3682.0',
716 '72.0.3626.74',
717 '71.0.3578.140',
718 '73.0.3681.4',
719 '73.0.3681.3',
720 '73.0.3681.2',
721 '73.0.3681.1',
722 '73.0.3681.0',
723 '72.0.3626.73',
724 '71.0.3578.139',
725 '72.0.3626.72',
726 '72.0.3626.71',
727 '73.0.3680.1',
728 '73.0.3680.0',
729 '72.0.3626.70',
730 '71.0.3578.138',
731 '73.0.3678.2',
732 '73.0.3679.1',
733 '73.0.3679.0',
734 '72.0.3626.69',
735 '71.0.3578.137',
736 '73.0.3678.1',
737 '73.0.3678.0',
738 '71.0.3578.136',
739 '73.0.3677.1',
740 '73.0.3677.0',
741 '72.0.3626.68',
742 '72.0.3626.67',
743 '71.0.3578.135',
744 '73.0.3676.1',
745 '73.0.3676.0',
746 '73.0.3674.2',
747 '72.0.3626.66',
748 '71.0.3578.134',
749 '73.0.3674.1',
750 '73.0.3674.0',
751 '72.0.3626.65',
752 '71.0.3578.133',
753 '73.0.3673.2',
754 '73.0.3673.1',
755 '73.0.3673.0',
756 '72.0.3626.64',
757 '71.0.3578.132',
758 '72.0.3626.63',
759 '72.0.3626.62',
760 '72.0.3626.61',
761 '72.0.3626.60',
762 '73.0.3672.1',
763 '73.0.3672.0',
764 '72.0.3626.59',
765 '71.0.3578.131',
766 '73.0.3671.3',
767 '73.0.3671.2',
768 '73.0.3671.1',
769 '73.0.3671.0',
770 '72.0.3626.58',
771 '71.0.3578.130',
772 '73.0.3670.1',
773 '73.0.3670.0',
774 '72.0.3626.57',
775 '71.0.3578.129',
776 '73.0.3669.1',
777 '73.0.3669.0',
778 '72.0.3626.56',
779 '71.0.3578.128',
780 '73.0.3668.2',
781 '73.0.3668.1',
782 '73.0.3668.0',
783 '72.0.3626.55',
784 '71.0.3578.127',
785 '73.0.3667.2',
786 '73.0.3667.1',
787 '73.0.3667.0',
788 '72.0.3626.54',
789 '71.0.3578.126',
790 '73.0.3666.1',
791 '73.0.3666.0',
792 '72.0.3626.53',
793 '71.0.3578.125',
794 '73.0.3665.4',
795 '73.0.3665.3',
796 '72.0.3626.52',
797 '73.0.3665.2',
798 '73.0.3664.4',
799 '73.0.3665.1',
800 '73.0.3665.0',
801 '72.0.3626.51',
802 '71.0.3578.124',
803 '72.0.3626.50',
804 '73.0.3664.3',
805 '73.0.3664.2',
806 '73.0.3664.1',
807 '73.0.3664.0',
808 '73.0.3663.2',
809 '72.0.3626.49',
810 '71.0.3578.123',
811 '73.0.3663.1',
812 '73.0.3663.0',
813 '72.0.3626.48',
814 '71.0.3578.122',
815 '73.0.3662.1',
816 '73.0.3662.0',
817 '72.0.3626.47',
818 '71.0.3578.121',
819 '73.0.3661.1',
820 '72.0.3626.46',
821 '73.0.3661.0',
822 '72.0.3626.45',
823 '71.0.3578.120',
824 '73.0.3660.2',
825 '73.0.3660.1',
826 '73.0.3660.0',
827 '72.0.3626.44',
828 '71.0.3578.119',
829 '73.0.3659.1',
830 '73.0.3659.0',
831 '72.0.3626.43',
832 '71.0.3578.118',
833 '73.0.3658.1',
834 '73.0.3658.0',
835 '72.0.3626.42',
836 '71.0.3578.117',
837 '73.0.3657.1',
838 '73.0.3657.0',
839 '72.0.3626.41',
840 '71.0.3578.116',
841 '73.0.3656.1',
842 '73.0.3656.0',
843 '72.0.3626.40',
844 '71.0.3578.115',
845 '73.0.3655.1',
846 '73.0.3655.0',
847 '72.0.3626.39',
848 '71.0.3578.114',
849 '73.0.3654.1',
850 '73.0.3654.0',
851 '72.0.3626.38',
852 '71.0.3578.113',
853 '73.0.3653.1',
854 '73.0.3653.0',
855 '72.0.3626.37',
856 '71.0.3578.112',
857 '73.0.3652.1',
858 '73.0.3652.0',
859 '72.0.3626.36',
860 '71.0.3578.111',
861 '73.0.3651.1',
862 '73.0.3651.0',
863 '72.0.3626.35',
864 '71.0.3578.110',
865 '73.0.3650.1',
866 '73.0.3650.0',
867 '72.0.3626.34',
868 '71.0.3578.109',
869 '73.0.3649.1',
870 '73.0.3649.0',
871 '72.0.3626.33',
872 '71.0.3578.108',
873 '73.0.3648.2',
874 '73.0.3648.1',
875 '73.0.3648.0',
876 '72.0.3626.32',
877 '71.0.3578.107',
878 '73.0.3647.2',
879 '73.0.3647.1',
880 '73.0.3647.0',
881 '72.0.3626.31',
882 '71.0.3578.106',
883 '73.0.3635.3',
884 '73.0.3646.2',
885 '73.0.3646.1',
886 '73.0.3646.0',
887 '72.0.3626.30',
888 '71.0.3578.105',
889 '72.0.3626.29',
890 '73.0.3645.2',
891 '73.0.3645.1',
892 '73.0.3645.0',
893 '72.0.3626.28',
894 '71.0.3578.104',
895 '72.0.3626.27',
896 '72.0.3626.26',
897 '72.0.3626.25',
898 '72.0.3626.24',
899 '73.0.3644.0',
900 '73.0.3643.2',
901 '72.0.3626.23',
902 '71.0.3578.103',
903 '73.0.3643.1',
904 '73.0.3643.0',
905 '72.0.3626.22',
906 '71.0.3578.102',
907 '73.0.3642.1',
908 '73.0.3642.0',
909 '72.0.3626.21',
910 '71.0.3578.101',
911 '73.0.3641.1',
912 '73.0.3641.0',
913 '72.0.3626.20',
914 '71.0.3578.100',
915 '72.0.3626.19',
916 '73.0.3640.1',
917 '73.0.3640.0',
918 '72.0.3626.18',
919 '73.0.3639.1',
920 '71.0.3578.99',
921 '73.0.3639.0',
922 '72.0.3626.17',
923 '73.0.3638.2',
924 '72.0.3626.16',
925 '73.0.3638.1',
926 '73.0.3638.0',
927 '72.0.3626.15',
928 '71.0.3578.98',
929 '73.0.3635.2',
930 '71.0.3578.97',
931 '73.0.3637.1',
932 '73.0.3637.0',
933 '72.0.3626.14',
934 '71.0.3578.96',
935 '71.0.3578.95',
936 '72.0.3626.13',
937 '71.0.3578.94',
938 '73.0.3636.2',
939 '71.0.3578.93',
940 '73.0.3636.1',
941 '73.0.3636.0',
942 '72.0.3626.12',
943 '71.0.3578.92',
944 '73.0.3635.1',
945 '73.0.3635.0',
946 '72.0.3626.11',
947 '71.0.3578.91',
948 '73.0.3634.2',
949 '73.0.3634.1',
950 '73.0.3634.0',
951 '72.0.3626.10',
952 '71.0.3578.90',
953 '71.0.3578.89',
954 '73.0.3633.2',
955 '73.0.3633.1',
956 '73.0.3633.0',
957 '72.0.3610.4',
958 '72.0.3626.9',
959 '71.0.3578.88',
960 '73.0.3632.5',
961 '73.0.3632.4',
962 '73.0.3632.3',
963 '73.0.3632.2',
964 '73.0.3632.1',
965 '73.0.3632.0',
966 '72.0.3626.8',
967 '71.0.3578.87',
968 '73.0.3631.2',
969 '73.0.3631.1',
970 '73.0.3631.0',
971 '72.0.3626.7',
972 '71.0.3578.86',
973 '72.0.3626.6',
974 '73.0.3630.1',
975 '73.0.3630.0',
976 '72.0.3626.5',
977 '71.0.3578.85',
978 '72.0.3626.4',
979 '73.0.3628.3',
980 '73.0.3628.2',
981 '73.0.3629.1',
982 '73.0.3629.0',
983 '72.0.3626.3',
984 '71.0.3578.84',
985 '73.0.3628.1',
986 '73.0.3628.0',
987 '71.0.3578.83',
988 '73.0.3627.1',
989 '73.0.3627.0',
990 '72.0.3626.2',
991 '71.0.3578.82',
992 '71.0.3578.81',
993 '71.0.3578.80',
994 '72.0.3626.1',
995 '72.0.3626.0',
996 '71.0.3578.79',
997 '70.0.3538.124',
998 '71.0.3578.78',
999 '72.0.3623.4',
1000 '72.0.3625.2',
1001 '72.0.3625.1',
1002 '72.0.3625.0',
1003 '71.0.3578.77',
1004 '70.0.3538.123',
1005 '72.0.3624.4',
1006 '72.0.3624.3',
1007 '72.0.3624.2',
1008 '71.0.3578.76',
1009 '72.0.3624.1',
1010 '72.0.3624.0',
1011 '72.0.3623.3',
1012 '71.0.3578.75',
1013 '70.0.3538.122',
1014 '71.0.3578.74',
1015 '72.0.3623.2',
1016 '72.0.3610.3',
1017 '72.0.3623.1',
1018 '72.0.3623.0',
1019 '72.0.3622.3',
1020 '72.0.3622.2',
1021 '71.0.3578.73',
1022 '70.0.3538.121',
1023 '72.0.3622.1',
1024 '72.0.3622.0',
1025 '71.0.3578.72',
1026 '70.0.3538.120',
1027 '72.0.3621.1',
1028 '72.0.3621.0',
1029 '71.0.3578.71',
1030 '70.0.3538.119',
1031 '72.0.3620.1',
1032 '72.0.3620.0',
1033 '71.0.3578.70',
1034 '70.0.3538.118',
1035 '71.0.3578.69',
1036 '72.0.3619.1',
1037 '72.0.3619.0',
1038 '71.0.3578.68',
1039 '70.0.3538.117',
1040 '71.0.3578.67',
1041 '72.0.3618.1',
1042 '72.0.3618.0',
1043 '71.0.3578.66',
1044 '70.0.3538.116',
1045 '72.0.3617.1',
1046 '72.0.3617.0',
1047 '71.0.3578.65',
1048 '70.0.3538.115',
1049 '72.0.3602.3',
1050 '71.0.3578.64',
1051 '72.0.3616.1',
1052 '72.0.3616.0',
1053 '71.0.3578.63',
1054 '70.0.3538.114',
1055 '71.0.3578.62',
1056 '72.0.3615.1',
1057 '72.0.3615.0',
1058 '71.0.3578.61',
1059 '70.0.3538.113',
1060 '72.0.3614.1',
1061 '72.0.3614.0',
1062 '71.0.3578.60',
1063 '70.0.3538.112',
1064 '72.0.3613.1',
1065 '72.0.3613.0',
1066 '71.0.3578.59',
1067 '70.0.3538.111',
1068 '72.0.3612.2',
1069 '72.0.3612.1',
1070 '72.0.3612.0',
1071 '70.0.3538.110',
1072 '71.0.3578.58',
1073 '70.0.3538.109',
1074 '72.0.3611.2',
1075 '72.0.3611.1',
1076 '72.0.3611.0',
1077 '71.0.3578.57',
1078 '70.0.3538.108',
1079 '72.0.3610.2',
1080 '71.0.3578.56',
1081 '71.0.3578.55',
1082 '72.0.3610.1',
1083 '72.0.3610.0',
1084 '71.0.3578.54',
1085 '70.0.3538.107',
1086 '71.0.3578.53',
1087 '72.0.3609.3',
1088 '71.0.3578.52',
1089 '72.0.3609.2',
1090 '71.0.3578.51',
1091 '72.0.3608.5',
1092 '72.0.3609.1',
1093 '72.0.3609.0',
1094 '71.0.3578.50',
1095 '70.0.3538.106',
1096 '72.0.3608.4',
1097 '72.0.3608.3',
1098 '72.0.3608.2',
1099 '71.0.3578.49',
1100 '72.0.3608.1',
1101 '72.0.3608.0',
1102 '70.0.3538.105',
1103 '71.0.3578.48',
1104 '72.0.3607.1',
1105 '72.0.3607.0',
1106 '71.0.3578.47',
1107 '70.0.3538.104',
1108 '72.0.3606.2',
1109 '72.0.3606.1',
1110 '72.0.3606.0',
1111 '71.0.3578.46',
1112 '70.0.3538.103',
1113 '70.0.3538.102',
1114 '72.0.3605.3',
1115 '72.0.3605.2',
1116 '72.0.3605.1',
1117 '72.0.3605.0',
1118 '71.0.3578.45',
1119 '70.0.3538.101',
1120 '71.0.3578.44',
1121 '71.0.3578.43',
1122 '70.0.3538.100',
1123 '70.0.3538.99',
1124 '71.0.3578.42',
1125 '72.0.3604.1',
1126 '72.0.3604.0',
1127 '71.0.3578.41',
1128 '70.0.3538.98',
1129 '71.0.3578.40',
1130 '72.0.3603.2',
1131 '72.0.3603.1',
1132 '72.0.3603.0',
1133 '71.0.3578.39',
1134 '70.0.3538.97',
1135 '72.0.3602.2',
1136 '71.0.3578.38',
1137 '71.0.3578.37',
1138 '72.0.3602.1',
1139 '72.0.3602.0',
1140 '71.0.3578.36',
1141 '70.0.3538.96',
1142 '72.0.3601.1',
1143 '72.0.3601.0',
1144 '71.0.3578.35',
1145 '70.0.3538.95',
1146 '72.0.3600.1',
1147 '72.0.3600.0',
1148 '71.0.3578.34',
1149 '70.0.3538.94',
1150 '72.0.3599.3',
1151 '72.0.3599.2',
1152 '72.0.3599.1',
1153 '72.0.3599.0',
1154 '71.0.3578.33',
1155 '70.0.3538.93',
1156 '72.0.3598.1',
1157 '72.0.3598.0',
1158 '71.0.3578.32',
1159 '70.0.3538.87',
1160 '72.0.3597.1',
1161 '72.0.3597.0',
1162 '72.0.3596.2',
1163 '71.0.3578.31',
1164 '70.0.3538.86',
1165 '71.0.3578.30',
1166 '71.0.3578.29',
1167 '72.0.3596.1',
1168 '72.0.3596.0',
1169 '71.0.3578.28',
1170 '70.0.3538.85',
1171 '72.0.3595.2',
1172 '72.0.3591.3',
1173 '72.0.3595.1',
1174 '72.0.3595.0',
1175 '71.0.3578.27',
1176 '70.0.3538.84',
1177 '72.0.3594.1',
1178 '72.0.3594.0',
1179 '71.0.3578.26',
1180 '70.0.3538.83',
1181 '72.0.3593.2',
1182 '72.0.3593.1',
1183 '72.0.3593.0',
1184 '71.0.3578.25',
1185 '70.0.3538.82',
1186 '72.0.3589.3',
1187 '72.0.3592.2',
1188 '72.0.3592.1',
1189 '72.0.3592.0',
1190 '71.0.3578.24',
1191 '72.0.3589.2',
1192 '70.0.3538.81',
1193 '70.0.3538.80',
1194 '72.0.3591.2',
1195 '72.0.3591.1',
1196 '72.0.3591.0',
1197 '71.0.3578.23',
1198 '70.0.3538.79',
1199 '71.0.3578.22',
1200 '72.0.3590.1',
1201 '72.0.3590.0',
1202 '71.0.3578.21',
1203 '70.0.3538.78',
1204 '70.0.3538.77',
1205 '72.0.3589.1',
1206 '72.0.3589.0',
1207 '71.0.3578.20',
1208 '70.0.3538.76',
1209 '71.0.3578.19',
1210 '70.0.3538.75',
1211 '72.0.3588.1',
1212 '72.0.3588.0',
1213 '71.0.3578.18',
1214 '70.0.3538.74',
1215 '72.0.3586.2',
1216 '72.0.3587.0',
1217 '71.0.3578.17',
1218 '70.0.3538.73',
1219 '72.0.3586.1',
1220 '72.0.3586.0',
1221 '71.0.3578.16',
1222 '70.0.3538.72',
1223 '72.0.3585.1',
1224 '72.0.3585.0',
1225 '71.0.3578.15',
1226 '70.0.3538.71',
1227 '71.0.3578.14',
1228 '72.0.3584.1',
1229 '72.0.3584.0',
1230 '71.0.3578.13',
1231 '70.0.3538.70',
1232 '72.0.3583.2',
1233 '71.0.3578.12',
1234 '72.0.3583.1',
1235 '72.0.3583.0',
1236 '71.0.3578.11',
1237 '70.0.3538.69',
1238 '71.0.3578.10',
1239 '72.0.3582.0',
1240 '72.0.3581.4',
1241 '71.0.3578.9',
1242 '70.0.3538.67',
1243 '72.0.3581.3',
1244 '72.0.3581.2',
1245 '72.0.3581.1',
1246 '72.0.3581.0',
1247 '71.0.3578.8',
1248 '70.0.3538.66',
1249 '72.0.3580.1',
1250 '72.0.3580.0',
1251 '71.0.3578.7',
1252 '70.0.3538.65',
1253 '71.0.3578.6',
1254 '72.0.3579.1',
1255 '72.0.3579.0',
1256 '71.0.3578.5',
1257 '70.0.3538.64',
1258 '71.0.3578.4',
1259 '71.0.3578.3',
1260 '71.0.3578.2',
1261 '71.0.3578.1',
1262 '71.0.3578.0',
1263 '70.0.3538.63',
1264 '69.0.3497.128',
1265 '70.0.3538.62',
1266 '70.0.3538.61',
1267 '70.0.3538.60',
1268 '70.0.3538.59',
1269 '71.0.3577.1',
1270 '71.0.3577.0',
1271 '70.0.3538.58',
1272 '69.0.3497.127',
1273 '71.0.3576.2',
1274 '71.0.3576.1',
1275 '71.0.3576.0',
1276 '70.0.3538.57',
1277 '70.0.3538.56',
1278 '71.0.3575.2',
1279 '70.0.3538.55',
1280 '69.0.3497.126',
1281 '70.0.3538.54',
1282 '71.0.3575.1',
1283 '71.0.3575.0',
1284 '71.0.3574.1',
1285 '71.0.3574.0',
1286 '70.0.3538.53',
1287 '69.0.3497.125',
1288 '70.0.3538.52',
1289 '71.0.3573.1',
1290 '71.0.3573.0',
1291 '70.0.3538.51',
1292 '69.0.3497.124',
1293 '71.0.3572.1',
1294 '71.0.3572.0',
1295 '70.0.3538.50',
1296 '69.0.3497.123',
1297 '71.0.3571.2',
1298 '70.0.3538.49',
1299 '69.0.3497.122',
1300 '71.0.3571.1',
1301 '71.0.3571.0',
1302 '70.0.3538.48',
1303 '69.0.3497.121',
1304 '71.0.3570.1',
1305 '71.0.3570.0',
1306 '70.0.3538.47',
1307 '69.0.3497.120',
1308 '71.0.3568.2',
1309 '71.0.3569.1',
1310 '71.0.3569.0',
1311 '70.0.3538.46',
1312 '69.0.3497.119',
1313 '70.0.3538.45',
1314 '71.0.3568.1',
1315 '71.0.3568.0',
1316 '70.0.3538.44',
1317 '69.0.3497.118',
1318 '70.0.3538.43',
1319 '70.0.3538.42',
1320 '71.0.3567.1',
1321 '71.0.3567.0',
1322 '70.0.3538.41',
1323 '69.0.3497.117',
1324 '71.0.3566.1',
1325 '71.0.3566.0',
1326 '70.0.3538.40',
1327 '69.0.3497.116',
1328 '71.0.3565.1',
1329 '71.0.3565.0',
1330 '70.0.3538.39',
1331 '69.0.3497.115',
1332 '71.0.3564.1',
1333 '71.0.3564.0',
1334 '70.0.3538.38',
1335 '69.0.3497.114',
1336 '71.0.3563.0',
1337 '71.0.3562.2',
1338 '70.0.3538.37',
1339 '69.0.3497.113',
1340 '70.0.3538.36',
1341 '70.0.3538.35',
1342 '71.0.3562.1',
1343 '71.0.3562.0',
1344 '70.0.3538.34',
1345 '69.0.3497.112',
1346 '70.0.3538.33',
1347 '71.0.3561.1',
1348 '71.0.3561.0',
1349 '70.0.3538.32',
1350 '69.0.3497.111',
1351 '71.0.3559.6',
1352 '71.0.3560.1',
1353 '71.0.3560.0',
1354 '71.0.3559.5',
1355 '71.0.3559.4',
1356 '70.0.3538.31',
1357 '69.0.3497.110',
1358 '71.0.3559.3',
1359 '70.0.3538.30',
1360 '69.0.3497.109',
1361 '71.0.3559.2',
1362 '71.0.3559.1',
1363 '71.0.3559.0',
1364 '70.0.3538.29',
1365 '69.0.3497.108',
1366 '71.0.3558.2',
1367 '71.0.3558.1',
1368 '71.0.3558.0',
1369 '70.0.3538.28',
1370 '69.0.3497.107',
1371 '71.0.3557.2',
1372 '71.0.3557.1',
1373 '71.0.3557.0',
1374 '70.0.3538.27',
1375 '69.0.3497.106',
1376 '71.0.3554.4',
1377 '70.0.3538.26',
1378 '71.0.3556.1',
1379 '71.0.3556.0',
1380 '70.0.3538.25',
1381 '71.0.3554.3',
1382 '69.0.3497.105',
1383 '71.0.3554.2',
1384 '70.0.3538.24',
1385 '69.0.3497.104',
1386 '71.0.3555.2',
1387 '70.0.3538.23',
1388 '71.0.3555.1',
1389 '71.0.3555.0',
1390 '70.0.3538.22',
1391 '69.0.3497.103',
1392 '71.0.3554.1',
1393 '71.0.3554.0',
1394 '70.0.3538.21',
1395 '69.0.3497.102',
1396 '71.0.3553.3',
1397 '70.0.3538.20',
1398 '69.0.3497.101',
1399 '71.0.3553.2',
1400 '69.0.3497.100',
1401 '71.0.3553.1',
1402 '71.0.3553.0',
1403 '70.0.3538.19',
1404 '69.0.3497.99',
1405 '69.0.3497.98',
1406 '69.0.3497.97',
1407 '71.0.3552.6',
1408 '71.0.3552.5',
1409 '71.0.3552.4',
1410 '71.0.3552.3',
1411 '71.0.3552.2',
1412 '71.0.3552.1',
1413 '71.0.3552.0',
1414 '70.0.3538.18',
1415 '69.0.3497.96',
1416 '71.0.3551.3',
1417 '71.0.3551.2',
1418 '71.0.3551.1',
1419 '71.0.3551.0',
1420 '70.0.3538.17',
1421 '69.0.3497.95',
1422 '71.0.3550.3',
1423 '71.0.3550.2',
1424 '71.0.3550.1',
1425 '71.0.3550.0',
1426 '70.0.3538.16',
1427 '69.0.3497.94',
1428 '71.0.3549.1',
1429 '71.0.3549.0',
1430 '70.0.3538.15',
1431 '69.0.3497.93',
1432 '69.0.3497.92',
1433 '71.0.3548.1',
1434 '71.0.3548.0',
1435 '70.0.3538.14',
1436 '69.0.3497.91',
1437 '71.0.3547.1',
1438 '71.0.3547.0',
1439 '70.0.3538.13',
1440 '69.0.3497.90',
1441 '71.0.3546.2',
1442 '69.0.3497.89',
1443 '71.0.3546.1',
1444 '71.0.3546.0',
1445 '70.0.3538.12',
1446 '69.0.3497.88',
1447 '71.0.3545.4',
1448 '71.0.3545.3',
1449 '71.0.3545.2',
1450 '71.0.3545.1',
1451 '71.0.3545.0',
1452 '70.0.3538.11',
1453 '69.0.3497.87',
1454 '71.0.3544.5',
1455 '71.0.3544.4',
1456 '71.0.3544.3',
1457 '71.0.3544.2',
1458 '71.0.3544.1',
1459 '71.0.3544.0',
1460 '69.0.3497.86',
1461 '70.0.3538.10',
1462 '69.0.3497.85',
1463 '70.0.3538.9',
1464 '69.0.3497.84',
1465 '71.0.3543.4',
1466 '70.0.3538.8',
1467 '71.0.3543.3',
1468 '71.0.3543.2',
1469 '71.0.3543.1',
1470 '71.0.3543.0',
1471 '70.0.3538.7',
1472 '69.0.3497.83',
1473 '71.0.3542.2',
1474 '71.0.3542.1',
1475 '71.0.3542.0',
1476 '70.0.3538.6',
1477 '69.0.3497.82',
1478 '69.0.3497.81',
1479 '71.0.3541.1',
1480 '71.0.3541.0',
1481 '70.0.3538.5',
1482 '69.0.3497.80',
1483 '71.0.3540.1',
1484 '71.0.3540.0',
1485 '70.0.3538.4',
1486 '69.0.3497.79',
1487 '70.0.3538.3',
1488 '71.0.3539.1',
1489 '71.0.3539.0',
1490 '69.0.3497.78',
1491 '68.0.3440.134',
1492 '69.0.3497.77',
1493 '70.0.3538.2',
1494 '70.0.3538.1',
1495 '70.0.3538.0',
1496 '69.0.3497.76',
1497 '68.0.3440.133',
1498 '69.0.3497.75',
1499 '70.0.3537.2',
1500 '70.0.3537.1',
1501 '70.0.3537.0',
1502 '69.0.3497.74',
1503 '68.0.3440.132',
1504 '70.0.3536.0',
1505 '70.0.3535.5',
1506 '70.0.3535.4',
1507 '70.0.3535.3',
1508 '69.0.3497.73',
1509 '68.0.3440.131',
1510 '70.0.3532.8',
1511 '70.0.3532.7',
1512 '69.0.3497.72',
1513 '69.0.3497.71',
1514 '70.0.3535.2',
1515 '70.0.3535.1',
1516 '70.0.3535.0',
1517 '69.0.3497.70',
1518 '68.0.3440.130',
1519 '69.0.3497.69',
1520 '68.0.3440.129',
1521 '70.0.3534.4',
1522 '70.0.3534.3',
1523 '70.0.3534.2',
1524 '70.0.3534.1',
1525 '70.0.3534.0',
1526 '69.0.3497.68',
1527 '68.0.3440.128',
1528 '70.0.3533.2',
1529 '70.0.3533.1',
1530 '70.0.3533.0',
1531 '69.0.3497.67',
1532 '68.0.3440.127',
1533 '70.0.3532.6',
1534 '70.0.3532.5',
1535 '70.0.3532.4',
1536 '69.0.3497.66',
1537 '68.0.3440.126',
1538 '70.0.3532.3',
1539 '70.0.3532.2',
1540 '70.0.3532.1',
1541 '69.0.3497.60',
1542 '69.0.3497.65',
1543 '69.0.3497.64',
1544 '70.0.3532.0',
1545 '70.0.3531.0',
1546 '70.0.3530.4',
1547 '70.0.3530.3',
1548 '70.0.3530.2',
1549 '69.0.3497.58',
1550 '68.0.3440.125',
1551 '69.0.3497.57',
1552 '69.0.3497.56',
1553 '69.0.3497.55',
1554 '69.0.3497.54',
1555 '70.0.3530.1',
1556 '70.0.3530.0',
1557 '69.0.3497.53',
1558 '68.0.3440.124',
1559 '69.0.3497.52',
1560 '70.0.3529.3',
1561 '70.0.3529.2',
1562 '70.0.3529.1',
1563 '70.0.3529.0',
1564 '69.0.3497.51',
1565 '70.0.3528.4',
1566 '68.0.3440.123',
1567 '70.0.3528.3',
1568 '70.0.3528.2',
1569 '70.0.3528.1',
1570 '70.0.3528.0',
1571 '69.0.3497.50',
1572 '68.0.3440.122',
1573 '70.0.3527.1',
1574 '70.0.3527.0',
1575 '69.0.3497.49',
1576 '68.0.3440.121',
1577 '70.0.3526.1',
1578 '70.0.3526.0',
1579 '68.0.3440.120',
1580 '69.0.3497.48',
1581 '69.0.3497.47',
1582 '68.0.3440.119',
1583 '68.0.3440.118',
1584 '70.0.3525.5',
1585 '70.0.3525.4',
1586 '70.0.3525.3',
1587 '68.0.3440.117',
1588 '69.0.3497.46',
1589 '70.0.3525.2',
1590 '70.0.3525.1',
1591 '70.0.3525.0',
1592 '69.0.3497.45',
1593 '68.0.3440.116',
1594 '70.0.3524.4',
1595 '70.0.3524.3',
1596 '69.0.3497.44',
1597 '70.0.3524.2',
1598 '70.0.3524.1',
1599 '70.0.3524.0',
1600 '70.0.3523.2',
1601 '69.0.3497.43',
1602 '68.0.3440.115',
1603 '70.0.3505.9',
1604 '69.0.3497.42',
1605 '70.0.3505.8',
1606 '70.0.3523.1',
1607 '70.0.3523.0',
1608 '69.0.3497.41',
1609 '68.0.3440.114',
1610 '70.0.3505.7',
1611 '69.0.3497.40',
1612 '70.0.3522.1',
1613 '70.0.3522.0',
1614 '70.0.3521.2',
1615 '69.0.3497.39',
1616 '68.0.3440.113',
1617 '70.0.3505.6',
1618 '70.0.3521.1',
1619 '70.0.3521.0',
1620 '69.0.3497.38',
1621 '68.0.3440.112',
1622 '70.0.3520.1',
1623 '70.0.3520.0',
1624 '69.0.3497.37',
1625 '68.0.3440.111',
1626 '70.0.3519.3',
1627 '70.0.3519.2',
1628 '70.0.3519.1',
1629 '70.0.3519.0',
1630 '69.0.3497.36',
1631 '68.0.3440.110',
1632 '70.0.3518.1',
1633 '70.0.3518.0',
1634 '69.0.3497.35',
1635 '69.0.3497.34',
1636 '68.0.3440.109',
1637 '70.0.3517.1',
1638 '70.0.3517.0',
1639 '69.0.3497.33',
1640 '68.0.3440.108',
1641 '69.0.3497.32',
1642 '70.0.3516.3',
1643 '70.0.3516.2',
1644 '70.0.3516.1',
1645 '70.0.3516.0',
1646 '69.0.3497.31',
1647 '68.0.3440.107',
1648 '70.0.3515.4',
1649 '68.0.3440.106',
1650 '70.0.3515.3',
1651 '70.0.3515.2',
1652 '70.0.3515.1',
1653 '70.0.3515.0',
1654 '69.0.3497.30',
1655 '68.0.3440.105',
1656 '68.0.3440.104',
1657 '70.0.3514.2',
1658 '70.0.3514.1',
1659 '70.0.3514.0',
1660 '69.0.3497.29',
1661 '68.0.3440.103',
1662 '70.0.3513.1',
1663 '70.0.3513.0',
1664 '69.0.3497.28',
1665 )
1666 return _USER_AGENT_TPL % random.choice(_CHROME_VERSIONS)
1667
1668
1669 std_headers = {
1670 'User-Agent': random_user_agent(),
1671 'Accept-Charset': 'ISO-8859-1,utf-8;q=0.7,*;q=0.7',
1672 'Accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8',
1673 'Accept-Encoding': 'gzip, deflate',
1674 'Accept-Language': 'en-us,en;q=0.5',
1675 }
1676
1677
1678 USER_AGENTS = {
1679 'Safari': 'Mozilla/5.0 (X11; Linux x86_64; rv:10.0) AppleWebKit/533.20.25 (KHTML, like Gecko) Version/5.0.4 Safari/533.20.27',
1680 }
1681
1682
1683 NO_DEFAULT = object()
1684
1685 ENGLISH_MONTH_NAMES = [
1686 'January', 'February', 'March', 'April', 'May', 'June',
1687 'July', 'August', 'September', 'October', 'November', 'December']
1688
1689 MONTH_NAMES = {
1690 'en': ENGLISH_MONTH_NAMES,
1691 'fr': [
1692 'janvier', 'fƩvrier', 'mars', 'avril', 'mai', 'juin',
1693 'juillet', 'aoƻt', 'septembre', 'octobre', 'novembre', 'dƩcembre'],
1694 }
1695
1696 KNOWN_EXTENSIONS = (
1697 'mp4', 'm4a', 'm4p', 'm4b', 'm4r', 'm4v', 'aac',
1698 'flv', 'f4v', 'f4a', 'f4b',
1699 'webm', 'ogg', 'ogv', 'oga', 'ogx', 'spx', 'opus',
1700 'mkv', 'mka', 'mk3d',
1701 'avi', 'divx',
1702 'mov',
1703 'asf', 'wmv', 'wma',
1704 '3gp', '3g2',
1705 'mp3',
1706 'flac',
1707 'ape',
1708 'wav',
1709 'f4f', 'f4m', 'm3u8', 'smil')
1710
1711 # needed for sanitizing filenames in restricted mode
1712 ACCENT_CHARS = dict(zip('Ć‚ĆƒĆ„Ć€ĆĆ…Ć†Ć‡ĆˆĆ‰ĆŠĆ‹ĆŒĆĆŽĆĆĆ‘Ć’Ć“Ć”Ć•Ć–ÅĆ˜Å’Ć™ĆšĆ›ĆœÅ°ĆĆžĆŸĆ Ć”Ć¢Ć£Ć¤Ć„Ć¦Ć§ĆØĆ©ĆŖƫƬƭƮĆÆĆ°Ć±Ć²Ć³Ć“ĆµĆ¶Å‘ĆøÅ“Ć¹ĆŗĆ»Ć¼Å±Ć½Ć¾Ćæ',
1713 itertools.chain('AAAAAA', ['AE'], 'CEEEEIIIIDNOOOOOOO', ['OE'], 'UUUUUY', ['TH', 'ss'],
1714 'aaaaaa', ['ae'], 'ceeeeiiiionooooooo', ['oe'], 'uuuuuy', ['th'], 'y')))
1715
1716 DATE_FORMATS = (
1717 '%d %B %Y',
1718 '%d %b %Y',
1719 '%B %d %Y',
1720 '%B %dst %Y',
1721 '%B %dnd %Y',
1722 '%B %drd %Y',
1723 '%B %dth %Y',
1724 '%b %d %Y',
1725 '%b %dst %Y',
1726 '%b %dnd %Y',
1727 '%b %drd %Y',
1728 '%b %dth %Y',
1729 '%b %dst %Y %I:%M',
1730 '%b %dnd %Y %I:%M',
1731 '%b %drd %Y %I:%M',
1732 '%b %dth %Y %I:%M',
1733 '%Y %m %d',
1734 '%Y-%m-%d',
1735 '%Y/%m/%d',
1736 '%Y/%m/%d %H:%M',
1737 '%Y/%m/%d %H:%M:%S',
1738 '%Y-%m-%d %H:%M',
1739 '%Y-%m-%d %H:%M:%S',
1740 '%Y-%m-%d %H:%M:%S.%f',
1741 '%d.%m.%Y %H:%M',
1742 '%d.%m.%Y %H.%M',
1743 '%Y-%m-%dT%H:%M:%SZ',
1744 '%Y-%m-%dT%H:%M:%S.%fZ',
1745 '%Y-%m-%dT%H:%M:%S.%f0Z',
1746 '%Y-%m-%dT%H:%M:%S',
1747 '%Y-%m-%dT%H:%M:%S.%f',
1748 '%Y-%m-%dT%H:%M',
1749 '%b %d %Y at %H:%M',
1750 '%b %d %Y at %H:%M:%S',
1751 '%B %d %Y at %H:%M',
1752 '%B %d %Y at %H:%M:%S',
1753 )
1754
1755 DATE_FORMATS_DAY_FIRST = list(DATE_FORMATS)
1756 DATE_FORMATS_DAY_FIRST.extend([
1757 '%d-%m-%Y',
1758 '%d.%m.%Y',
1759 '%d.%m.%y',
1760 '%d/%m/%Y',
1761 '%d/%m/%y',
1762 '%d/%m/%Y %H:%M:%S',
1763 ])
1764
1765 DATE_FORMATS_MONTH_FIRST = list(DATE_FORMATS)
1766 DATE_FORMATS_MONTH_FIRST.extend([
1767 '%m-%d-%Y',
1768 '%m.%d.%Y',
1769 '%m/%d/%Y',
1770 '%m/%d/%y',
1771 '%m/%d/%Y %H:%M:%S',
1772 ])
1773
1774 PACKED_CODES_RE = r"}\('(.+)',(\d+),(\d+),'([^']+)'\.split\('\|'\)"
1775 JSON_LD_RE = r'(?is)<script[^>]+type=(["\']?)application/ld\+json\1[^>]*>(?P<json_ld>.+?)</script>'
1776
1777
1778 def preferredencoding():
1779 """Get preferred encoding.
1780
1781 Returns the best encoding scheme for the system, based on
1782 locale.getpreferredencoding() and some further tweaks.
1783 """
1784 try:
1785 pref = locale.getpreferredencoding()
1786 'TEST'.encode(pref)
1787 except Exception:
1788 pref = 'UTF-8'
1789
1790 return pref
1791
1792
1793 def write_json_file(obj, fn):
1794 """ Encode obj as JSON and write it to fn, atomically if possible """
1795
1796 fn = encodeFilename(fn)
1797 if sys.version_info < (3, 0) and sys.platform != 'win32':
1798 encoding = get_filesystem_encoding()
1799 # os.path.basename returns a bytes object, but NamedTemporaryFile
1800 # will fail if the filename contains non ascii characters unless we
1801 # use a unicode object
1802 path_basename = lambda f: os.path.basename(fn).decode(encoding)
1803 # the same for os.path.dirname
1804 path_dirname = lambda f: os.path.dirname(fn).decode(encoding)
1805 else:
1806 path_basename = os.path.basename
1807 path_dirname = os.path.dirname
1808
1809 args = {
1810 'suffix': '.tmp',
1811 'prefix': path_basename(fn) + '.',
1812 'dir': path_dirname(fn),
1813 'delete': False,
1814 }
1815
1816 # In Python 2.x, json.dump expects a bytestream.
1817 # In Python 3.x, it writes to a character stream
1818 if sys.version_info < (3, 0):
1819 args['mode'] = 'wb'
1820 else:
1821 args.update({
1822 'mode': 'w',
1823 'encoding': 'utf-8',
1824 })
1825
1826 tf = tempfile.NamedTemporaryFile(**compat_kwargs(args))
1827
1828 try:
1829 with tf:
1830 json.dump(obj, tf)
1831 if sys.platform == 'win32':
1832 # Need to remove existing file on Windows, else os.rename raises
1833 # WindowsError or FileExistsError.
1834 try:
1835 os.unlink(fn)
1836 except OSError:
1837 pass
1838 os.rename(tf.name, fn)
1839 except Exception:
1840 try:
1841 os.remove(tf.name)
1842 except OSError:
1843 pass
1844 raise
1845
1846
1847 if sys.version_info >= (2, 7):
1848 def find_xpath_attr(node, xpath, key, val=None):
1849 """ Find the xpath xpath[@key=val] """
1850 assert re.match(r'^[a-zA-Z_-]+$', key)
1851 expr = xpath + ('[@%s]' % key if val is None else "[@%s='%s']" % (key, val))
1852 return node.find(expr)
1853 else:
1854 def find_xpath_attr(node, xpath, key, val=None):
1855 for f in node.findall(compat_xpath(xpath)):
1856 if key not in f.attrib:
1857 continue
1858 if val is None or f.attrib.get(key) == val:
1859 return f
1860 return None
1861
1862 # On python2.6 the xml.etree.ElementTree.Element methods don't support
1863 # the namespace parameter
1864
1865
1866 def xpath_with_ns(path, ns_map):
1867 components = [c.split(':') for c in path.split('/')]
1868 replaced = []
1869 for c in components:
1870 if len(c) == 1:
1871 replaced.append(c[0])
1872 else:
1873 ns, tag = c
1874 replaced.append('{%s}%s' % (ns_map[ns], tag))
1875 return '/'.join(replaced)
1876
1877
1878 def xpath_element(node, xpath, name=None, fatal=False, default=NO_DEFAULT):
1879 def _find_xpath(xpath):
1880 return node.find(compat_xpath(xpath))
1881
1882 if isinstance(xpath, (str, compat_str)):
1883 n = _find_xpath(xpath)
1884 else:
1885 for xp in xpath:
1886 n = _find_xpath(xp)
1887 if n is not None:
1888 break
1889
1890 if n is None:
1891 if default is not NO_DEFAULT:
1892 return default
1893 elif fatal:
1894 name = xpath if name is None else name
1895 raise ExtractorError('Could not find XML element %s' % name)
1896 else:
1897 return None
1898 return n
1899
1900
1901 def xpath_text(node, xpath, name=None, fatal=False, default=NO_DEFAULT):
1902 n = xpath_element(node, xpath, name, fatal=fatal, default=default)
1903 if n is None or n == default:
1904 return n
1905 if n.text is None:
1906 if default is not NO_DEFAULT:
1907 return default
1908 elif fatal:
1909 name = xpath if name is None else name
1910 raise ExtractorError('Could not find XML element\'s text %s' % name)
1911 else:
1912 return None
1913 return n.text
1914
1915
1916 def xpath_attr(node, xpath, key, name=None, fatal=False, default=NO_DEFAULT):
1917 n = find_xpath_attr(node, xpath, key)
1918 if n is None:
1919 if default is not NO_DEFAULT:
1920 return default
1921 elif fatal:
1922 name = '%s[@%s]' % (xpath, key) if name is None else name
1923 raise ExtractorError('Could not find XML attribute %s' % name)
1924 else:
1925 return None
1926 return n.attrib[key]
1927
1928
1929 def get_element_by_id(id, html):
1930 """Return the content of the tag with the specified ID in the passed HTML document"""
1931 return get_element_by_attribute('id', id, html)
1932
1933
1934 def get_element_by_class(class_name, html):
1935 """Return the content of the first tag with the specified class in the passed HTML document"""
1936 retval = get_elements_by_class(class_name, html)
1937 return retval[0] if retval else None
1938
1939
1940 def get_element_by_attribute(attribute, value, html, escape_value=True):
1941 retval = get_elements_by_attribute(attribute, value, html, escape_value)
1942 return retval[0] if retval else None
1943
1944
1945 def get_elements_by_class(class_name, html):
1946 """Return the content of all tags with the specified class in the passed HTML document as a list"""
1947 return get_elements_by_attribute(
1948 'class', r'[^\'"]*\b%s\b[^\'"]*' % re.escape(class_name),
1949 html, escape_value=False)
1950
1951
1952 def get_elements_by_attribute(attribute, value, html, escape_value=True):
1953 """Return the content of the tag with the specified attribute in the passed HTML document"""
1954
1955 value = re.escape(value) if escape_value else value
1956
1957 retlist = []
1958 for m in re.finditer(r'''(?xs)
1959 <([a-zA-Z0-9:._-]+)
1960 (?:\s+[a-zA-Z0-9:._-]+(?:=[a-zA-Z0-9:._-]*|="[^"]*"|='[^']*'|))*?
1961 \s+%s=['"]?%s['"]?
1962 (?:\s+[a-zA-Z0-9:._-]+(?:=[a-zA-Z0-9:._-]*|="[^"]*"|='[^']*'|))*?
1963 \s*>
1964 (?P<content>.*?)
1965 </\1>
1966 ''' % (re.escape(attribute), value), html):
1967 res = m.group('content')
1968
1969 if res.startswith('"') or res.startswith("'"):
1970 res = res[1:-1]
1971
1972 retlist.append(unescapeHTML(res))
1973
1974 return retlist
1975
1976
1977 class HTMLAttributeParser(compat_HTMLParser):
1978 """Trivial HTML parser to gather the attributes for a single element"""
1979 def __init__(self):
1980 self.attrs = {}
1981 compat_HTMLParser.__init__(self)
1982
1983 def handle_starttag(self, tag, attrs):
1984 self.attrs = dict(attrs)
1985
1986
1987 def extract_attributes(html_element):
1988 """Given a string for an HTML element such as
1989 <el
1990 a="foo" B="bar" c="&98;az" d=boz
1991 empty= noval entity="&amp;"
1992 sq='"' dq="'"
1993 >
1994 Decode and return a dictionary of attributes.
1995 {
1996 'a': 'foo', 'b': 'bar', c: 'baz', d: 'boz',
1997 'empty': '', 'noval': None, 'entity': '&',
1998 'sq': '"', 'dq': '\''
1999 }.
2000 NB HTMLParser is stricter in Python 2.6 & 3.2 than in later versions,
2001 but the cases in the unit test will work for all of 2.6, 2.7, 3.2-3.5.
2002 """
2003 parser = HTMLAttributeParser()
2004 try:
2005 parser.feed(html_element)
2006 parser.close()
2007 # Older Python may throw HTMLParseError in case of malformed HTML
2008 except compat_HTMLParseError:
2009 pass
2010 return parser.attrs
2011
2012
2013 def clean_html(html):
2014 """Clean an HTML snippet into a readable string"""
2015
2016 if html is None: # Convenience for sanitizing descriptions etc.
2017 return html
2018
2019 # Newline vs <br />
2020 html = html.replace('\n', ' ')
2021 html = re.sub(r'(?u)\s*<\s*br\s*/?\s*>\s*', '\n', html)
2022 html = re.sub(r'(?u)<\s*/\s*p\s*>\s*<\s*p[^>]*>', '\n', html)
2023 # Strip html tags
2024 html = re.sub('<.*?>', '', html)
2025 # Replace html entities
2026 html = unescapeHTML(html)
2027 return html.strip()
2028
2029
2030 def sanitize_open(filename, open_mode):
2031 """Try to open the given filename, and slightly tweak it if this fails.
2032
2033 Attempts to open the given filename. If this fails, it tries to change
2034 the filename slightly, step by step, until it's either able to open it
2035 or it fails and raises a final exception, like the standard open()
2036 function.
2037
2038 It returns the tuple (stream, definitive_file_name).
2039 """
2040 try:
2041 if filename == '-':
2042 if sys.platform == 'win32':
2043 import msvcrt
2044 msvcrt.setmode(sys.stdout.fileno(), os.O_BINARY)
2045 return (sys.stdout.buffer if hasattr(sys.stdout, 'buffer') else sys.stdout, filename)
2046 stream = open(encodeFilename(filename), open_mode)
2047 return (stream, filename)
2048 except (IOError, OSError) as err:
2049 if err.errno in (errno.EACCES,):
2050 raise
2051
2052 # In case of error, try to remove win32 forbidden chars
2053 alt_filename = sanitize_path(filename)
2054 if alt_filename == filename:
2055 raise
2056 else:
2057 # An exception here should be caught in the caller
2058 stream = open(encodeFilename(alt_filename), open_mode)
2059 return (stream, alt_filename)
2060
2061
2062 def timeconvert(timestr):
2063 """Convert RFC 2822 defined time string into system timestamp"""
2064 timestamp = None
2065 timetuple = email.utils.parsedate_tz(timestr)
2066 if timetuple is not None:
2067 timestamp = email.utils.mktime_tz(timetuple)
2068 return timestamp
2069
2070
2071 def sanitize_filename(s, restricted=False, is_id=False):
2072 """Sanitizes a string so it could be used as part of a filename.
2073 If restricted is set, use a stricter subset of allowed characters.
2074 Set is_id if this is not an arbitrary string, but an ID that should be kept
2075 if possible.
2076 """
2077 def replace_insane(char):
2078 if restricted and char in ACCENT_CHARS:
2079 return ACCENT_CHARS[char]
2080 if char == '?' or ord(char) < 32 or ord(char) == 127:
2081 return ''
2082 elif char == '"':
2083 return '' if restricted else '\''
2084 elif char == ':':
2085 return '_-' if restricted else ' -'
2086 elif char in '\\/|*<>':
2087 return '_'
2088 if restricted and (char in '!&\'()[]{}$;`^,#' or char.isspace()):
2089 return '_'
2090 if restricted and ord(char) > 127:
2091 return '_'
2092 return char
2093
2094 # Handle timestamps
2095 s = re.sub(r'[0-9]+(?::[0-9]+)+', lambda m: m.group(0).replace(':', '_'), s)
2096 result = ''.join(map(replace_insane, s))
2097 if not is_id:
2098 while '__' in result:
2099 result = result.replace('__', '_')
2100 result = result.strip('_')
2101 # Common case of "Foreign band name - English song title"
2102 if restricted and result.startswith('-_'):
2103 result = result[2:]
2104 if result.startswith('-'):
2105 result = '_' + result[len('-'):]
2106 result = result.lstrip('.')
2107 if not result:
2108 result = '_'
2109 return result
2110
2111
2112 def sanitize_path(s):
2113 """Sanitizes and normalizes path on Windows"""
2114 if sys.platform != 'win32':
2115 return s
2116 drive_or_unc, _ = os.path.splitdrive(s)
2117 if sys.version_info < (2, 7) and not drive_or_unc:
2118 drive_or_unc, _ = os.path.splitunc(s)
2119 norm_path = os.path.normpath(remove_start(s, drive_or_unc)).split(os.path.sep)
2120 if drive_or_unc:
2121 norm_path.pop(0)
2122 sanitized_path = [
2123 path_part if path_part in ['.', '..'] else re.sub(r'(?:[/<>:"\|\\?\*]|[\s.]$)', '#', path_part)
2124 for path_part in norm_path]
2125 if drive_or_unc:
2126 sanitized_path.insert(0, drive_or_unc + os.path.sep)
2127 return os.path.join(*sanitized_path)
2128
2129
2130 def sanitize_url(url):
2131 # Prepend protocol-less URLs with `http:` scheme in order to mitigate
2132 # the number of unwanted failures due to missing protocol
2133 if url.startswith('//'):
2134 return 'http:%s' % url
2135 # Fix some common typos seen so far
2136 COMMON_TYPOS = (
2137 # https://github.com/ytdl-org/youtube-dl/issues/15649
2138 (r'^httpss://', r'https://'),
2139 # https://bx1.be/lives/direct-tv/
2140 (r'^rmtp([es]?)://', r'rtmp\1://'),
2141 )
2142 for mistake, fixup in COMMON_TYPOS:
2143 if re.match(mistake, url):
2144 return re.sub(mistake, fixup, url)
2145 return url
2146
2147
2148 def sanitized_Request(url, *args, **kwargs):
2149 return compat_urllib_request.Request(sanitize_url(url), *args, **kwargs)
2150
2151
2152 def expand_path(s):
2153 """Expand shell variables and ~"""
2154 return os.path.expandvars(compat_expanduser(s))
2155
2156
2157 def orderedSet(iterable):
2158 """ Remove all duplicates from the input iterable """
2159 res = []
2160 for el in iterable:
2161 if el not in res:
2162 res.append(el)
2163 return res
2164
2165
2166 def _htmlentity_transform(entity_with_semicolon):
2167 """Transforms an HTML entity to a character."""
2168 entity = entity_with_semicolon[:-1]
2169
2170 # Known non-numeric HTML entity
2171 if entity in compat_html_entities.name2codepoint:
2172 return compat_chr(compat_html_entities.name2codepoint[entity])
2173
2174 # TODO: HTML5 allows entities without a semicolon. For example,
2175 # '&Eacuteric' should be decoded as 'Ɖric'.
2176 if entity_with_semicolon in compat_html_entities_html5:
2177 return compat_html_entities_html5[entity_with_semicolon]
2178
2179 mobj = re.match(r'#(x[0-9a-fA-F]+|[0-9]+)', entity)
2180 if mobj is not None:
2181 numstr = mobj.group(1)
2182 if numstr.startswith('x'):
2183 base = 16
2184 numstr = '0%s' % numstr
2185 else:
2186 base = 10
2187 # See https://github.com/ytdl-org/youtube-dl/issues/7518
2188 try:
2189 return compat_chr(int(numstr, base))
2190 except ValueError:
2191 pass
2192
2193 # Unknown entity in name, return its literal representation
2194 return '&%s;' % entity
2195
2196
2197 def unescapeHTML(s):
2198 if s is None:
2199 return None
2200 assert type(s) == compat_str
2201
2202 return re.sub(
2203 r'&([^&;]+;)', lambda m: _htmlentity_transform(m.group(1)), s)
2204
2205
2206 def get_subprocess_encoding():
2207 if sys.platform == 'win32' and sys.getwindowsversion()[0] >= 5:
2208 # For subprocess calls, encode with locale encoding
2209 # Refer to http://stackoverflow.com/a/9951851/35070
2210 encoding = preferredencoding()
2211 else:
2212 encoding = sys.getfilesystemencoding()
2213 if encoding is None:
2214 encoding = 'utf-8'
2215 return encoding
2216
2217
2218 def encodeFilename(s, for_subprocess=False):
2219 """
2220 @param s The name of the file
2221 """
2222
2223 assert type(s) == compat_str
2224
2225 # Python 3 has a Unicode API
2226 if sys.version_info >= (3, 0):
2227 return s
2228
2229 # Pass '' directly to use Unicode APIs on Windows 2000 and up
2230 # (Detecting Windows NT 4 is tricky because 'major >= 4' would
2231 # match Windows 9x series as well. Besides, NT 4 is obsolete.)
2232 if not for_subprocess and sys.platform == 'win32' and sys.getwindowsversion()[0] >= 5:
2233 return s
2234
2235 # Jython assumes filenames are Unicode strings though reported as Python 2.x compatible
2236 if sys.platform.startswith('java'):
2237 return s
2238
2239 return s.encode(get_subprocess_encoding(), 'ignore')
2240
2241
2242 def decodeFilename(b, for_subprocess=False):
2243
2244 if sys.version_info >= (3, 0):
2245 return b
2246
2247 if not isinstance(b, bytes):
2248 return b
2249
2250 return b.decode(get_subprocess_encoding(), 'ignore')
2251
2252
2253 def encodeArgument(s):
2254 if not isinstance(s, compat_str):
2255 # Legacy code that uses byte strings
2256 # Uncomment the following line after fixing all post processors
2257 # assert False, 'Internal error: %r should be of type %r, is %r' % (s, compat_str, type(s))
2258 s = s.decode('ascii')
2259 return encodeFilename(s, True)
2260
2261
2262 def decodeArgument(b):
2263 return decodeFilename(b, True)
2264
2265
2266 def decodeOption(optval):
2267 if optval is None:
2268 return optval
2269 if isinstance(optval, bytes):
2270 optval = optval.decode(preferredencoding())
2271
2272 assert isinstance(optval, compat_str)
2273 return optval
2274
2275
2276 def formatSeconds(secs):
2277 if secs > 3600:
2278 return '%d:%02d:%02d' % (secs // 3600, (secs % 3600) // 60, secs % 60)
2279 elif secs > 60:
2280 return '%d:%02d' % (secs // 60, secs % 60)
2281 else:
2282 return '%d' % secs
2283
2284
2285 def make_HTTPS_handler(params, **kwargs):
2286 opts_no_check_certificate = params.get('nocheckcertificate', False)
2287 if hasattr(ssl, 'create_default_context'): # Python >= 3.4 or 2.7.9
2288 context = ssl.create_default_context(ssl.Purpose.SERVER_AUTH)
2289 if opts_no_check_certificate:
2290 context.check_hostname = False
2291 context.verify_mode = ssl.CERT_NONE
2292 try:
2293 return YoutubeDLHTTPSHandler(params, context=context, **kwargs)
2294 except TypeError:
2295 # Python 2.7.8
2296 # (create_default_context present but HTTPSHandler has no context=)
2297 pass
2298
2299 if sys.version_info < (3, 2):
2300 return YoutubeDLHTTPSHandler(params, **kwargs)
2301 else: # Python < 3.4
2302 context = ssl.SSLContext(ssl.PROTOCOL_TLSv1)
2303 context.verify_mode = (ssl.CERT_NONE
2304 if opts_no_check_certificate
2305 else ssl.CERT_REQUIRED)
2306 context.set_default_verify_paths()
2307 return YoutubeDLHTTPSHandler(params, context=context, **kwargs)
2308
2309
2310 def bug_reports_message():
2311 if ytdl_is_updateable():
2312 update_cmd = 'type youtube-dl -U to update'
2313 else:
2314 update_cmd = 'see https://yt-dl.org/update on how to update'
2315 msg = '; please report this issue on https://yt-dl.org/bug .'
2316 msg += ' Make sure you are using the latest version; %s.' % update_cmd
2317 msg += ' Be sure to call youtube-dl with the --verbose flag and include its complete output.'
2318 return msg
2319
2320
2321 class YoutubeDLError(Exception):
2322 """Base exception for YoutubeDL errors."""
2323 pass
2324
2325
2326 class ExtractorError(YoutubeDLError):
2327 """Error during info extraction."""
2328
2329 def __init__(self, msg, tb=None, expected=False, cause=None, video_id=None):
2330 """ tb, if given, is the original traceback (so that it can be printed out).
2331 If expected is set, this is a normal error message and most likely not a bug in youtube-dl.
2332 """
2333
2334 if sys.exc_info()[0] in (compat_urllib_error.URLError, socket.timeout, UnavailableVideoError):
2335 expected = True
2336 if video_id is not None:
2337 msg = video_id + ': ' + msg
2338 if cause:
2339 msg += ' (caused by %r)' % cause
2340 if not expected:
2341 msg += bug_reports_message()
2342 super(ExtractorError, self).__init__(msg)
2343
2344 self.traceback = tb
2345 self.exc_info = sys.exc_info() # preserve original exception
2346 self.cause = cause
2347 self.video_id = video_id
2348
2349 def format_traceback(self):
2350 if self.traceback is None:
2351 return None
2352 return ''.join(traceback.format_tb(self.traceback))
2353
2354
2355 class UnsupportedError(ExtractorError):
2356 def __init__(self, url):
2357 super(UnsupportedError, self).__init__(
2358 'Unsupported URL: %s' % url, expected=True)
2359 self.url = url
2360
2361
2362 class RegexNotFoundError(ExtractorError):
2363 """Error when a regex didn't match"""
2364 pass
2365
2366
2367 class GeoRestrictedError(ExtractorError):
2368 """Geographic restriction Error exception.
2369
2370 This exception may be thrown when a video is not available from your
2371 geographic location due to geographic restrictions imposed by a website.
2372 """
2373 def __init__(self, msg, countries=None):
2374 super(GeoRestrictedError, self).__init__(msg, expected=True)
2375 self.msg = msg
2376 self.countries = countries
2377
2378
2379 class DownloadError(YoutubeDLError):
2380 """Download Error exception.
2381
2382 This exception may be thrown by FileDownloader objects if they are not
2383 configured to continue on errors. They will contain the appropriate
2384 error message.
2385 """
2386
2387 def __init__(self, msg, exc_info=None):
2388 """ exc_info, if given, is the original exception that caused the trouble (as returned by sys.exc_info()). """
2389 super(DownloadError, self).__init__(msg)
2390 self.exc_info = exc_info
2391
2392
2393 class SameFileError(YoutubeDLError):
2394 """Same File exception.
2395
2396 This exception will be thrown by FileDownloader objects if they detect
2397 multiple files would have to be downloaded to the same file on disk.
2398 """
2399 pass
2400
2401
2402 class PostProcessingError(YoutubeDLError):
2403 """Post Processing exception.
2404
2405 This exception may be raised by PostProcessor's .run() method to
2406 indicate an error in the postprocessing task.
2407 """
2408
2409 def __init__(self, msg):
2410 super(PostProcessingError, self).__init__(msg)
2411 self.msg = msg
2412
2413
2414 class MaxDownloadsReached(YoutubeDLError):
2415 """ --max-downloads limit has been reached. """
2416 pass
2417
2418
2419 class UnavailableVideoError(YoutubeDLError):
2420 """Unavailable Format exception.
2421
2422 This exception will be thrown when a video is requested
2423 in a format that is not available for that video.
2424 """
2425 pass
2426
2427
2428 class ContentTooShortError(YoutubeDLError):
2429 """Content Too Short exception.
2430
2431 This exception may be raised by FileDownloader objects when a file they
2432 download is too small for what the server announced first, indicating
2433 the connection was probably interrupted.
2434 """
2435
2436 def __init__(self, downloaded, expected):
2437 super(ContentTooShortError, self).__init__(
2438 'Downloaded {0} bytes, expected {1} bytes'.format(downloaded, expected)
2439 )
2440 # Both in bytes
2441 self.downloaded = downloaded
2442 self.expected = expected
2443
2444
2445 class XAttrMetadataError(YoutubeDLError):
2446 def __init__(self, code=None, msg='Unknown error'):
2447 super(XAttrMetadataError, self).__init__(msg)
2448 self.code = code
2449 self.msg = msg
2450
2451 # Parsing code and msg
2452 if (self.code in (errno.ENOSPC, errno.EDQUOT)
2453 or 'No space left' in self.msg or 'Disk quota excedded' in self.msg):
2454 self.reason = 'NO_SPACE'
2455 elif self.code == errno.E2BIG or 'Argument list too long' in self.msg:
2456 self.reason = 'VALUE_TOO_LONG'
2457 else:
2458 self.reason = 'NOT_SUPPORTED'
2459
2460
2461 class XAttrUnavailableError(YoutubeDLError):
2462 pass
2463
2464
2465 def _create_http_connection(ydl_handler, http_class, is_https, *args, **kwargs):
2466 # Working around python 2 bug (see http://bugs.python.org/issue17849) by limiting
2467 # expected HTTP responses to meet HTTP/1.0 or later (see also
2468 # https://github.com/ytdl-org/youtube-dl/issues/6727)
2469 if sys.version_info < (3, 0):
2470 kwargs['strict'] = True
2471 hc = http_class(*args, **compat_kwargs(kwargs))
2472 source_address = ydl_handler._params.get('source_address')
2473
2474 if source_address is not None:
2475 # This is to workaround _create_connection() from socket where it will try all
2476 # address data from getaddrinfo() including IPv6. This filters the result from
2477 # getaddrinfo() based on the source_address value.
2478 # This is based on the cpython socket.create_connection() function.
2479 # https://github.com/python/cpython/blob/master/Lib/socket.py#L691
2480 def _create_connection(address, timeout=socket._GLOBAL_DEFAULT_TIMEOUT, source_address=None):
2481 host, port = address
2482 err = None
2483 addrs = socket.getaddrinfo(host, port, 0, socket.SOCK_STREAM)
2484 af = socket.AF_INET if '.' in source_address[0] else socket.AF_INET6
2485 ip_addrs = [addr for addr in addrs if addr[0] == af]
2486 if addrs and not ip_addrs:
2487 ip_version = 'v4' if af == socket.AF_INET else 'v6'
2488 raise socket.error(
2489 "No remote IP%s addresses available for connect, can't use '%s' as source address"
2490 % (ip_version, source_address[0]))
2491 for res in ip_addrs:
2492 af, socktype, proto, canonname, sa = res
2493 sock = None
2494 try:
2495 sock = socket.socket(af, socktype, proto)
2496 if timeout is not socket._GLOBAL_DEFAULT_TIMEOUT:
2497 sock.settimeout(timeout)
2498 sock.bind(source_address)
2499 sock.connect(sa)
2500 err = None # Explicitly break reference cycle
2501 return sock
2502 except socket.error as _:
2503 err = _
2504 if sock is not None:
2505 sock.close()
2506 if err is not None:
2507 raise err
2508 else:
2509 raise socket.error('getaddrinfo returns an empty list')
2510 if hasattr(hc, '_create_connection'):
2511 hc._create_connection = _create_connection
2512 sa = (source_address, 0)
2513 if hasattr(hc, 'source_address'): # Python 2.7+
2514 hc.source_address = sa
2515 else: # Python 2.6
2516 def _hc_connect(self, *args, **kwargs):
2517 sock = _create_connection(
2518 (self.host, self.port), self.timeout, sa)
2519 if is_https:
2520 self.sock = ssl.wrap_socket(
2521 sock, self.key_file, self.cert_file,
2522 ssl_version=ssl.PROTOCOL_TLSv1)
2523 else:
2524 self.sock = sock
2525 hc.connect = functools.partial(_hc_connect, hc)
2526
2527 return hc
2528
2529
2530 def handle_youtubedl_headers(headers):
2531 filtered_headers = headers
2532
2533 if 'Youtubedl-no-compression' in filtered_headers:
2534 filtered_headers = dict((k, v) for k, v in filtered_headers.items() if k.lower() != 'accept-encoding')
2535 del filtered_headers['Youtubedl-no-compression']
2536
2537 return filtered_headers
2538
2539
2540 class YoutubeDLHandler(compat_urllib_request.HTTPHandler):
2541 """Handler for HTTP requests and responses.
2542
2543 This class, when installed with an OpenerDirector, automatically adds
2544 the standard headers to every HTTP request and handles gzipped and
2545 deflated responses from web servers. If compression is to be avoided in
2546 a particular request, the original request in the program code only has
2547 to include the HTTP header "Youtubedl-no-compression", which will be
2548 removed before making the real request.
2549
2550 Part of this code was copied from:
2551
2552 http://techknack.net/python-urllib2-handlers/
2553
2554 Andrew Rowls, the author of that code, agreed to release it to the
2555 public domain.
2556 """
2557
2558 def __init__(self, params, *args, **kwargs):
2559 compat_urllib_request.HTTPHandler.__init__(self, *args, **kwargs)
2560 self._params = params
2561
2562 def http_open(self, req):
2563 conn_class = compat_http_client.HTTPConnection
2564
2565 socks_proxy = req.headers.get('Ytdl-socks-proxy')
2566 if socks_proxy:
2567 conn_class = make_socks_conn_class(conn_class, socks_proxy)
2568 del req.headers['Ytdl-socks-proxy']
2569
2570 return self.do_open(functools.partial(
2571 _create_http_connection, self, conn_class, False),
2572 req)
2573
2574 @staticmethod
2575 def deflate(data):
2576 try:
2577 return zlib.decompress(data, -zlib.MAX_WBITS)
2578 except zlib.error:
2579 return zlib.decompress(data)
2580
2581 def http_request(self, req):
2582 # According to RFC 3986, URLs can not contain non-ASCII characters, however this is not
2583 # always respected by websites, some tend to give out URLs with non percent-encoded
2584 # non-ASCII characters (see telemb.py, ard.py [#3412])
2585 # urllib chokes on URLs with non-ASCII characters (see http://bugs.python.org/issue3991)
2586 # To work around aforementioned issue we will replace request's original URL with
2587 # percent-encoded one
2588 # Since redirects are also affected (e.g. http://www.southpark.de/alle-episoden/s18e09)
2589 # the code of this workaround has been moved here from YoutubeDL.urlopen()
2590 url = req.get_full_url()
2591 url_escaped = escape_url(url)
2592
2593 # Substitute URL if any change after escaping
2594 if url != url_escaped:
2595 req = update_Request(req, url=url_escaped)
2596
2597 for h, v in std_headers.items():
2598 # Capitalize is needed because of Python bug 2275: http://bugs.python.org/issue2275
2599 # The dict keys are capitalized because of this bug by urllib
2600 if h.capitalize() not in req.headers:
2601 req.add_header(h, v)
2602
2603 req.headers = handle_youtubedl_headers(req.headers)
2604
2605 if sys.version_info < (2, 7) and '#' in req.get_full_url():
2606 # Python 2.6 is brain-dead when it comes to fragments
2607 req._Request__original = req._Request__original.partition('#')[0]
2608 req._Request__r_type = req._Request__r_type.partition('#')[0]
2609
2610 return req
2611
2612 def http_response(self, req, resp):
2613 old_resp = resp
2614 # gzip
2615 if resp.headers.get('Content-encoding', '') == 'gzip':
2616 content = resp.read()
2617 gz = gzip.GzipFile(fileobj=io.BytesIO(content), mode='rb')
2618 try:
2619 uncompressed = io.BytesIO(gz.read())
2620 except IOError as original_ioerror:
2621 # There may be junk add the end of the file
2622 # See http://stackoverflow.com/q/4928560/35070 for details
2623 for i in range(1, 1024):
2624 try:
2625 gz = gzip.GzipFile(fileobj=io.BytesIO(content[:-i]), mode='rb')
2626 uncompressed = io.BytesIO(gz.read())
2627 except IOError:
2628 continue
2629 break
2630 else:
2631 raise original_ioerror
2632 resp = compat_urllib_request.addinfourl(uncompressed, old_resp.headers, old_resp.url, old_resp.code)
2633 resp.msg = old_resp.msg
2634 del resp.headers['Content-encoding']
2635 # deflate
2636 if resp.headers.get('Content-encoding', '') == 'deflate':
2637 gz = io.BytesIO(self.deflate(resp.read()))
2638 resp = compat_urllib_request.addinfourl(gz, old_resp.headers, old_resp.url, old_resp.code)
2639 resp.msg = old_resp.msg
2640 del resp.headers['Content-encoding']
2641 # Percent-encode redirect URL of Location HTTP header to satisfy RFC 3986 (see
2642 # https://github.com/ytdl-org/youtube-dl/issues/6457).
2643 if 300 <= resp.code < 400:
2644 location = resp.headers.get('Location')
2645 if location:
2646 # As of RFC 2616 default charset is iso-8859-1 that is respected by python 3
2647 if sys.version_info >= (3, 0):
2648 location = location.encode('iso-8859-1').decode('utf-8')
2649 else:
2650 location = location.decode('utf-8')
2651 location_escaped = escape_url(location)
2652 if location != location_escaped:
2653 del resp.headers['Location']
2654 if sys.version_info < (3, 0):
2655 location_escaped = location_escaped.encode('utf-8')
2656 resp.headers['Location'] = location_escaped
2657 return resp
2658
2659 https_request = http_request
2660 https_response = http_response
2661
2662
2663 def make_socks_conn_class(base_class, socks_proxy):
2664 assert issubclass(base_class, (
2665 compat_http_client.HTTPConnection, compat_http_client.HTTPSConnection))
2666
2667 url_components = compat_urlparse.urlparse(socks_proxy)
2668 if url_components.scheme.lower() == 'socks5':
2669 socks_type = ProxyType.SOCKS5
2670 elif url_components.scheme.lower() in ('socks', 'socks4'):
2671 socks_type = ProxyType.SOCKS4
2672 elif url_components.scheme.lower() == 'socks4a':
2673 socks_type = ProxyType.SOCKS4A
2674
2675 def unquote_if_non_empty(s):
2676 if not s:
2677 return s
2678 return compat_urllib_parse_unquote_plus(s)
2679
2680 proxy_args = (
2681 socks_type,
2682 url_components.hostname, url_components.port or 1080,
2683 True, # Remote DNS
2684 unquote_if_non_empty(url_components.username),
2685 unquote_if_non_empty(url_components.password),
2686 )
2687
2688 class SocksConnection(base_class):
2689 def connect(self):
2690 self.sock = sockssocket()
2691 self.sock.setproxy(*proxy_args)
2692 if type(self.timeout) in (int, float):
2693 self.sock.settimeout(self.timeout)
2694 self.sock.connect((self.host, self.port))
2695
2696 if isinstance(self, compat_http_client.HTTPSConnection):
2697 if hasattr(self, '_context'): # Python > 2.6
2698 self.sock = self._context.wrap_socket(
2699 self.sock, server_hostname=self.host)
2700 else:
2701 self.sock = ssl.wrap_socket(self.sock)
2702
2703 return SocksConnection
2704
2705
2706 class YoutubeDLHTTPSHandler(compat_urllib_request.HTTPSHandler):
2707 def __init__(self, params, https_conn_class=None, *args, **kwargs):
2708 compat_urllib_request.HTTPSHandler.__init__(self, *args, **kwargs)
2709 self._https_conn_class = https_conn_class or compat_http_client.HTTPSConnection
2710 self._params = params
2711
2712 def https_open(self, req):
2713 kwargs = {}
2714 conn_class = self._https_conn_class
2715
2716 if hasattr(self, '_context'): # python > 2.6
2717 kwargs['context'] = self._context
2718 if hasattr(self, '_check_hostname'): # python 3.x
2719 kwargs['check_hostname'] = self._check_hostname
2720
2721 socks_proxy = req.headers.get('Ytdl-socks-proxy')
2722 if socks_proxy:
2723 conn_class = make_socks_conn_class(conn_class, socks_proxy)
2724 del req.headers['Ytdl-socks-proxy']
2725
2726 return self.do_open(functools.partial(
2727 _create_http_connection, self, conn_class, True),
2728 req, **kwargs)
2729
2730
2731 class YoutubeDLCookieJar(compat_cookiejar.MozillaCookieJar):
2732 _HTTPONLY_PREFIX = '#HttpOnly_'
2733
2734 def save(self, filename=None, ignore_discard=False, ignore_expires=False):
2735 # Store session cookies with `expires` set to 0 instead of an empty
2736 # string
2737 for cookie in self:
2738 if cookie.expires is None:
2739 cookie.expires = 0
2740 compat_cookiejar.MozillaCookieJar.save(self, filename, ignore_discard, ignore_expires)
2741
2742 def load(self, filename=None, ignore_discard=False, ignore_expires=False):
2743 """Load cookies from a file."""
2744 if filename is None:
2745 if self.filename is not None:
2746 filename = self.filename
2747 else:
2748 raise ValueError(compat_cookiejar.MISSING_FILENAME_TEXT)
2749
2750 cf = io.StringIO()
2751 with open(filename) as f:
2752 for line in f:
2753 if line.startswith(self._HTTPONLY_PREFIX):
2754 line = line[len(self._HTTPONLY_PREFIX):]
2755 cf.write(compat_str(line))
2756 cf.seek(0)
2757 self._really_load(cf, filename, ignore_discard, ignore_expires)
2758 # Session cookies are denoted by either `expires` field set to
2759 # an empty string or 0. MozillaCookieJar only recognizes the former
2760 # (see [1]). So we need force the latter to be recognized as session
2761 # cookies on our own.
2762 # Session cookies may be important for cookies-based authentication,
2763 # e.g. usually, when user does not check 'Remember me' check box while
2764 # logging in on a site, some important cookies are stored as session
2765 # cookies so that not recognizing them will result in failed login.
2766 # 1. https://bugs.python.org/issue17164
2767 for cookie in self:
2768 # Treat `expires=0` cookies as session cookies
2769 if cookie.expires == 0:
2770 cookie.expires = None
2771 cookie.discard = True
2772
2773
2774 class YoutubeDLCookieProcessor(compat_urllib_request.HTTPCookieProcessor):
2775 def __init__(self, cookiejar=None):
2776 compat_urllib_request.HTTPCookieProcessor.__init__(self, cookiejar)
2777
2778 def http_response(self, request, response):
2779 # Python 2 will choke on next HTTP request in row if there are non-ASCII
2780 # characters in Set-Cookie HTTP header of last response (see
2781 # https://github.com/ytdl-org/youtube-dl/issues/6769).
2782 # In order to at least prevent crashing we will percent encode Set-Cookie
2783 # header before HTTPCookieProcessor starts processing it.
2784 # if sys.version_info < (3, 0) and response.headers:
2785 # for set_cookie_header in ('Set-Cookie', 'Set-Cookie2'):
2786 # set_cookie = response.headers.get(set_cookie_header)
2787 # if set_cookie:
2788 # set_cookie_escaped = compat_urllib_parse.quote(set_cookie, b"%/;:@&=+$,!~*'()?#[] ")
2789 # if set_cookie != set_cookie_escaped:
2790 # del response.headers[set_cookie_header]
2791 # response.headers[set_cookie_header] = set_cookie_escaped
2792 return compat_urllib_request.HTTPCookieProcessor.http_response(self, request, response)
2793
2794 https_request = compat_urllib_request.HTTPCookieProcessor.http_request
2795 https_response = http_response
2796
2797
2798 def extract_timezone(date_str):
2799 m = re.search(
2800 r'^.{8,}?(?P<tz>Z$| ?(?P<sign>\+|-)(?P<hours>[0-9]{2}):?(?P<minutes>[0-9]{2})$)',
2801 date_str)
2802 if not m:
2803 timezone = datetime.timedelta()
2804 else:
2805 date_str = date_str[:-len(m.group('tz'))]
2806 if not m.group('sign'):
2807 timezone = datetime.timedelta()
2808 else:
2809 sign = 1 if m.group('sign') == '+' else -1
2810 timezone = datetime.timedelta(
2811 hours=sign * int(m.group('hours')),
2812 minutes=sign * int(m.group('minutes')))
2813 return timezone, date_str
2814
2815
2816 def parse_iso8601(date_str, delimiter='T', timezone=None):
2817 """ Return a UNIX timestamp from the given date """
2818
2819 if date_str is None:
2820 return None
2821
2822 date_str = re.sub(r'\.[0-9]+', '', date_str)
2823
2824 if timezone is None:
2825 timezone, date_str = extract_timezone(date_str)
2826
2827 try:
2828 date_format = '%Y-%m-%d{0}%H:%M:%S'.format(delimiter)
2829 dt = datetime.datetime.strptime(date_str, date_format) - timezone
2830 return calendar.timegm(dt.timetuple())
2831 except ValueError:
2832 pass
2833
2834
2835 def date_formats(day_first=True):
2836 return DATE_FORMATS_DAY_FIRST if day_first else DATE_FORMATS_MONTH_FIRST
2837
2838
2839 def unified_strdate(date_str, day_first=True):
2840 """Return a string with the date in the format YYYYMMDD"""
2841
2842 if date_str is None:
2843 return None
2844 upload_date = None
2845 # Replace commas
2846 date_str = date_str.replace(',', ' ')
2847 # Remove AM/PM + timezone
2848 date_str = re.sub(r'(?i)\s*(?:AM|PM)(?:\s+[A-Z]+)?', '', date_str)
2849 _, date_str = extract_timezone(date_str)
2850
2851 for expression in date_formats(day_first):
2852 try:
2853 upload_date = datetime.datetime.strptime(date_str, expression).strftime('%Y%m%d')
2854 except ValueError:
2855 pass
2856 if upload_date is None:
2857 timetuple = email.utils.parsedate_tz(date_str)
2858 if timetuple:
2859 try:
2860 upload_date = datetime.datetime(*timetuple[:6]).strftime('%Y%m%d')
2861 except ValueError:
2862 pass
2863 if upload_date is not None:
2864 return compat_str(upload_date)
2865
2866
2867 def unified_timestamp(date_str, day_first=True):
2868 if date_str is None:
2869 return None
2870
2871 date_str = re.sub(r'[,|]', '', date_str)
2872
2873 pm_delta = 12 if re.search(r'(?i)PM', date_str) else 0
2874 timezone, date_str = extract_timezone(date_str)
2875
2876 # Remove AM/PM + timezone
2877 date_str = re.sub(r'(?i)\s*(?:AM|PM)(?:\s+[A-Z]+)?', '', date_str)
2878
2879 # Remove unrecognized timezones from ISO 8601 alike timestamps
2880 m = re.search(r'\d{1,2}:\d{1,2}(?:\.\d+)?(?P<tz>\s*[A-Z]+)$', date_str)
2881 if m:
2882 date_str = date_str[:-len(m.group('tz'))]
2883
2884 # Python only supports microseconds, so remove nanoseconds
2885 m = re.search(r'^([0-9]{4,}-[0-9]{1,2}-[0-9]{1,2}T[0-9]{1,2}:[0-9]{1,2}:[0-9]{1,2}\.[0-9]{6})[0-9]+$', date_str)
2886 if m:
2887 date_str = m.group(1)
2888
2889 for expression in date_formats(day_first):
2890 try:
2891 dt = datetime.datetime.strptime(date_str, expression) - timezone + datetime.timedelta(hours=pm_delta)
2892 return calendar.timegm(dt.timetuple())
2893 except ValueError:
2894 pass
2895 timetuple = email.utils.parsedate_tz(date_str)
2896 if timetuple:
2897 return calendar.timegm(timetuple) + pm_delta * 3600
2898
2899
2900 def determine_ext(url, default_ext='unknown_video'):
2901 if url is None or '.' not in url:
2902 return default_ext
2903 guess = url.partition('?')[0].rpartition('.')[2]
2904 if re.match(r'^[A-Za-z0-9]+$', guess):
2905 return guess
2906 # Try extract ext from URLs like http://example.com/foo/bar.mp4/?download
2907 elif guess.rstrip('/') in KNOWN_EXTENSIONS:
2908 return guess.rstrip('/')
2909 else:
2910 return default_ext
2911
2912
2913 def subtitles_filename(filename, sub_lang, sub_format, expected_real_ext=None):
2914 return replace_extension(filename, sub_lang + '.' + sub_format, expected_real_ext)
2915
2916
2917 def date_from_str(date_str):
2918 """
2919 Return a datetime object from a string in the format YYYYMMDD or
2920 (now|today)[+-][0-9](day|week|month|year)(s)?"""
2921 today = datetime.date.today()
2922 if date_str in ('now', 'today'):
2923 return today
2924 if date_str == 'yesterday':
2925 return today - datetime.timedelta(days=1)
2926 match = re.match(r'(now|today)(?P<sign>[+-])(?P<time>\d+)(?P<unit>day|week|month|year)(s)?', date_str)
2927 if match is not None:
2928 sign = match.group('sign')
2929 time = int(match.group('time'))
2930 if sign == '-':
2931 time = -time
2932 unit = match.group('unit')
2933 # A bad approximation?
2934 if unit == 'month':
2935 unit = 'day'
2936 time *= 30
2937 elif unit == 'year':
2938 unit = 'day'
2939 time *= 365
2940 unit += 's'
2941 delta = datetime.timedelta(**{unit: time})
2942 return today + delta
2943 return datetime.datetime.strptime(date_str, '%Y%m%d').date()
2944
2945
2946 def hyphenate_date(date_str):
2947 """
2948 Convert a date in 'YYYYMMDD' format to 'YYYY-MM-DD' format"""
2949 match = re.match(r'^(\d\d\d\d)(\d\d)(\d\d)$', date_str)
2950 if match is not None:
2951 return '-'.join(match.groups())
2952 else:
2953 return date_str
2954
2955
2956 class DateRange(object):
2957 """Represents a time interval between two dates"""
2958
2959 def __init__(self, start=None, end=None):
2960 """start and end must be strings in the format accepted by date"""
2961 if start is not None:
2962 self.start = date_from_str(start)
2963 else:
2964 self.start = datetime.datetime.min.date()
2965 if end is not None:
2966 self.end = date_from_str(end)
2967 else:
2968 self.end = datetime.datetime.max.date()
2969 if self.start > self.end:
2970 raise ValueError('Date range: "%s" , the start date must be before the end date' % self)
2971
2972 @classmethod
2973 def day(cls, day):
2974 """Returns a range that only contains the given day"""
2975 return cls(day, day)
2976
2977 def __contains__(self, date):
2978 """Check if the date is in the range"""
2979 if not isinstance(date, datetime.date):
2980 date = date_from_str(date)
2981 return self.start <= date <= self.end
2982
2983 def __str__(self):
2984 return '%s - %s' % (self.start.isoformat(), self.end.isoformat())
2985
2986
2987 def platform_name():
2988 """ Returns the platform name as a compat_str """
2989 res = platform.platform()
2990 if isinstance(res, bytes):
2991 res = res.decode(preferredencoding())
2992
2993 assert isinstance(res, compat_str)
2994 return res
2995
2996
2997 def _windows_write_string(s, out):
2998 """ Returns True if the string was written using special methods,
2999 False if it has yet to be written out."""
3000 # Adapted from http://stackoverflow.com/a/3259271/35070
3001
3002 import ctypes
3003 import ctypes.wintypes
3004
3005 WIN_OUTPUT_IDS = {
3006 1: -11,
3007 2: -12,
3008 }
3009
3010 try:
3011 fileno = out.fileno()
3012 except AttributeError:
3013 # If the output stream doesn't have a fileno, it's virtual
3014 return False
3015 except io.UnsupportedOperation:
3016 # Some strange Windows pseudo files?
3017 return False
3018 if fileno not in WIN_OUTPUT_IDS:
3019 return False
3020
3021 GetStdHandle = compat_ctypes_WINFUNCTYPE(
3022 ctypes.wintypes.HANDLE, ctypes.wintypes.DWORD)(
3023 ('GetStdHandle', ctypes.windll.kernel32))
3024 h = GetStdHandle(WIN_OUTPUT_IDS[fileno])
3025
3026 WriteConsoleW = compat_ctypes_WINFUNCTYPE(
3027 ctypes.wintypes.BOOL, ctypes.wintypes.HANDLE, ctypes.wintypes.LPWSTR,
3028 ctypes.wintypes.DWORD, ctypes.POINTER(ctypes.wintypes.DWORD),
3029 ctypes.wintypes.LPVOID)(('WriteConsoleW', ctypes.windll.kernel32))
3030 written = ctypes.wintypes.DWORD(0)
3031
3032 GetFileType = compat_ctypes_WINFUNCTYPE(ctypes.wintypes.DWORD, ctypes.wintypes.DWORD)(('GetFileType', ctypes.windll.kernel32))
3033 FILE_TYPE_CHAR = 0x0002
3034 FILE_TYPE_REMOTE = 0x8000
3035 GetConsoleMode = compat_ctypes_WINFUNCTYPE(
3036 ctypes.wintypes.BOOL, ctypes.wintypes.HANDLE,
3037 ctypes.POINTER(ctypes.wintypes.DWORD))(
3038 ('GetConsoleMode', ctypes.windll.kernel32))
3039 INVALID_HANDLE_VALUE = ctypes.wintypes.DWORD(-1).value
3040
3041 def not_a_console(handle):
3042 if handle == INVALID_HANDLE_VALUE or handle is None:
3043 return True
3044 return ((GetFileType(handle) & ~FILE_TYPE_REMOTE) != FILE_TYPE_CHAR
3045 or GetConsoleMode(handle, ctypes.byref(ctypes.wintypes.DWORD())) == 0)
3046
3047 if not_a_console(h):
3048 return False
3049
3050 def next_nonbmp_pos(s):
3051 try:
3052 return next(i for i, c in enumerate(s) if ord(c) > 0xffff)
3053 except StopIteration:
3054 return len(s)
3055
3056 while s:
3057 count = min(next_nonbmp_pos(s), 1024)
3058
3059 ret = WriteConsoleW(
3060 h, s, count if count else 2, ctypes.byref(written), None)
3061 if ret == 0:
3062 raise OSError('Failed to write string')
3063 if not count: # We just wrote a non-BMP character
3064 assert written.value == 2
3065 s = s[1:]
3066 else:
3067 assert written.value > 0
3068 s = s[written.value:]
3069 return True
3070
3071
3072 def write_string(s, out=None, encoding=None):
3073 if out is None:
3074 out = sys.stderr
3075 assert type(s) == compat_str
3076
3077 if sys.platform == 'win32' and encoding is None and hasattr(out, 'fileno'):
3078 if _windows_write_string(s, out):
3079 return
3080
3081 if ('b' in getattr(out, 'mode', '')
3082 or sys.version_info[0] < 3): # Python 2 lies about mode of sys.stderr
3083 byt = s.encode(encoding or preferredencoding(), 'ignore')
3084 out.write(byt)
3085 elif hasattr(out, 'buffer'):
3086 enc = encoding or getattr(out, 'encoding', None) or preferredencoding()
3087 byt = s.encode(enc, 'ignore')
3088 out.buffer.write(byt)
3089 else:
3090 out.write(s)
3091 out.flush()
3092
3093
3094 def bytes_to_intlist(bs):
3095 if not bs:
3096 return []
3097 if isinstance(bs[0], int): # Python 3
3098 return list(bs)
3099 else:
3100 return [ord(c) for c in bs]
3101
3102
3103 def intlist_to_bytes(xs):
3104 if not xs:
3105 return b''
3106 return compat_struct_pack('%dB' % len(xs), *xs)
3107
3108
3109 # Cross-platform file locking
3110 if sys.platform == 'win32':
3111 import ctypes.wintypes
3112 import msvcrt
3113
3114 class OVERLAPPED(ctypes.Structure):
3115 _fields_ = [
3116 ('Internal', ctypes.wintypes.LPVOID),
3117 ('InternalHigh', ctypes.wintypes.LPVOID),
3118 ('Offset', ctypes.wintypes.DWORD),
3119 ('OffsetHigh', ctypes.wintypes.DWORD),
3120 ('hEvent', ctypes.wintypes.HANDLE),
3121 ]
3122
3123 kernel32 = ctypes.windll.kernel32
3124 LockFileEx = kernel32.LockFileEx
3125 LockFileEx.argtypes = [
3126 ctypes.wintypes.HANDLE, # hFile
3127 ctypes.wintypes.DWORD, # dwFlags
3128 ctypes.wintypes.DWORD, # dwReserved
3129 ctypes.wintypes.DWORD, # nNumberOfBytesToLockLow
3130 ctypes.wintypes.DWORD, # nNumberOfBytesToLockHigh
3131 ctypes.POINTER(OVERLAPPED) # Overlapped
3132 ]
3133 LockFileEx.restype = ctypes.wintypes.BOOL
3134 UnlockFileEx = kernel32.UnlockFileEx
3135 UnlockFileEx.argtypes = [
3136 ctypes.wintypes.HANDLE, # hFile
3137 ctypes.wintypes.DWORD, # dwReserved
3138 ctypes.wintypes.DWORD, # nNumberOfBytesToLockLow
3139 ctypes.wintypes.DWORD, # nNumberOfBytesToLockHigh
3140 ctypes.POINTER(OVERLAPPED) # Overlapped
3141 ]
3142 UnlockFileEx.restype = ctypes.wintypes.BOOL
3143 whole_low = 0xffffffff
3144 whole_high = 0x7fffffff
3145
3146 def _lock_file(f, exclusive):
3147 overlapped = OVERLAPPED()
3148 overlapped.Offset = 0
3149 overlapped.OffsetHigh = 0
3150 overlapped.hEvent = 0
3151 f._lock_file_overlapped_p = ctypes.pointer(overlapped)
3152 handle = msvcrt.get_osfhandle(f.fileno())
3153 if not LockFileEx(handle, 0x2 if exclusive else 0x0, 0,
3154 whole_low, whole_high, f._lock_file_overlapped_p):
3155 raise OSError('Locking file failed: %r' % ctypes.FormatError())
3156
3157 def _unlock_file(f):
3158 assert f._lock_file_overlapped_p
3159 handle = msvcrt.get_osfhandle(f.fileno())
3160 if not UnlockFileEx(handle, 0,
3161 whole_low, whole_high, f._lock_file_overlapped_p):
3162 raise OSError('Unlocking file failed: %r' % ctypes.FormatError())
3163
3164 else:
3165 # Some platforms, such as Jython, is missing fcntl
3166 try:
3167 import fcntl
3168
3169 def _lock_file(f, exclusive):
3170 fcntl.flock(f, fcntl.LOCK_EX if exclusive else fcntl.LOCK_SH)
3171
3172 def _unlock_file(f):
3173 fcntl.flock(f, fcntl.LOCK_UN)
3174 except ImportError:
3175 UNSUPPORTED_MSG = 'file locking is not supported on this platform'
3176
3177 def _lock_file(f, exclusive):
3178 raise IOError(UNSUPPORTED_MSG)
3179
3180 def _unlock_file(f):
3181 raise IOError(UNSUPPORTED_MSG)
3182
3183
3184 class locked_file(object):
3185 def __init__(self, filename, mode, encoding=None):
3186 assert mode in ['r', 'a', 'w']
3187 self.f = io.open(filename, mode, encoding=encoding)
3188 self.mode = mode
3189
3190 def __enter__(self):
3191 exclusive = self.mode != 'r'
3192 try:
3193 _lock_file(self.f, exclusive)
3194 except IOError:
3195 self.f.close()
3196 raise
3197 return self
3198
3199 def __exit__(self, etype, value, traceback):
3200 try:
3201 _unlock_file(self.f)
3202 finally:
3203 self.f.close()
3204
3205 def __iter__(self):
3206 return iter(self.f)
3207
3208 def write(self, *args):
3209 return self.f.write(*args)
3210
3211 def read(self, *args):
3212 return self.f.read(*args)
3213
3214
3215 def get_filesystem_encoding():
3216 encoding = sys.getfilesystemencoding()
3217 return encoding if encoding is not None else 'utf-8'
3218
3219
3220 def shell_quote(args):
3221 quoted_args = []
3222 encoding = get_filesystem_encoding()
3223 for a in args:
3224 if isinstance(a, bytes):
3225 # We may get a filename encoded with 'encodeFilename'
3226 a = a.decode(encoding)
3227 quoted_args.append(compat_shlex_quote(a))
3228 return ' '.join(quoted_args)
3229
3230
3231 def smuggle_url(url, data):
3232 """ Pass additional data in a URL for internal use. """
3233
3234 url, idata = unsmuggle_url(url, {})
3235 data.update(idata)
3236 sdata = compat_urllib_parse_urlencode(
3237 {'__youtubedl_smuggle': json.dumps(data)})
3238 return url + '#' + sdata
3239
3240
3241 def unsmuggle_url(smug_url, default=None):
3242 if '#__youtubedl_smuggle' not in smug_url:
3243 return smug_url, default
3244 url, _, sdata = smug_url.rpartition('#')
3245 jsond = compat_parse_qs(sdata)['__youtubedl_smuggle'][0]
3246 data = json.loads(jsond)
3247 return url, data
3248
3249
3250 def format_bytes(bytes):
3251 if bytes is None:
3252 return 'N/A'
3253 if type(bytes) is str:
3254 bytes = float(bytes)
3255 if bytes == 0.0:
3256 exponent = 0
3257 else:
3258 exponent = int(math.log(bytes, 1024.0))
3259 suffix = ['B', 'KiB', 'MiB', 'GiB', 'TiB', 'PiB', 'EiB', 'ZiB', 'YiB'][exponent]
3260 converted = float(bytes) / float(1024 ** exponent)
3261 return '%.2f%s' % (converted, suffix)
3262
3263
3264 def lookup_unit_table(unit_table, s):
3265 units_re = '|'.join(re.escape(u) for u in unit_table)
3266 m = re.match(
3267 r'(?P<num>[0-9]+(?:[,.][0-9]*)?)\s*(?P<unit>%s)\b' % units_re, s)
3268 if not m:
3269 return None
3270 num_str = m.group('num').replace(',', '.')
3271 mult = unit_table[m.group('unit')]
3272 return int(float(num_str) * mult)
3273
3274
3275 def parse_filesize(s):
3276 if s is None:
3277 return None
3278
3279 # The lower-case forms are of course incorrect and unofficial,
3280 # but we support those too
3281 _UNIT_TABLE = {
3282 'B': 1,
3283 'b': 1,
3284 'bytes': 1,
3285 'KiB': 1024,
3286 'KB': 1000,
3287 'kB': 1024,
3288 'Kb': 1000,
3289 'kb': 1000,
3290 'kilobytes': 1000,
3291 'kibibytes': 1024,
3292 'MiB': 1024 ** 2,
3293 'MB': 1000 ** 2,
3294 'mB': 1024 ** 2,
3295 'Mb': 1000 ** 2,
3296 'mb': 1000 ** 2,
3297 'megabytes': 1000 ** 2,
3298 'mebibytes': 1024 ** 2,
3299 'GiB': 1024 ** 3,
3300 'GB': 1000 ** 3,
3301 'gB': 1024 ** 3,
3302 'Gb': 1000 ** 3,
3303 'gb': 1000 ** 3,
3304 'gigabytes': 1000 ** 3,
3305 'gibibytes': 1024 ** 3,
3306 'TiB': 1024 ** 4,
3307 'TB': 1000 ** 4,
3308 'tB': 1024 ** 4,
3309 'Tb': 1000 ** 4,
3310 'tb': 1000 ** 4,
3311 'terabytes': 1000 ** 4,
3312 'tebibytes': 1024 ** 4,
3313 'PiB': 1024 ** 5,
3314 'PB': 1000 ** 5,
3315 'pB': 1024 ** 5,
3316 'Pb': 1000 ** 5,
3317 'pb': 1000 ** 5,
3318 'petabytes': 1000 ** 5,
3319 'pebibytes': 1024 ** 5,
3320 'EiB': 1024 ** 6,
3321 'EB': 1000 ** 6,
3322 'eB': 1024 ** 6,
3323 'Eb': 1000 ** 6,
3324 'eb': 1000 ** 6,
3325 'exabytes': 1000 ** 6,
3326 'exbibytes': 1024 ** 6,
3327 'ZiB': 1024 ** 7,
3328 'ZB': 1000 ** 7,
3329 'zB': 1024 ** 7,
3330 'Zb': 1000 ** 7,
3331 'zb': 1000 ** 7,
3332 'zettabytes': 1000 ** 7,
3333 'zebibytes': 1024 ** 7,
3334 'YiB': 1024 ** 8,
3335 'YB': 1000 ** 8,
3336 'yB': 1024 ** 8,
3337 'Yb': 1000 ** 8,
3338 'yb': 1000 ** 8,
3339 'yottabytes': 1000 ** 8,
3340 'yobibytes': 1024 ** 8,
3341 }
3342
3343 return lookup_unit_table(_UNIT_TABLE, s)
3344
3345
3346 def parse_count(s):
3347 if s is None:
3348 return None
3349
3350 s = s.strip()
3351
3352 if re.match(r'^[\d,.]+$', s):
3353 return str_to_int(s)
3354
3355 _UNIT_TABLE = {
3356 'k': 1000,
3357 'K': 1000,
3358 'm': 1000 ** 2,
3359 'M': 1000 ** 2,
3360 'kk': 1000 ** 2,
3361 'KK': 1000 ** 2,
3362 }
3363
3364 return lookup_unit_table(_UNIT_TABLE, s)
3365
3366
3367 def parse_resolution(s):
3368 if s is None:
3369 return {}
3370
3371 mobj = re.search(r'\b(?P<w>\d+)\s*[xXƗ]\s*(?P<h>\d+)\b', s)
3372 if mobj:
3373 return {
3374 'width': int(mobj.group('w')),
3375 'height': int(mobj.group('h')),
3376 }
3377
3378 mobj = re.search(r'\b(\d+)[pPiI]\b', s)
3379 if mobj:
3380 return {'height': int(mobj.group(1))}
3381
3382 mobj = re.search(r'\b([48])[kK]\b', s)
3383 if mobj:
3384 return {'height': int(mobj.group(1)) * 540}
3385
3386 return {}
3387
3388
3389 def parse_bitrate(s):
3390 if not isinstance(s, compat_str):
3391 return
3392 mobj = re.search(r'\b(\d+)\s*kbps', s)
3393 if mobj:
3394 return int(mobj.group(1))
3395
3396
3397 def month_by_name(name, lang='en'):
3398 """ Return the number of a month by (locale-independently) English name """
3399
3400 month_names = MONTH_NAMES.get(lang, MONTH_NAMES['en'])
3401
3402 try:
3403 return month_names.index(name) + 1
3404 except ValueError:
3405 return None
3406
3407
3408 def month_by_abbreviation(abbrev):
3409 """ Return the number of a month by (locale-independently) English
3410 abbreviations """
3411
3412 try:
3413 return [s[:3] for s in ENGLISH_MONTH_NAMES].index(abbrev) + 1
3414 except ValueError:
3415 return None
3416
3417
3418 def fix_xml_ampersands(xml_str):
3419 """Replace all the '&' by '&amp;' in XML"""
3420 return re.sub(
3421 r'&(?!amp;|lt;|gt;|apos;|quot;|#x[0-9a-fA-F]{,4};|#[0-9]{,4};)',
3422 '&amp;',
3423 xml_str)
3424
3425
3426 def setproctitle(title):
3427 assert isinstance(title, compat_str)
3428
3429 # ctypes in Jython is not complete
3430 # http://bugs.jython.org/issue2148
3431 if sys.platform.startswith('java'):
3432 return
3433
3434 try:
3435 libc = ctypes.cdll.LoadLibrary('libc.so.6')
3436 except OSError:
3437 return
3438 except TypeError:
3439 # LoadLibrary in Windows Python 2.7.13 only expects
3440 # a bytestring, but since unicode_literals turns
3441 # every string into a unicode string, it fails.
3442 return
3443 title_bytes = title.encode('utf-8')
3444 buf = ctypes.create_string_buffer(len(title_bytes))
3445 buf.value = title_bytes
3446 try:
3447 libc.prctl(15, buf, 0, 0, 0)
3448 except AttributeError:
3449 return # Strange libc, just skip this
3450
3451
3452 def remove_start(s, start):
3453 return s[len(start):] if s is not None and s.startswith(start) else s
3454
3455
3456 def remove_end(s, end):
3457 return s[:-len(end)] if s is not None and s.endswith(end) else s
3458
3459
3460 def remove_quotes(s):
3461 if s is None or len(s) < 2:
3462 return s
3463 for quote in ('"', "'", ):
3464 if s[0] == quote and s[-1] == quote:
3465 return s[1:-1]
3466 return s
3467
3468
3469 def url_basename(url):
3470 path = compat_urlparse.urlparse(url).path
3471 return path.strip('/').split('/')[-1]
3472
3473
3474 def base_url(url):
3475 return re.match(r'https?://[^?#&]+/', url).group()
3476
3477
3478 def urljoin(base, path):
3479 if isinstance(path, bytes):
3480 path = path.decode('utf-8')
3481 if not isinstance(path, compat_str) or not path:
3482 return None
3483 if re.match(r'^(?:[a-zA-Z][a-zA-Z0-9+-.]*:)?//', path):
3484 return path
3485 if isinstance(base, bytes):
3486 base = base.decode('utf-8')
3487 if not isinstance(base, compat_str) or not re.match(
3488 r'^(?:https?:)?//', base):
3489 return None
3490 return compat_urlparse.urljoin(base, path)
3491
3492
3493 class HEADRequest(compat_urllib_request.Request):
3494 def get_method(self):
3495 return 'HEAD'
3496
3497
3498 class PUTRequest(compat_urllib_request.Request):
3499 def get_method(self):
3500 return 'PUT'
3501
3502
3503 def int_or_none(v, scale=1, default=None, get_attr=None, invscale=1):
3504 if get_attr:
3505 if v is not None:
3506 v = getattr(v, get_attr, None)
3507 if v == '':
3508 v = None
3509 if v is None:
3510 return default
3511 try:
3512 return int(v) * invscale // scale
3513 except (ValueError, TypeError):
3514 return default
3515
3516
3517 def str_or_none(v, default=None):
3518 return default if v is None else compat_str(v)
3519
3520
3521 def str_to_int(int_str):
3522 """ A more relaxed version of int_or_none """
3523 if isinstance(int_str, compat_integer_types):
3524 return int_str
3525 elif isinstance(int_str, compat_str):
3526 int_str = re.sub(r'[,\.\+]', '', int_str)
3527 return int_or_none(int_str)
3528
3529
3530 def float_or_none(v, scale=1, invscale=1, default=None):
3531 if v is None:
3532 return default
3533 try:
3534 return float(v) * invscale / scale
3535 except (ValueError, TypeError):
3536 return default
3537
3538
3539 def bool_or_none(v, default=None):
3540 return v if isinstance(v, bool) else default
3541
3542
3543 def strip_or_none(v, default=None):
3544 return v.strip() if isinstance(v, compat_str) else default
3545
3546
3547 def url_or_none(url):
3548 if not url or not isinstance(url, compat_str):
3549 return None
3550 url = url.strip()
3551 return url if re.match(r'^(?:[a-zA-Z][\da-zA-Z.+-]*:)?//', url) else None
3552
3553
3554 def parse_duration(s):
3555 if not isinstance(s, compat_basestring):
3556 return None
3557
3558 s = s.strip()
3559
3560 days, hours, mins, secs, ms = [None] * 5
3561 m = re.match(r'(?:(?:(?:(?P<days>[0-9]+):)?(?P<hours>[0-9]+):)?(?P<mins>[0-9]+):)?(?P<secs>[0-9]+)(?P<ms>\.[0-9]+)?Z?$', s)
3562 if m:
3563 days, hours, mins, secs, ms = m.groups()
3564 else:
3565 m = re.match(
3566 r'''(?ix)(?:P?
3567 (?:
3568 [0-9]+\s*y(?:ears?)?\s*
3569 )?
3570 (?:
3571 [0-9]+\s*m(?:onths?)?\s*
3572 )?
3573 (?:
3574 [0-9]+\s*w(?:eeks?)?\s*
3575 )?
3576 (?:
3577 (?P<days>[0-9]+)\s*d(?:ays?)?\s*
3578 )?
3579 T)?
3580 (?:
3581 (?P<hours>[0-9]+)\s*h(?:ours?)?\s*
3582 )?
3583 (?:
3584 (?P<mins>[0-9]+)\s*m(?:in(?:ute)?s?)?\s*
3585 )?
3586 (?:
3587 (?P<secs>[0-9]+)(?P<ms>\.[0-9]+)?\s*s(?:ec(?:ond)?s?)?\s*
3588 )?Z?$''', s)
3589 if m:
3590 days, hours, mins, secs, ms = m.groups()
3591 else:
3592 m = re.match(r'(?i)(?:(?P<hours>[0-9.]+)\s*(?:hours?)|(?P<mins>[0-9.]+)\s*(?:mins?\.?|minutes?)\s*)Z?$', s)
3593 if m:
3594 hours, mins = m.groups()
3595 else:
3596 return None
3597
3598 duration = 0
3599 if secs:
3600 duration += float(secs)
3601 if mins:
3602 duration += float(mins) * 60
3603 if hours:
3604 duration += float(hours) * 60 * 60
3605 if days:
3606 duration += float(days) * 24 * 60 * 60
3607 if ms:
3608 duration += float(ms)
3609 return duration
3610
3611
3612 def prepend_extension(filename, ext, expected_real_ext=None):
3613 name, real_ext = os.path.splitext(filename)
3614 return (
3615 '{0}.{1}{2}'.format(name, ext, real_ext)
3616 if not expected_real_ext or real_ext[1:] == expected_real_ext
3617 else '{0}.{1}'.format(filename, ext))
3618
3619
3620 def replace_extension(filename, ext, expected_real_ext=None):
3621 name, real_ext = os.path.splitext(filename)
3622 return '{0}.{1}'.format(
3623 name if not expected_real_ext or real_ext[1:] == expected_real_ext else filename,
3624 ext)
3625
3626
3627 def check_executable(exe, args=[]):
3628 """ Checks if the given binary is installed somewhere in PATH, and returns its name.
3629 args can be a list of arguments for a short output (like -version) """
3630 try:
3631 subprocess.Popen([exe] + args, stdout=subprocess.PIPE, stderr=subprocess.PIPE).communicate()
3632 except OSError:
3633 return False
3634 return exe
3635
3636
3637 def get_exe_version(exe, args=['--version'],
3638 version_re=None, unrecognized='present'):
3639 """ Returns the version of the specified executable,
3640 or False if the executable is not present """
3641 try:
3642 # STDIN should be redirected too. On UNIX-like systems, ffmpeg triggers
3643 # SIGTTOU if youtube-dl is run in the background.
3644 # See https://github.com/ytdl-org/youtube-dl/issues/955#issuecomment-209789656
3645 out, _ = subprocess.Popen(
3646 [encodeArgument(exe)] + args,
3647 stdin=subprocess.PIPE,
3648 stdout=subprocess.PIPE, stderr=subprocess.STDOUT).communicate()
3649 except OSError:
3650 return False
3651 if isinstance(out, bytes): # Python 2.x
3652 out = out.decode('ascii', 'ignore')
3653 return detect_exe_version(out, version_re, unrecognized)
3654
3655
3656 def detect_exe_version(output, version_re=None, unrecognized='present'):
3657 assert isinstance(output, compat_str)
3658 if version_re is None:
3659 version_re = r'version\s+([-0-9._a-zA-Z]+)'
3660 m = re.search(version_re, output)
3661 if m:
3662 return m.group(1)
3663 else:
3664 return unrecognized
3665
3666
3667 class PagedList(object):
3668 def __len__(self):
3669 # This is only useful for tests
3670 return len(self.getslice())
3671
3672
3673 class OnDemandPagedList(PagedList):
3674 def __init__(self, pagefunc, pagesize, use_cache=True):
3675 self._pagefunc = pagefunc
3676 self._pagesize = pagesize
3677 self._use_cache = use_cache
3678 if use_cache:
3679 self._cache = {}
3680
3681 def getslice(self, start=0, end=None):
3682 res = []
3683 for pagenum in itertools.count(start // self._pagesize):
3684 firstid = pagenum * self._pagesize
3685 nextfirstid = pagenum * self._pagesize + self._pagesize
3686 if start >= nextfirstid:
3687 continue
3688
3689 page_results = None
3690 if self._use_cache:
3691 page_results = self._cache.get(pagenum)
3692 if page_results is None:
3693 page_results = list(self._pagefunc(pagenum))
3694 if self._use_cache:
3695 self._cache[pagenum] = page_results
3696
3697 startv = (
3698 start % self._pagesize
3699 if firstid <= start < nextfirstid
3700 else 0)
3701
3702 endv = (
3703 ((end - 1) % self._pagesize) + 1
3704 if (end is not None and firstid <= end <= nextfirstid)
3705 else None)
3706
3707 if startv != 0 or endv is not None:
3708 page_results = page_results[startv:endv]
3709 res.extend(page_results)
3710
3711 # A little optimization - if current page is not "full", ie. does
3712 # not contain page_size videos then we can assume that this page
3713 # is the last one - there are no more ids on further pages -
3714 # i.e. no need to query again.
3715 if len(page_results) + startv < self._pagesize:
3716 break
3717
3718 # If we got the whole page, but the next page is not interesting,
3719 # break out early as well
3720 if end == nextfirstid:
3721 break
3722 return res
3723
3724
3725 class InAdvancePagedList(PagedList):
3726 def __init__(self, pagefunc, pagecount, pagesize):
3727 self._pagefunc = pagefunc
3728 self._pagecount = pagecount
3729 self._pagesize = pagesize
3730
3731 def getslice(self, start=0, end=None):
3732 res = []
3733 start_page = start // self._pagesize
3734 end_page = (
3735 self._pagecount if end is None else (end // self._pagesize + 1))
3736 skip_elems = start - start_page * self._pagesize
3737 only_more = None if end is None else end - start
3738 for pagenum in range(start_page, end_page):
3739 page = list(self._pagefunc(pagenum))
3740 if skip_elems:
3741 page = page[skip_elems:]
3742 skip_elems = None
3743 if only_more is not None:
3744 if len(page) < only_more:
3745 only_more -= len(page)
3746 else:
3747 page = page[:only_more]
3748 res.extend(page)
3749 break
3750 res.extend(page)
3751 return res
3752
3753
3754 def uppercase_escape(s):
3755 unicode_escape = codecs.getdecoder('unicode_escape')
3756 return re.sub(
3757 r'\\U[0-9a-fA-F]{8}',
3758 lambda m: unicode_escape(m.group(0))[0],
3759 s)
3760
3761
3762 def lowercase_escape(s):
3763 unicode_escape = codecs.getdecoder('unicode_escape')
3764 return re.sub(
3765 r'\\u[0-9a-fA-F]{4}',
3766 lambda m: unicode_escape(m.group(0))[0],
3767 s)
3768
3769
3770 def escape_rfc3986(s):
3771 """Escape non-ASCII characters as suggested by RFC 3986"""
3772 if sys.version_info < (3, 0) and isinstance(s, compat_str):
3773 s = s.encode('utf-8')
3774 return compat_urllib_parse.quote(s, b"%/;:@&=+$,!~*'()?#[]")
3775
3776
3777 def escape_url(url):
3778 """Escape URL as suggested by RFC 3986"""
3779 url_parsed = compat_urllib_parse_urlparse(url)
3780 return url_parsed._replace(
3781 netloc=url_parsed.netloc.encode('idna').decode('ascii'),
3782 path=escape_rfc3986(url_parsed.path),
3783 params=escape_rfc3986(url_parsed.params),
3784 query=escape_rfc3986(url_parsed.query),
3785 fragment=escape_rfc3986(url_parsed.fragment)
3786 ).geturl()
3787
3788
3789 def read_batch_urls(batch_fd):
3790 def fixup(url):
3791 if not isinstance(url, compat_str):
3792 url = url.decode('utf-8', 'replace')
3793 BOM_UTF8 = '\xef\xbb\xbf'
3794 if url.startswith(BOM_UTF8):
3795 url = url[len(BOM_UTF8):]
3796 url = url.strip()
3797 if url.startswith(('#', ';', ']')):
3798 return False
3799 return url
3800
3801 with contextlib.closing(batch_fd) as fd:
3802 return [url for url in map(fixup, fd) if url]
3803
3804
3805 def urlencode_postdata(*args, **kargs):
3806 return compat_urllib_parse_urlencode(*args, **kargs).encode('ascii')
3807
3808
3809 def update_url_query(url, query):
3810 if not query:
3811 return url
3812 parsed_url = compat_urlparse.urlparse(url)
3813 qs = compat_parse_qs(parsed_url.query)
3814 qs.update(query)
3815 return compat_urlparse.urlunparse(parsed_url._replace(
3816 query=compat_urllib_parse_urlencode(qs, True)))
3817
3818
3819 def update_Request(req, url=None, data=None, headers={}, query={}):
3820 req_headers = req.headers.copy()
3821 req_headers.update(headers)
3822 req_data = data or req.data
3823 req_url = update_url_query(url or req.get_full_url(), query)
3824 req_get_method = req.get_method()
3825 if req_get_method == 'HEAD':
3826 req_type = HEADRequest
3827 elif req_get_method == 'PUT':
3828 req_type = PUTRequest
3829 else:
3830 req_type = compat_urllib_request.Request
3831 new_req = req_type(
3832 req_url, data=req_data, headers=req_headers,
3833 origin_req_host=req.origin_req_host, unverifiable=req.unverifiable)
3834 if hasattr(req, 'timeout'):
3835 new_req.timeout = req.timeout
3836 return new_req
3837
3838
3839 def _multipart_encode_impl(data, boundary):
3840 content_type = 'multipart/form-data; boundary=%s' % boundary
3841
3842 out = b''
3843 for k, v in data.items():
3844 out += b'--' + boundary.encode('ascii') + b'\r\n'
3845 if isinstance(k, compat_str):
3846 k = k.encode('utf-8')
3847 if isinstance(v, compat_str):
3848 v = v.encode('utf-8')
3849 # RFC 2047 requires non-ASCII field names to be encoded, while RFC 7578
3850 # suggests sending UTF-8 directly. Firefox sends UTF-8, too
3851 content = b'Content-Disposition: form-data; name="' + k + b'"\r\n\r\n' + v + b'\r\n'
3852 if boundary.encode('ascii') in content:
3853 raise ValueError('Boundary overlaps with data')
3854 out += content
3855
3856 out += b'--' + boundary.encode('ascii') + b'--\r\n'
3857
3858 return out, content_type
3859
3860
3861 def multipart_encode(data, boundary=None):
3862 '''
3863 Encode a dict to RFC 7578-compliant form-data
3864
3865 data:
3866 A dict where keys and values can be either Unicode or bytes-like
3867 objects.
3868 boundary:
3869 If specified a Unicode object, it's used as the boundary. Otherwise
3870 a random boundary is generated.
3871
3872 Reference: https://tools.ietf.org/html/rfc7578
3873 '''
3874 has_specified_boundary = boundary is not None
3875
3876 while True:
3877 if boundary is None:
3878 boundary = '---------------' + str(random.randrange(0x0fffffff, 0xffffffff))
3879
3880 try:
3881 out, content_type = _multipart_encode_impl(data, boundary)
3882 break
3883 except ValueError:
3884 if has_specified_boundary:
3885 raise
3886 boundary = None
3887
3888 return out, content_type
3889
3890
3891 def dict_get(d, key_or_keys, default=None, skip_false_values=True):
3892 if isinstance(key_or_keys, (list, tuple)):
3893 for key in key_or_keys:
3894 if key not in d or d[key] is None or skip_false_values and not d[key]:
3895 continue
3896 return d[key]
3897 return default
3898 return d.get(key_or_keys, default)
3899
3900
3901 def try_get(src, getter, expected_type=None):
3902 if not isinstance(getter, (list, tuple)):
3903 getter = [getter]
3904 for get in getter:
3905 try:
3906 v = get(src)
3907 except (AttributeError, KeyError, TypeError, IndexError):
3908 pass
3909 else:
3910 if expected_type is None or isinstance(v, expected_type):
3911 return v
3912
3913
3914 def merge_dicts(*dicts):
3915 merged = {}
3916 for a_dict in dicts:
3917 for k, v in a_dict.items():
3918 if v is None:
3919 continue
3920 if (k not in merged
3921 or (isinstance(v, compat_str) and v
3922 and isinstance(merged[k], compat_str)
3923 and not merged[k])):
3924 merged[k] = v
3925 return merged
3926
3927
3928 def encode_compat_str(string, encoding=preferredencoding(), errors='strict'):
3929 return string if isinstance(string, compat_str) else compat_str(string, encoding, errors)
3930
3931
3932 US_RATINGS = {
3933 'G': 0,
3934 'PG': 10,
3935 'PG-13': 13,
3936 'R': 16,
3937 'NC': 18,
3938 }
3939
3940
3941 TV_PARENTAL_GUIDELINES = {
3942 'TV-Y': 0,
3943 'TV-Y7': 7,
3944 'TV-G': 0,
3945 'TV-PG': 0,
3946 'TV-14': 14,
3947 'TV-MA': 17,
3948 }
3949
3950
3951 def parse_age_limit(s):
3952 if type(s) == int:
3953 return s if 0 <= s <= 21 else None
3954 if not isinstance(s, compat_basestring):
3955 return None
3956 m = re.match(r'^(?P<age>\d{1,2})\+?$', s)
3957 if m:
3958 return int(m.group('age'))
3959 if s in US_RATINGS:
3960 return US_RATINGS[s]
3961 m = re.match(r'^TV[_-]?(%s)$' % '|'.join(k[3:] for k in TV_PARENTAL_GUIDELINES), s)
3962 if m:
3963 return TV_PARENTAL_GUIDELINES['TV-' + m.group(1)]
3964 return None
3965
3966
3967 def strip_jsonp(code):
3968 return re.sub(
3969 r'''(?sx)^
3970 (?:window\.)?(?P<func_name>[a-zA-Z0-9_.$]*)
3971 (?:\s*&&\s*(?P=func_name))?
3972 \s*\(\s*(?P<callback_data>.*)\);?
3973 \s*?(?://[^\n]*)*$''',
3974 r'\g<callback_data>', code)
3975
3976
3977 def js_to_json(code):
3978 COMMENT_RE = r'/\*(?:(?!\*/).)*?\*/|//[^\n]*'
3979 SKIP_RE = r'\s*(?:{comment})?\s*'.format(comment=COMMENT_RE)
3980 INTEGER_TABLE = (
3981 (r'(?s)^(0[xX][0-9a-fA-F]+){skip}:?$'.format(skip=SKIP_RE), 16),
3982 (r'(?s)^(0+[0-7]+){skip}:?$'.format(skip=SKIP_RE), 8),
3983 )
3984
3985 def fix_kv(m):
3986 v = m.group(0)
3987 if v in ('true', 'false', 'null'):
3988 return v
3989 elif v.startswith('/*') or v.startswith('//') or v == ',':
3990 return ""
3991
3992 if v[0] in ("'", '"'):
3993 v = re.sub(r'(?s)\\.|"', lambda m: {
3994 '"': '\\"',
3995 "\\'": "'",
3996 '\\\n': '',
3997 '\\x': '\\u00',
3998 }.get(m.group(0), m.group(0)), v[1:-1])
3999
4000 for regex, base in INTEGER_TABLE:
4001 im = re.match(regex, v)
4002 if im:
4003 i = int(im.group(1), base)
4004 return '"%d":' % i if v.endswith(':') else '%d' % i
4005
4006 return '"%s"' % v
4007
4008 return re.sub(r'''(?sx)
4009 "(?:[^"\\]*(?:\\\\|\\['"nurtbfx/\n]))*[^"\\]*"|
4010 '(?:[^'\\]*(?:\\\\|\\['"nurtbfx/\n]))*[^'\\]*'|
4011 {comment}|,(?={skip}[\]}}])|
4012 (?:(?<![0-9])[eE]|[a-df-zA-DF-Z_])[.a-zA-Z_0-9]*|
4013 \b(?:0[xX][0-9a-fA-F]+|0+[0-7]+)(?:{skip}:)?|
4014 [0-9]+(?={skip}:)
4015 '''.format(comment=COMMENT_RE, skip=SKIP_RE), fix_kv, code)
4016
4017
4018 def qualities(quality_ids):
4019 """ Get a numeric quality value out of a list of possible values """
4020 def q(qid):
4021 try:
4022 return quality_ids.index(qid)
4023 except ValueError:
4024 return -1
4025 return q
4026
4027
4028 DEFAULT_OUTTMPL = '%(title)s-%(id)s.%(ext)s'
4029
4030
4031 def limit_length(s, length):
4032 """ Add ellipses to overly long strings """
4033 if s is None:
4034 return None
4035 ELLIPSES = '...'
4036 if len(s) > length:
4037 return s[:length - len(ELLIPSES)] + ELLIPSES
4038 return s
4039
4040
4041 def version_tuple(v):
4042 return tuple(int(e) for e in re.split(r'[-.]', v))
4043
4044
4045 def is_outdated_version(version, limit, assume_new=True):
4046 if not version:
4047 return not assume_new
4048 try:
4049 return version_tuple(version) < version_tuple(limit)
4050 except ValueError:
4051 return not assume_new
4052
4053
4054 def ytdl_is_updateable():
4055 """ Returns if youtube-dl can be updated with -U """
4056 from zipimport import zipimporter
4057
4058 return isinstance(globals().get('__loader__'), zipimporter) or hasattr(sys, 'frozen')
4059
4060
4061 def args_to_str(args):
4062 # Get a short string representation for a subprocess command
4063 return ' '.join(compat_shlex_quote(a) for a in args)
4064
4065
4066 def error_to_compat_str(err):
4067 err_str = str(err)
4068 # On python 2 error byte string must be decoded with proper
4069 # encoding rather than ascii
4070 if sys.version_info[0] < 3:
4071 err_str = err_str.decode(preferredencoding())
4072 return err_str
4073
4074
4075 def mimetype2ext(mt):
4076 if mt is None:
4077 return None
4078
4079 ext = {
4080 'audio/mp4': 'm4a',
4081 # Per RFC 3003, audio/mpeg can be .mp1, .mp2 or .mp3. Here use .mp3 as
4082 # it's the most popular one
4083 'audio/mpeg': 'mp3',
4084 }.get(mt)
4085 if ext is not None:
4086 return ext
4087
4088 _, _, res = mt.rpartition('/')
4089 res = res.split(';')[0].strip().lower()
4090
4091 return {
4092 '3gpp': '3gp',
4093 'smptett+xml': 'tt',
4094 'ttaf+xml': 'dfxp',
4095 'ttml+xml': 'ttml',
4096 'x-flv': 'flv',
4097 'x-mp4-fragmented': 'mp4',
4098 'x-ms-sami': 'sami',
4099 'x-ms-wmv': 'wmv',
4100 'mpegurl': 'm3u8',
4101 'x-mpegurl': 'm3u8',
4102 'vnd.apple.mpegurl': 'm3u8',
4103 'dash+xml': 'mpd',
4104 'f4m+xml': 'f4m',
4105 'hds+xml': 'f4m',
4106 'vnd.ms-sstr+xml': 'ism',
4107 'quicktime': 'mov',
4108 'mp2t': 'ts',
4109 }.get(res, res)
4110
4111
4112 def parse_codecs(codecs_str):
4113 # http://tools.ietf.org/html/rfc6381
4114 if not codecs_str:
4115 return {}
4116 splited_codecs = list(filter(None, map(
4117 lambda str: str.strip(), codecs_str.strip().strip(',').split(','))))
4118 vcodec, acodec = None, None
4119 for full_codec in splited_codecs:
4120 codec = full_codec.split('.')[0]
4121 if codec in ('avc1', 'avc2', 'avc3', 'avc4', 'vp9', 'vp8', 'hev1', 'hev2', 'h263', 'h264', 'mp4v', 'hvc1', 'av01', 'theora'):
4122 if not vcodec:
4123 vcodec = full_codec
4124 elif codec in ('mp4a', 'opus', 'vorbis', 'mp3', 'aac', 'ac-3', 'ec-3', 'eac3', 'dtsc', 'dtse', 'dtsh', 'dtsl'):
4125 if not acodec:
4126 acodec = full_codec
4127 else:
4128 write_string('WARNING: Unknown codec %s\n' % full_codec, sys.stderr)
4129 if not vcodec and not acodec:
4130 if len(splited_codecs) == 2:
4131 return {
4132 'vcodec': splited_codecs[0],
4133 'acodec': splited_codecs[1],
4134 }
4135 else:
4136 return {
4137 'vcodec': vcodec or 'none',
4138 'acodec': acodec or 'none',
4139 }
4140 return {}
4141
4142
4143 def urlhandle_detect_ext(url_handle):
4144 getheader = url_handle.headers.get
4145
4146 cd = getheader('Content-Disposition')
4147 if cd:
4148 m = re.match(r'attachment;\s*filename="(?P<filename>[^"]+)"', cd)
4149 if m:
4150 e = determine_ext(m.group('filename'), default_ext=None)
4151 if e:
4152 return e
4153
4154 return mimetype2ext(getheader('Content-Type'))
4155
4156
4157 def encode_data_uri(data, mime_type):
4158 return 'data:%s;base64,%s' % (mime_type, base64.b64encode(data).decode('ascii'))
4159
4160
4161 def age_restricted(content_limit, age_limit):
4162 """ Returns True iff the content should be blocked """
4163
4164 if age_limit is None: # No limit set
4165 return False
4166 if content_limit is None:
4167 return False # Content available for everyone
4168 return age_limit < content_limit
4169
4170
4171 def is_html(first_bytes):
4172 """ Detect whether a file contains HTML by examining its first bytes. """
4173
4174 BOMS = [
4175 (b'\xef\xbb\xbf', 'utf-8'),
4176 (b'\x00\x00\xfe\xff', 'utf-32-be'),
4177 (b'\xff\xfe\x00\x00', 'utf-32-le'),
4178 (b'\xff\xfe', 'utf-16-le'),
4179 (b'\xfe\xff', 'utf-16-be'),
4180 ]
4181 for bom, enc in BOMS:
4182 if first_bytes.startswith(bom):
4183 s = first_bytes[len(bom):].decode(enc, 'replace')
4184 break
4185 else:
4186 s = first_bytes.decode('utf-8', 'replace')
4187
4188 return re.match(r'^\s*<', s)
4189
4190
4191 def determine_protocol(info_dict):
4192 protocol = info_dict.get('protocol')
4193 if protocol is not None:
4194 return protocol
4195
4196 url = info_dict['url']
4197 if url.startswith('rtmp'):
4198 return 'rtmp'
4199 elif url.startswith('mms'):
4200 return 'mms'
4201 elif url.startswith('rtsp'):
4202 return 'rtsp'
4203
4204 ext = determine_ext(url)
4205 if ext == 'm3u8':
4206 return 'm3u8'
4207 elif ext == 'f4m':
4208 return 'f4m'
4209
4210 return compat_urllib_parse_urlparse(url).scheme
4211
4212
4213 def render_table(header_row, data):
4214 """ Render a list of rows, each as a list of values """
4215 table = [header_row] + data
4216 max_lens = [max(len(compat_str(v)) for v in col) for col in zip(*table)]
4217 format_str = ' '.join('%-' + compat_str(ml + 1) + 's' for ml in max_lens[:-1]) + '%s'
4218 return '\n'.join(format_str % tuple(row) for row in table)
4219
4220
4221 def _match_one(filter_part, dct):
4222 COMPARISON_OPERATORS = {
4223 '<': operator.lt,
4224 '<=': operator.le,
4225 '>': operator.gt,
4226 '>=': operator.ge,
4227 '=': operator.eq,
4228 '!=': operator.ne,
4229 }
4230 operator_rex = re.compile(r'''(?x)\s*
4231 (?P<key>[a-z_]+)
4232 \s*(?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
4233 (?:
4234 (?P<intval>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)|
4235 (?P<quote>["\'])(?P<quotedstrval>(?:\\.|(?!(?P=quote)|\\).)+?)(?P=quote)|
4236 (?P<strval>(?![0-9.])[a-z0-9A-Z]*)
4237 )
4238 \s*$
4239 ''' % '|'.join(map(re.escape, COMPARISON_OPERATORS.keys())))
4240 m = operator_rex.search(filter_part)
4241 if m:
4242 op = COMPARISON_OPERATORS[m.group('op')]
4243 actual_value = dct.get(m.group('key'))
4244 if (m.group('quotedstrval') is not None
4245 or m.group('strval') is not None
4246 # If the original field is a string and matching comparisonvalue is
4247 # a number we should respect the origin of the original field
4248 # and process comparison value as a string (see
4249 # https://github.com/ytdl-org/youtube-dl/issues/11082).
4250 or actual_value is not None and m.group('intval') is not None
4251 and isinstance(actual_value, compat_str)):
4252 if m.group('op') not in ('=', '!='):
4253 raise ValueError(
4254 'Operator %s does not support string values!' % m.group('op'))
4255 comparison_value = m.group('quotedstrval') or m.group('strval') or m.group('intval')
4256 quote = m.group('quote')
4257 if quote is not None:
4258 comparison_value = comparison_value.replace(r'\%s' % quote, quote)
4259 else:
4260 try:
4261 comparison_value = int(m.group('intval'))
4262 except ValueError:
4263 comparison_value = parse_filesize(m.group('intval'))
4264 if comparison_value is None:
4265 comparison_value = parse_filesize(m.group('intval') + 'B')
4266 if comparison_value is None:
4267 raise ValueError(
4268 'Invalid integer value %r in filter part %r' % (
4269 m.group('intval'), filter_part))
4270 if actual_value is None:
4271 return m.group('none_inclusive')
4272 return op(actual_value, comparison_value)
4273
4274 UNARY_OPERATORS = {
4275 '': lambda v: (v is True) if isinstance(v, bool) else (v is not None),
4276 '!': lambda v: (v is False) if isinstance(v, bool) else (v is None),
4277 }
4278 operator_rex = re.compile(r'''(?x)\s*
4279 (?P<op>%s)\s*(?P<key>[a-z_]+)
4280 \s*$
4281 ''' % '|'.join(map(re.escape, UNARY_OPERATORS.keys())))
4282 m = operator_rex.search(filter_part)
4283 if m:
4284 op = UNARY_OPERATORS[m.group('op')]
4285 actual_value = dct.get(m.group('key'))
4286 return op(actual_value)
4287
4288 raise ValueError('Invalid filter part %r' % filter_part)
4289
4290
4291 def match_str(filter_str, dct):
4292 """ Filter a dictionary with a simple string syntax. Returns True (=passes filter) or false """
4293
4294 return all(
4295 _match_one(filter_part, dct) for filter_part in filter_str.split('&'))
4296
4297
4298 def match_filter_func(filter_str):
4299 def _match_func(info_dict):
4300 if match_str(filter_str, info_dict):
4301 return None
4302 else:
4303 video_title = info_dict.get('title', info_dict.get('id', 'video'))
4304 return '%s does not pass filter %s, skipping ..' % (video_title, filter_str)
4305 return _match_func
4306
4307
4308 def parse_dfxp_time_expr(time_expr):
4309 if not time_expr:
4310 return
4311
4312 mobj = re.match(r'^(?P<time_offset>\d+(?:\.\d+)?)s?$', time_expr)
4313 if mobj:
4314 return float(mobj.group('time_offset'))
4315
4316 mobj = re.match(r'^(\d+):(\d\d):(\d\d(?:(?:\.|:)\d+)?)$', time_expr)
4317 if mobj:
4318 return 3600 * int(mobj.group(1)) + 60 * int(mobj.group(2)) + float(mobj.group(3).replace(':', '.'))
4319
4320
4321 def srt_subtitles_timecode(seconds):
4322 return '%02d:%02d:%02d,%03d' % (seconds / 3600, (seconds % 3600) / 60, seconds % 60, (seconds % 1) * 1000)
4323
4324
4325 def dfxp2srt(dfxp_data):
4326 '''
4327 @param dfxp_data A bytes-like object containing DFXP data
4328 @returns A unicode object containing converted SRT data
4329 '''
4330 LEGACY_NAMESPACES = (
4331 (b'http://www.w3.org/ns/ttml', [
4332 b'http://www.w3.org/2004/11/ttaf1',
4333 b'http://www.w3.org/2006/04/ttaf1',
4334 b'http://www.w3.org/2006/10/ttaf1',
4335 ]),
4336 (b'http://www.w3.org/ns/ttml#styling', [
4337 b'http://www.w3.org/ns/ttml#style',
4338 ]),
4339 )
4340
4341 SUPPORTED_STYLING = [
4342 'color',
4343 'fontFamily',
4344 'fontSize',
4345 'fontStyle',
4346 'fontWeight',
4347 'textDecoration'
4348 ]
4349
4350 _x = functools.partial(xpath_with_ns, ns_map={
4351 'xml': 'http://www.w3.org/XML/1998/namespace',
4352 'ttml': 'http://www.w3.org/ns/ttml',
4353 'tts': 'http://www.w3.org/ns/ttml#styling',
4354 })
4355
4356 styles = {}
4357 default_style = {}
4358
4359 class TTMLPElementParser(object):
4360 _out = ''
4361 _unclosed_elements = []
4362 _applied_styles = []
4363
4364 def start(self, tag, attrib):
4365 if tag in (_x('ttml:br'), 'br'):
4366 self._out += '\n'
4367 else:
4368 unclosed_elements = []
4369 style = {}
4370 element_style_id = attrib.get('style')
4371 if default_style:
4372 style.update(default_style)
4373 if element_style_id:
4374 style.update(styles.get(element_style_id, {}))
4375 for prop in SUPPORTED_STYLING:
4376 prop_val = attrib.get(_x('tts:' + prop))
4377 if prop_val:
4378 style[prop] = prop_val
4379 if style:
4380 font = ''
4381 for k, v in sorted(style.items()):
4382 if self._applied_styles and self._applied_styles[-1].get(k) == v:
4383 continue
4384 if k == 'color':
4385 font += ' color="%s"' % v
4386 elif k == 'fontSize':
4387 font += ' size="%s"' % v
4388 elif k == 'fontFamily':
4389 font += ' face="%s"' % v
4390 elif k == 'fontWeight' and v == 'bold':
4391 self._out += '<b>'
4392 unclosed_elements.append('b')
4393 elif k == 'fontStyle' and v == 'italic':
4394 self._out += '<i>'
4395 unclosed_elements.append('i')
4396 elif k == 'textDecoration' and v == 'underline':
4397 self._out += '<u>'
4398 unclosed_elements.append('u')
4399 if font:
4400 self._out += '<font' + font + '>'
4401 unclosed_elements.append('font')
4402 applied_style = {}
4403 if self._applied_styles:
4404 applied_style.update(self._applied_styles[-1])
4405 applied_style.update(style)
4406 self._applied_styles.append(applied_style)
4407 self._unclosed_elements.append(unclosed_elements)
4408
4409 def end(self, tag):
4410 if tag not in (_x('ttml:br'), 'br'):
4411 unclosed_elements = self._unclosed_elements.pop()
4412 for element in reversed(unclosed_elements):
4413 self._out += '</%s>' % element
4414 if unclosed_elements and self._applied_styles:
4415 self._applied_styles.pop()
4416
4417 def data(self, data):
4418 self._out += data
4419
4420 def close(self):
4421 return self._out.strip()
4422
4423 def parse_node(node):
4424 target = TTMLPElementParser()
4425 parser = xml.etree.ElementTree.XMLParser(target=target)
4426 parser.feed(xml.etree.ElementTree.tostring(node))
4427 return parser.close()
4428
4429 for k, v in LEGACY_NAMESPACES:
4430 for ns in v:
4431 dfxp_data = dfxp_data.replace(ns, k)
4432
4433 dfxp = compat_etree_fromstring(dfxp_data)
4434 out = []
4435 paras = dfxp.findall(_x('.//ttml:p')) or dfxp.findall('.//p')
4436
4437 if not paras:
4438 raise ValueError('Invalid dfxp/TTML subtitle')
4439
4440 repeat = False
4441 while True:
4442 for style in dfxp.findall(_x('.//ttml:style')):
4443 style_id = style.get('id') or style.get(_x('xml:id'))
4444 if not style_id:
4445 continue
4446 parent_style_id = style.get('style')
4447 if parent_style_id:
4448 if parent_style_id not in styles:
4449 repeat = True
4450 continue
4451 styles[style_id] = styles[parent_style_id].copy()
4452 for prop in SUPPORTED_STYLING:
4453 prop_val = style.get(_x('tts:' + prop))
4454 if prop_val:
4455 styles.setdefault(style_id, {})[prop] = prop_val
4456 if repeat:
4457 repeat = False
4458 else:
4459 break
4460
4461 for p in ('body', 'div'):
4462 ele = xpath_element(dfxp, [_x('.//ttml:' + p), './/' + p])
4463 if ele is None:
4464 continue
4465 style = styles.get(ele.get('style'))
4466 if not style:
4467 continue
4468 default_style.update(style)
4469
4470 for para, index in zip(paras, itertools.count(1)):
4471 begin_time = parse_dfxp_time_expr(para.attrib.get('begin'))
4472 end_time = parse_dfxp_time_expr(para.attrib.get('end'))
4473 dur = parse_dfxp_time_expr(para.attrib.get('dur'))
4474 if begin_time is None:
4475 continue
4476 if not end_time:
4477 if not dur:
4478 continue
4479 end_time = begin_time + dur
4480 out.append('%d\n%s --> %s\n%s\n\n' % (
4481 index,
4482 srt_subtitles_timecode(begin_time),
4483 srt_subtitles_timecode(end_time),
4484 parse_node(para)))
4485
4486 return ''.join(out)
4487
4488
4489 def cli_option(params, command_option, param):
4490 param = params.get(param)
4491 if param:
4492 param = compat_str(param)
4493 return [command_option, param] if param is not None else []
4494
4495
4496 def cli_bool_option(params, command_option, param, true_value='true', false_value='false', separator=None):
4497 param = params.get(param)
4498 if param is None:
4499 return []
4500 assert isinstance(param, bool)
4501 if separator:
4502 return [command_option + separator + (true_value if param else false_value)]
4503 return [command_option, true_value if param else false_value]
4504
4505
4506 def cli_valueless_option(params, command_option, param, expected_value=True):
4507 param = params.get(param)
4508 return [command_option] if param == expected_value else []
4509
4510
4511 def cli_configuration_args(params, param, default=[]):
4512 ex_args = params.get(param)
4513 if ex_args is None:
4514 return default
4515 assert isinstance(ex_args, list)
4516 return ex_args
4517
4518
4519 class ISO639Utils(object):
4520 # See http://www.loc.gov/standards/iso639-2/ISO-639-2_utf-8.txt
4521 _lang_map = {
4522 'aa': 'aar',
4523 'ab': 'abk',
4524 'ae': 'ave',
4525 'af': 'afr',
4526 'ak': 'aka',
4527 'am': 'amh',
4528 'an': 'arg',
4529 'ar': 'ara',
4530 'as': 'asm',
4531 'av': 'ava',
4532 'ay': 'aym',
4533 'az': 'aze',
4534 'ba': 'bak',
4535 'be': 'bel',
4536 'bg': 'bul',
4537 'bh': 'bih',
4538 'bi': 'bis',
4539 'bm': 'bam',
4540 'bn': 'ben',
4541 'bo': 'bod',
4542 'br': 'bre',
4543 'bs': 'bos',
4544 'ca': 'cat',
4545 'ce': 'che',
4546 'ch': 'cha',
4547 'co': 'cos',
4548 'cr': 'cre',
4549 'cs': 'ces',
4550 'cu': 'chu',
4551 'cv': 'chv',
4552 'cy': 'cym',
4553 'da': 'dan',
4554 'de': 'deu',
4555 'dv': 'div',
4556 'dz': 'dzo',
4557 'ee': 'ewe',
4558 'el': 'ell',
4559 'en': 'eng',
4560 'eo': 'epo',
4561 'es': 'spa',
4562 'et': 'est',
4563 'eu': 'eus',
4564 'fa': 'fas',
4565 'ff': 'ful',
4566 'fi': 'fin',
4567 'fj': 'fij',
4568 'fo': 'fao',
4569 'fr': 'fra',
4570 'fy': 'fry',
4571 'ga': 'gle',
4572 'gd': 'gla',
4573 'gl': 'glg',
4574 'gn': 'grn',
4575 'gu': 'guj',
4576 'gv': 'glv',
4577 'ha': 'hau',
4578 'he': 'heb',
4579 'iw': 'heb', # Replaced by he in 1989 revision
4580 'hi': 'hin',
4581 'ho': 'hmo',
4582 'hr': 'hrv',
4583 'ht': 'hat',
4584 'hu': 'hun',
4585 'hy': 'hye',
4586 'hz': 'her',
4587 'ia': 'ina',
4588 'id': 'ind',
4589 'in': 'ind', # Replaced by id in 1989 revision
4590 'ie': 'ile',
4591 'ig': 'ibo',
4592 'ii': 'iii',
4593 'ik': 'ipk',
4594 'io': 'ido',
4595 'is': 'isl',
4596 'it': 'ita',
4597 'iu': 'iku',
4598 'ja': 'jpn',
4599 'jv': 'jav',
4600 'ka': 'kat',
4601 'kg': 'kon',
4602 'ki': 'kik',
4603 'kj': 'kua',
4604 'kk': 'kaz',
4605 'kl': 'kal',
4606 'km': 'khm',
4607 'kn': 'kan',
4608 'ko': 'kor',
4609 'kr': 'kau',
4610 'ks': 'kas',
4611 'ku': 'kur',
4612 'kv': 'kom',
4613 'kw': 'cor',
4614 'ky': 'kir',
4615 'la': 'lat',
4616 'lb': 'ltz',
4617 'lg': 'lug',
4618 'li': 'lim',
4619 'ln': 'lin',
4620 'lo': 'lao',
4621 'lt': 'lit',
4622 'lu': 'lub',
4623 'lv': 'lav',
4624 'mg': 'mlg',
4625 'mh': 'mah',
4626 'mi': 'mri',
4627 'mk': 'mkd',
4628 'ml': 'mal',
4629 'mn': 'mon',
4630 'mr': 'mar',
4631 'ms': 'msa',
4632 'mt': 'mlt',
4633 'my': 'mya',
4634 'na': 'nau',
4635 'nb': 'nob',
4636 'nd': 'nde',
4637 'ne': 'nep',
4638 'ng': 'ndo',
4639 'nl': 'nld',
4640 'nn': 'nno',
4641 'no': 'nor',
4642 'nr': 'nbl',
4643 'nv': 'nav',
4644 'ny': 'nya',
4645 'oc': 'oci',
4646 'oj': 'oji',
4647 'om': 'orm',
4648 'or': 'ori',
4649 'os': 'oss',
4650 'pa': 'pan',
4651 'pi': 'pli',
4652 'pl': 'pol',
4653 'ps': 'pus',
4654 'pt': 'por',
4655 'qu': 'que',
4656 'rm': 'roh',
4657 'rn': 'run',
4658 'ro': 'ron',
4659 'ru': 'rus',
4660 'rw': 'kin',
4661 'sa': 'san',
4662 'sc': 'srd',
4663 'sd': 'snd',
4664 'se': 'sme',
4665 'sg': 'sag',
4666 'si': 'sin',
4667 'sk': 'slk',
4668 'sl': 'slv',
4669 'sm': 'smo',
4670 'sn': 'sna',
4671 'so': 'som',
4672 'sq': 'sqi',
4673 'sr': 'srp',
4674 'ss': 'ssw',
4675 'st': 'sot',
4676 'su': 'sun',
4677 'sv': 'swe',
4678 'sw': 'swa',
4679 'ta': 'tam',
4680 'te': 'tel',
4681 'tg': 'tgk',
4682 'th': 'tha',
4683 'ti': 'tir',
4684 'tk': 'tuk',
4685 'tl': 'tgl',
4686 'tn': 'tsn',
4687 'to': 'ton',
4688 'tr': 'tur',
4689 'ts': 'tso',
4690 'tt': 'tat',
4691 'tw': 'twi',
4692 'ty': 'tah',
4693 'ug': 'uig',
4694 'uk': 'ukr',
4695 'ur': 'urd',
4696 'uz': 'uzb',
4697 've': 'ven',
4698 'vi': 'vie',
4699 'vo': 'vol',
4700 'wa': 'wln',
4701 'wo': 'wol',
4702 'xh': 'xho',
4703 'yi': 'yid',
4704 'ji': 'yid', # Replaced by yi in 1989 revision
4705 'yo': 'yor',
4706 'za': 'zha',
4707 'zh': 'zho',
4708 'zu': 'zul',
4709 }
4710
4711 @classmethod
4712 def short2long(cls, code):
4713 """Convert language code from ISO 639-1 to ISO 639-2/T"""
4714 return cls._lang_map.get(code[:2])
4715
4716 @classmethod
4717 def long2short(cls, code):
4718 """Convert language code from ISO 639-2/T to ISO 639-1"""
4719 for short_name, long_name in cls._lang_map.items():
4720 if long_name == code:
4721 return short_name
4722
4723
4724 class ISO3166Utils(object):
4725 # From http://data.okfn.org/data/core/country-list
4726 _country_map = {
4727 'AF': 'Afghanistan',
4728 'AX': 'ƅland Islands',
4729 'AL': 'Albania',
4730 'DZ': 'Algeria',
4731 'AS': 'American Samoa',
4732 'AD': 'Andorra',
4733 'AO': 'Angola',
4734 'AI': 'Anguilla',
4735 'AQ': 'Antarctica',
4736 'AG': 'Antigua and Barbuda',
4737 'AR': 'Argentina',
4738 'AM': 'Armenia',
4739 'AW': 'Aruba',
4740 'AU': 'Australia',
4741 'AT': 'Austria',
4742 'AZ': 'Azerbaijan',
4743 'BS': 'Bahamas',
4744 'BH': 'Bahrain',
4745 'BD': 'Bangladesh',
4746 'BB': 'Barbados',
4747 'BY': 'Belarus',
4748 'BE': 'Belgium',
4749 'BZ': 'Belize',
4750 'BJ': 'Benin',
4751 'BM': 'Bermuda',
4752 'BT': 'Bhutan',
4753 'BO': 'Bolivia, Plurinational State of',
4754 'BQ': 'Bonaire, Sint Eustatius and Saba',
4755 'BA': 'Bosnia and Herzegovina',
4756 'BW': 'Botswana',
4757 'BV': 'Bouvet Island',
4758 'BR': 'Brazil',
4759 'IO': 'British Indian Ocean Territory',
4760 'BN': 'Brunei Darussalam',
4761 'BG': 'Bulgaria',
4762 'BF': 'Burkina Faso',
4763 'BI': 'Burundi',
4764 'KH': 'Cambodia',
4765 'CM': 'Cameroon',
4766 'CA': 'Canada',
4767 'CV': 'Cape Verde',
4768 'KY': 'Cayman Islands',
4769 'CF': 'Central African Republic',
4770 'TD': 'Chad',
4771 'CL': 'Chile',
4772 'CN': 'China',
4773 'CX': 'Christmas Island',
4774 'CC': 'Cocos (Keeling) Islands',
4775 'CO': 'Colombia',
4776 'KM': 'Comoros',
4777 'CG': 'Congo',
4778 'CD': 'Congo, the Democratic Republic of the',
4779 'CK': 'Cook Islands',
4780 'CR': 'Costa Rica',
4781 'CI': 'CĆ“te d\'Ivoire',
4782 'HR': 'Croatia',
4783 'CU': 'Cuba',
4784 'CW': 'CuraƧao',
4785 'CY': 'Cyprus',
4786 'CZ': 'Czech Republic',
4787 'DK': 'Denmark',
4788 'DJ': 'Djibouti',
4789 'DM': 'Dominica',
4790 'DO': 'Dominican Republic',
4791 'EC': 'Ecuador',
4792 'EG': 'Egypt',
4793 'SV': 'El Salvador',
4794 'GQ': 'Equatorial Guinea',
4795 'ER': 'Eritrea',
4796 'EE': 'Estonia',
4797 'ET': 'Ethiopia',
4798 'FK': 'Falkland Islands (Malvinas)',
4799 'FO': 'Faroe Islands',
4800 'FJ': 'Fiji',
4801 'FI': 'Finland',
4802 'FR': 'France',
4803 'GF': 'French Guiana',
4804 'PF': 'French Polynesia',
4805 'TF': 'French Southern Territories',
4806 'GA': 'Gabon',
4807 'GM': 'Gambia',
4808 'GE': 'Georgia',
4809 'DE': 'Germany',
4810 'GH': 'Ghana',
4811 'GI': 'Gibraltar',
4812 'GR': 'Greece',
4813 'GL': 'Greenland',
4814 'GD': 'Grenada',
4815 'GP': 'Guadeloupe',
4816 'GU': 'Guam',
4817 'GT': 'Guatemala',
4818 'GG': 'Guernsey',
4819 'GN': 'Guinea',
4820 'GW': 'Guinea-Bissau',
4821 'GY': 'Guyana',
4822 'HT': 'Haiti',
4823 'HM': 'Heard Island and McDonald Islands',
4824 'VA': 'Holy See (Vatican City State)',
4825 'HN': 'Honduras',
4826 'HK': 'Hong Kong',
4827 'HU': 'Hungary',
4828 'IS': 'Iceland',
4829 'IN': 'India',
4830 'ID': 'Indonesia',
4831 'IR': 'Iran, Islamic Republic of',
4832 'IQ': 'Iraq',
4833 'IE': 'Ireland',
4834 'IM': 'Isle of Man',
4835 'IL': 'Israel',
4836 'IT': 'Italy',
4837 'JM': 'Jamaica',
4838 'JP': 'Japan',
4839 'JE': 'Jersey',
4840 'JO': 'Jordan',
4841 'KZ': 'Kazakhstan',
4842 'KE': 'Kenya',
4843 'KI': 'Kiribati',
4844 'KP': 'Korea, Democratic People\'s Republic of',
4845 'KR': 'Korea, Republic of',
4846 'KW': 'Kuwait',
4847 'KG': 'Kyrgyzstan',
4848 'LA': 'Lao People\'s Democratic Republic',
4849 'LV': 'Latvia',
4850 'LB': 'Lebanon',
4851 'LS': 'Lesotho',
4852 'LR': 'Liberia',
4853 'LY': 'Libya',
4854 'LI': 'Liechtenstein',
4855 'LT': 'Lithuania',
4856 'LU': 'Luxembourg',
4857 'MO': 'Macao',
4858 'MK': 'Macedonia, the Former Yugoslav Republic of',
4859 'MG': 'Madagascar',
4860 'MW': 'Malawi',
4861 'MY': 'Malaysia',
4862 'MV': 'Maldives',
4863 'ML': 'Mali',
4864 'MT': 'Malta',
4865 'MH': 'Marshall Islands',
4866 'MQ': 'Martinique',
4867 'MR': 'Mauritania',
4868 'MU': 'Mauritius',
4869 'YT': 'Mayotte',
4870 'MX': 'Mexico',
4871 'FM': 'Micronesia, Federated States of',
4872 'MD': 'Moldova, Republic of',
4873 'MC': 'Monaco',
4874 'MN': 'Mongolia',
4875 'ME': 'Montenegro',
4876 'MS': 'Montserrat',
4877 'MA': 'Morocco',
4878 'MZ': 'Mozambique',
4879 'MM': 'Myanmar',
4880 'NA': 'Namibia',
4881 'NR': 'Nauru',
4882 'NP': 'Nepal',
4883 'NL': 'Netherlands',
4884 'NC': 'New Caledonia',
4885 'NZ': 'New Zealand',
4886 'NI': 'Nicaragua',
4887 'NE': 'Niger',
4888 'NG': 'Nigeria',
4889 'NU': 'Niue',
4890 'NF': 'Norfolk Island',
4891 'MP': 'Northern Mariana Islands',
4892 'NO': 'Norway',
4893 'OM': 'Oman',
4894 'PK': 'Pakistan',
4895 'PW': 'Palau',
4896 'PS': 'Palestine, State of',
4897 'PA': 'Panama',
4898 'PG': 'Papua New Guinea',
4899 'PY': 'Paraguay',
4900 'PE': 'Peru',
4901 'PH': 'Philippines',
4902 'PN': 'Pitcairn',
4903 'PL': 'Poland',
4904 'PT': 'Portugal',
4905 'PR': 'Puerto Rico',
4906 'QA': 'Qatar',
4907 'RE': 'RĆ©union',
4908 'RO': 'Romania',
4909 'RU': 'Russian Federation',
4910 'RW': 'Rwanda',
4911 'BL': 'Saint BarthƩlemy',
4912 'SH': 'Saint Helena, Ascension and Tristan da Cunha',
4913 'KN': 'Saint Kitts and Nevis',
4914 'LC': 'Saint Lucia',
4915 'MF': 'Saint Martin (French part)',
4916 'PM': 'Saint Pierre and Miquelon',
4917 'VC': 'Saint Vincent and the Grenadines',
4918 'WS': 'Samoa',
4919 'SM': 'San Marino',
4920 'ST': 'Sao Tome and Principe',
4921 'SA': 'Saudi Arabia',
4922 'SN': 'Senegal',
4923 'RS': 'Serbia',
4924 'SC': 'Seychelles',
4925 'SL': 'Sierra Leone',
4926 'SG': 'Singapore',
4927 'SX': 'Sint Maarten (Dutch part)',
4928 'SK': 'Slovakia',
4929 'SI': 'Slovenia',
4930 'SB': 'Solomon Islands',
4931 'SO': 'Somalia',
4932 'ZA': 'South Africa',
4933 'GS': 'South Georgia and the South Sandwich Islands',
4934 'SS': 'South Sudan',
4935 'ES': 'Spain',
4936 'LK': 'Sri Lanka',
4937 'SD': 'Sudan',
4938 'SR': 'Suriname',
4939 'SJ': 'Svalbard and Jan Mayen',
4940 'SZ': 'Swaziland',
4941 'SE': 'Sweden',
4942 'CH': 'Switzerland',
4943 'SY': 'Syrian Arab Republic',
4944 'TW': 'Taiwan, Province of China',
4945 'TJ': 'Tajikistan',
4946 'TZ': 'Tanzania, United Republic of',
4947 'TH': 'Thailand',
4948 'TL': 'Timor-Leste',
4949 'TG': 'Togo',
4950 'TK': 'Tokelau',
4951 'TO': 'Tonga',
4952 'TT': 'Trinidad and Tobago',
4953 'TN': 'Tunisia',
4954 'TR': 'Turkey',
4955 'TM': 'Turkmenistan',
4956 'TC': 'Turks and Caicos Islands',
4957 'TV': 'Tuvalu',
4958 'UG': 'Uganda',
4959 'UA': 'Ukraine',
4960 'AE': 'United Arab Emirates',
4961 'GB': 'United Kingdom',
4962 'US': 'United States',
4963 'UM': 'United States Minor Outlying Islands',
4964 'UY': 'Uruguay',
4965 'UZ': 'Uzbekistan',
4966 'VU': 'Vanuatu',
4967 'VE': 'Venezuela, Bolivarian Republic of',
4968 'VN': 'Viet Nam',
4969 'VG': 'Virgin Islands, British',
4970 'VI': 'Virgin Islands, U.S.',
4971 'WF': 'Wallis and Futuna',
4972 'EH': 'Western Sahara',
4973 'YE': 'Yemen',
4974 'ZM': 'Zambia',
4975 'ZW': 'Zimbabwe',
4976 }
4977
4978 @classmethod
4979 def short2full(cls, code):
4980 """Convert an ISO 3166-2 country code to the corresponding full name"""
4981 return cls._country_map.get(code.upper())
4982
4983
4984 class GeoUtils(object):
4985 # Major IPv4 address blocks per country
4986 _country_ip_map = {
4987 'AD': '46.172.224.0/19',
4988 'AE': '94.200.0.0/13',
4989 'AF': '149.54.0.0/17',
4990 'AG': '209.59.64.0/18',
4991 'AI': '204.14.248.0/21',
4992 'AL': '46.99.0.0/16',
4993 'AM': '46.70.0.0/15',
4994 'AO': '105.168.0.0/13',
4995 'AP': '182.50.184.0/21',
4996 'AQ': '23.154.160.0/24',
4997 'AR': '181.0.0.0/12',
4998 'AS': '202.70.112.0/20',
4999 'AT': '77.116.0.0/14',
5000 'AU': '1.128.0.0/11',
5001 'AW': '181.41.0.0/18',
5002 'AX': '185.217.4.0/22',
5003 'AZ': '5.197.0.0/16',
5004 'BA': '31.176.128.0/17',
5005 'BB': '65.48.128.0/17',
5006 'BD': '114.130.0.0/16',
5007 'BE': '57.0.0.0/8',
5008 'BF': '102.178.0.0/15',
5009 'BG': '95.42.0.0/15',
5010 'BH': '37.131.0.0/17',
5011 'BI': '154.117.192.0/18',
5012 'BJ': '137.255.0.0/16',
5013 'BL': '185.212.72.0/23',
5014 'BM': '196.12.64.0/18',
5015 'BN': '156.31.0.0/16',
5016 'BO': '161.56.0.0/16',
5017 'BQ': '161.0.80.0/20',
5018 'BR': '191.128.0.0/12',
5019 'BS': '24.51.64.0/18',
5020 'BT': '119.2.96.0/19',
5021 'BW': '168.167.0.0/16',
5022 'BY': '178.120.0.0/13',
5023 'BZ': '179.42.192.0/18',
5024 'CA': '99.224.0.0/11',
5025 'CD': '41.243.0.0/16',
5026 'CF': '197.242.176.0/21',
5027 'CG': '160.113.0.0/16',
5028 'CH': '85.0.0.0/13',
5029 'CI': '102.136.0.0/14',
5030 'CK': '202.65.32.0/19',
5031 'CL': '152.172.0.0/14',
5032 'CM': '102.244.0.0/14',
5033 'CN': '36.128.0.0/10',
5034 'CO': '181.240.0.0/12',
5035 'CR': '201.192.0.0/12',
5036 'CU': '152.206.0.0/15',
5037 'CV': '165.90.96.0/19',
5038 'CW': '190.88.128.0/17',
5039 'CY': '31.153.0.0/16',
5040 'CZ': '88.100.0.0/14',
5041 'DE': '53.0.0.0/8',
5042 'DJ': '197.241.0.0/17',
5043 'DK': '87.48.0.0/12',
5044 'DM': '192.243.48.0/20',
5045 'DO': '152.166.0.0/15',
5046 'DZ': '41.96.0.0/12',
5047 'EC': '186.68.0.0/15',
5048 'EE': '90.190.0.0/15',
5049 'EG': '156.160.0.0/11',
5050 'ER': '196.200.96.0/20',
5051 'ES': '88.0.0.0/11',
5052 'ET': '196.188.0.0/14',
5053 'EU': '2.16.0.0/13',
5054 'FI': '91.152.0.0/13',
5055 'FJ': '144.120.0.0/16',
5056 'FK': '80.73.208.0/21',
5057 'FM': '119.252.112.0/20',
5058 'FO': '88.85.32.0/19',
5059 'FR': '90.0.0.0/9',
5060 'GA': '41.158.0.0/15',
5061 'GB': '25.0.0.0/8',
5062 'GD': '74.122.88.0/21',
5063 'GE': '31.146.0.0/16',
5064 'GF': '161.22.64.0/18',
5065 'GG': '62.68.160.0/19',
5066 'GH': '154.160.0.0/12',
5067 'GI': '95.164.0.0/16',
5068 'GL': '88.83.0.0/19',
5069 'GM': '160.182.0.0/15',
5070 'GN': '197.149.192.0/18',
5071 'GP': '104.250.0.0/19',
5072 'GQ': '105.235.224.0/20',
5073 'GR': '94.64.0.0/13',
5074 'GT': '168.234.0.0/16',
5075 'GU': '168.123.0.0/16',
5076 'GW': '197.214.80.0/20',
5077 'GY': '181.41.64.0/18',
5078 'HK': '113.252.0.0/14',
5079 'HN': '181.210.0.0/16',
5080 'HR': '93.136.0.0/13',
5081 'HT': '148.102.128.0/17',
5082 'HU': '84.0.0.0/14',
5083 'ID': '39.192.0.0/10',
5084 'IE': '87.32.0.0/12',
5085 'IL': '79.176.0.0/13',
5086 'IM': '5.62.80.0/20',
5087 'IN': '117.192.0.0/10',
5088 'IO': '203.83.48.0/21',
5089 'IQ': '37.236.0.0/14',
5090 'IR': '2.176.0.0/12',
5091 'IS': '82.221.0.0/16',
5092 'IT': '79.0.0.0/10',
5093 'JE': '87.244.64.0/18',
5094 'JM': '72.27.0.0/17',
5095 'JO': '176.29.0.0/16',
5096 'JP': '133.0.0.0/8',
5097 'KE': '105.48.0.0/12',
5098 'KG': '158.181.128.0/17',
5099 'KH': '36.37.128.0/17',
5100 'KI': '103.25.140.0/22',
5101 'KM': '197.255.224.0/20',
5102 'KN': '198.167.192.0/19',
5103 'KP': '175.45.176.0/22',
5104 'KR': '175.192.0.0/10',
5105 'KW': '37.36.0.0/14',
5106 'KY': '64.96.0.0/15',
5107 'KZ': '2.72.0.0/13',
5108 'LA': '115.84.64.0/18',
5109 'LB': '178.135.0.0/16',
5110 'LC': '24.92.144.0/20',
5111 'LI': '82.117.0.0/19',
5112 'LK': '112.134.0.0/15',
5113 'LR': '102.183.0.0/16',
5114 'LS': '129.232.0.0/17',
5115 'LT': '78.56.0.0/13',
5116 'LU': '188.42.0.0/16',
5117 'LV': '46.109.0.0/16',
5118 'LY': '41.252.0.0/14',
5119 'MA': '105.128.0.0/11',
5120 'MC': '88.209.64.0/18',
5121 'MD': '37.246.0.0/16',
5122 'ME': '178.175.0.0/17',
5123 'MF': '74.112.232.0/21',
5124 'MG': '154.126.0.0/17',
5125 'MH': '117.103.88.0/21',
5126 'MK': '77.28.0.0/15',
5127 'ML': '154.118.128.0/18',
5128 'MM': '37.111.0.0/17',
5129 'MN': '49.0.128.0/17',
5130 'MO': '60.246.0.0/16',
5131 'MP': '202.88.64.0/20',
5132 'MQ': '109.203.224.0/19',
5133 'MR': '41.188.64.0/18',
5134 'MS': '208.90.112.0/22',
5135 'MT': '46.11.0.0/16',
5136 'MU': '105.16.0.0/12',
5137 'MV': '27.114.128.0/18',
5138 'MW': '102.70.0.0/15',
5139 'MX': '187.192.0.0/11',
5140 'MY': '175.136.0.0/13',
5141 'MZ': '197.218.0.0/15',
5142 'NA': '41.182.0.0/16',
5143 'NC': '101.101.0.0/18',
5144 'NE': '197.214.0.0/18',
5145 'NF': '203.17.240.0/22',
5146 'NG': '105.112.0.0/12',
5147 'NI': '186.76.0.0/15',
5148 'NL': '145.96.0.0/11',
5149 'NO': '84.208.0.0/13',
5150 'NP': '36.252.0.0/15',
5151 'NR': '203.98.224.0/19',
5152 'NU': '49.156.48.0/22',
5153 'NZ': '49.224.0.0/14',
5154 'OM': '5.36.0.0/15',
5155 'PA': '186.72.0.0/15',
5156 'PE': '186.160.0.0/14',
5157 'PF': '123.50.64.0/18',
5158 'PG': '124.240.192.0/19',
5159 'PH': '49.144.0.0/13',
5160 'PK': '39.32.0.0/11',
5161 'PL': '83.0.0.0/11',
5162 'PM': '70.36.0.0/20',
5163 'PR': '66.50.0.0/16',
5164 'PS': '188.161.0.0/16',
5165 'PT': '85.240.0.0/13',
5166 'PW': '202.124.224.0/20',
5167 'PY': '181.120.0.0/14',
5168 'QA': '37.210.0.0/15',
5169 'RE': '102.35.0.0/16',
5170 'RO': '79.112.0.0/13',
5171 'RS': '93.86.0.0/15',
5172 'RU': '5.136.0.0/13',
5173 'RW': '41.186.0.0/16',
5174 'SA': '188.48.0.0/13',
5175 'SB': '202.1.160.0/19',
5176 'SC': '154.192.0.0/11',
5177 'SD': '102.120.0.0/13',
5178 'SE': '78.64.0.0/12',
5179 'SG': '8.128.0.0/10',
5180 'SI': '188.196.0.0/14',
5181 'SK': '78.98.0.0/15',
5182 'SL': '102.143.0.0/17',
5183 'SM': '89.186.32.0/19',
5184 'SN': '41.82.0.0/15',
5185 'SO': '154.115.192.0/18',
5186 'SR': '186.179.128.0/17',
5187 'SS': '105.235.208.0/21',
5188 'ST': '197.159.160.0/19',
5189 'SV': '168.243.0.0/16',
5190 'SX': '190.102.0.0/20',
5191 'SY': '5.0.0.0/16',
5192 'SZ': '41.84.224.0/19',
5193 'TC': '65.255.48.0/20',
5194 'TD': '154.68.128.0/19',
5195 'TG': '196.168.0.0/14',
5196 'TH': '171.96.0.0/13',
5197 'TJ': '85.9.128.0/18',
5198 'TK': '27.96.24.0/21',
5199 'TL': '180.189.160.0/20',
5200 'TM': '95.85.96.0/19',
5201 'TN': '197.0.0.0/11',
5202 'TO': '175.176.144.0/21',
5203 'TR': '78.160.0.0/11',
5204 'TT': '186.44.0.0/15',
5205 'TV': '202.2.96.0/19',
5206 'TW': '120.96.0.0/11',
5207 'TZ': '156.156.0.0/14',
5208 'UA': '37.52.0.0/14',
5209 'UG': '102.80.0.0/13',
5210 'US': '6.0.0.0/8',
5211 'UY': '167.56.0.0/13',
5212 'UZ': '84.54.64.0/18',
5213 'VA': '212.77.0.0/19',
5214 'VC': '207.191.240.0/21',
5215 'VE': '186.88.0.0/13',
5216 'VG': '66.81.192.0/20',
5217 'VI': '146.226.0.0/16',
5218 'VN': '14.160.0.0/11',
5219 'VU': '202.80.32.0/20',
5220 'WF': '117.20.32.0/21',
5221 'WS': '202.4.32.0/19',
5222 'YE': '134.35.0.0/16',
5223 'YT': '41.242.116.0/22',
5224 'ZA': '41.0.0.0/11',
5225 'ZM': '102.144.0.0/13',
5226 'ZW': '102.177.192.0/18',
5227 }
5228
5229 @classmethod
5230 def random_ipv4(cls, code_or_block):
5231 if len(code_or_block) == 2:
5232 block = cls._country_ip_map.get(code_or_block.upper())
5233 if not block:
5234 return None
5235 else:
5236 block = code_or_block
5237 addr, preflen = block.split('/')
5238 addr_min = compat_struct_unpack('!L', socket.inet_aton(addr))[0]
5239 addr_max = addr_min | (0xffffffff >> int(preflen))
5240 return compat_str(socket.inet_ntoa(
5241 compat_struct_pack('!L', random.randint(addr_min, addr_max))))
5242
5243
5244 class PerRequestProxyHandler(compat_urllib_request.ProxyHandler):
5245 def __init__(self, proxies=None):
5246 # Set default handlers
5247 for type in ('http', 'https'):
5248 setattr(self, '%s_open' % type,
5249 lambda r, proxy='__noproxy__', type=type, meth=self.proxy_open:
5250 meth(r, proxy, type))
5251 compat_urllib_request.ProxyHandler.__init__(self, proxies)
5252
5253 def proxy_open(self, req, proxy, type):
5254 req_proxy = req.headers.get('Ytdl-request-proxy')
5255 if req_proxy is not None:
5256 proxy = req_proxy
5257 del req.headers['Ytdl-request-proxy']
5258
5259 if proxy == '__noproxy__':
5260 return None # No Proxy
5261 if compat_urlparse.urlparse(proxy).scheme.lower() in ('socks', 'socks4', 'socks4a', 'socks5'):
5262 req.add_header('Ytdl-socks-proxy', proxy)
5263 # youtube-dl's http/https handlers do wrapping the socket with socks
5264 return None
5265 return compat_urllib_request.ProxyHandler.proxy_open(
5266 self, req, proxy, type)
5267
5268
5269 # Both long_to_bytes and bytes_to_long are adapted from PyCrypto, which is
5270 # released into Public Domain
5271 # https://github.com/dlitz/pycrypto/blob/master/lib/Crypto/Util/number.py#L387
5272
5273 def long_to_bytes(n, blocksize=0):
5274 """long_to_bytes(n:long, blocksize:int) : string
5275 Convert a long integer to a byte string.
5276
5277 If optional blocksize is given and greater than zero, pad the front of the
5278 byte string with binary zeros so that the length is a multiple of
5279 blocksize.
5280 """
5281 # after much testing, this algorithm was deemed to be the fastest
5282 s = b''
5283 n = int(n)
5284 while n > 0:
5285 s = compat_struct_pack('>I', n & 0xffffffff) + s
5286 n = n >> 32
5287 # strip off leading zeros
5288 for i in range(len(s)):
5289 if s[i] != b'\000'[0]:
5290 break
5291 else:
5292 # only happens when n == 0
5293 s = b'\000'
5294 i = 0
5295 s = s[i:]
5296 # add back some pad bytes. this could be done more efficiently w.r.t. the
5297 # de-padding being done above, but sigh...
5298 if blocksize > 0 and len(s) % blocksize:
5299 s = (blocksize - len(s) % blocksize) * b'\000' + s
5300 return s
5301
5302
5303 def bytes_to_long(s):
5304 """bytes_to_long(string) : long
5305 Convert a byte string to a long integer.
5306
5307 This is (essentially) the inverse of long_to_bytes().
5308 """
5309 acc = 0
5310 length = len(s)
5311 if length % 4:
5312 extra = (4 - length % 4)
5313 s = b'\000' * extra + s
5314 length = length + extra
5315 for i in range(0, length, 4):
5316 acc = (acc << 32) + compat_struct_unpack('>I', s[i:i + 4])[0]
5317 return acc
5318
5319
5320 def ohdave_rsa_encrypt(data, exponent, modulus):
5321 '''
5322 Implement OHDave's RSA algorithm. See http://www.ohdave.com/rsa/
5323
5324 Input:
5325 data: data to encrypt, bytes-like object
5326 exponent, modulus: parameter e and N of RSA algorithm, both integer
5327 Output: hex string of encrypted data
5328
5329 Limitation: supports one block encryption only
5330 '''
5331
5332 payload = int(binascii.hexlify(data[::-1]), 16)
5333 encrypted = pow(payload, exponent, modulus)
5334 return '%x' % encrypted
5335
5336
5337 def pkcs1pad(data, length):
5338 """
5339 Padding input data with PKCS#1 scheme
5340
5341 @param {int[]} data input data
5342 @param {int} length target length
5343 @returns {int[]} padded data
5344 """
5345 if len(data) > length - 11:
5346 raise ValueError('Input data too long for PKCS#1 padding')
5347
5348 pseudo_random = [random.randint(0, 254) for _ in range(length - len(data) - 3)]
5349 return [0, 2] + pseudo_random + [0] + data
5350
5351
5352 def encode_base_n(num, n, table=None):
5353 FULL_TABLE = '0123456789abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ'
5354 if not table:
5355 table = FULL_TABLE[:n]
5356
5357 if n > len(table):
5358 raise ValueError('base %d exceeds table length %d' % (n, len(table)))
5359
5360 if num == 0:
5361 return table[0]
5362
5363 ret = ''
5364 while num:
5365 ret = table[num % n] + ret
5366 num = num // n
5367 return ret
5368
5369
5370 def decode_packed_codes(code):
5371 mobj = re.search(PACKED_CODES_RE, code)
5372 obfucasted_code, base, count, symbols = mobj.groups()
5373 base = int(base)
5374 count = int(count)
5375 symbols = symbols.split('|')
5376 symbol_table = {}
5377
5378 while count:
5379 count -= 1
5380 base_n_count = encode_base_n(count, base)
5381 symbol_table[base_n_count] = symbols[count] or base_n_count
5382
5383 return re.sub(
5384 r'\b(\w+)\b', lambda mobj: symbol_table[mobj.group(0)],
5385 obfucasted_code)
5386
5387
5388 def caesar(s, alphabet, shift):
5389 if shift == 0:
5390 return s
5391 l = len(alphabet)
5392 return ''.join(
5393 alphabet[(alphabet.index(c) + shift) % l] if c in alphabet else c
5394 for c in s)
5395
5396
5397 def rot47(s):
5398 return caesar(s, r'''!"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~''', 47)
5399
5400
5401 def parse_m3u8_attributes(attrib):
5402 info = {}
5403 for (key, val) in re.findall(r'(?P<key>[A-Z0-9-]+)=(?P<val>"[^"]+"|[^",]+)(?:,|$)', attrib):
5404 if val.startswith('"'):
5405 val = val[1:-1]
5406 info[key] = val
5407 return info
5408
5409
5410 def urshift(val, n):
5411 return val >> n if val >= 0 else (val + 0x100000000) >> n
5412
5413
5414 # Based on png2str() written by @gdkchan and improved by @yokrysty
5415 # Originally posted at https://github.com/ytdl-org/youtube-dl/issues/9706
5416 def decode_png(png_data):
5417 # Reference: https://www.w3.org/TR/PNG/
5418 header = png_data[8:]
5419
5420 if png_data[:8] != b'\x89PNG\x0d\x0a\x1a\x0a' or header[4:8] != b'IHDR':
5421 raise IOError('Not a valid PNG file.')
5422
5423 int_map = {1: '>B', 2: '>H', 4: '>I'}
5424 unpack_integer = lambda x: compat_struct_unpack(int_map[len(x)], x)[0]
5425
5426 chunks = []
5427
5428 while header:
5429 length = unpack_integer(header[:4])
5430 header = header[4:]
5431
5432 chunk_type = header[:4]
5433 header = header[4:]
5434
5435 chunk_data = header[:length]
5436 header = header[length:]
5437
5438 header = header[4:] # Skip CRC
5439
5440 chunks.append({
5441 'type': chunk_type,
5442 'length': length,
5443 'data': chunk_data
5444 })
5445
5446 ihdr = chunks[0]['data']
5447
5448 width = unpack_integer(ihdr[:4])
5449 height = unpack_integer(ihdr[4:8])
5450
5451 idat = b''
5452
5453 for chunk in chunks:
5454 if chunk['type'] == b'IDAT':
5455 idat += chunk['data']
5456
5457 if not idat:
5458 raise IOError('Unable to read PNG data.')
5459
5460 decompressed_data = bytearray(zlib.decompress(idat))
5461
5462 stride = width * 3
5463 pixels = []
5464
5465 def _get_pixel(idx):
5466 x = idx % stride
5467 y = idx // stride
5468 return pixels[y][x]
5469
5470 for y in range(height):
5471 basePos = y * (1 + stride)
5472 filter_type = decompressed_data[basePos]
5473
5474 current_row = []
5475
5476 pixels.append(current_row)
5477
5478 for x in range(stride):
5479 color = decompressed_data[1 + basePos + x]
5480 basex = y * stride + x
5481 left = 0
5482 up = 0
5483
5484 if x > 2:
5485 left = _get_pixel(basex - 3)
5486 if y > 0:
5487 up = _get_pixel(basex - stride)
5488
5489 if filter_type == 1: # Sub
5490 color = (color + left) & 0xff
5491 elif filter_type == 2: # Up
5492 color = (color + up) & 0xff
5493 elif filter_type == 3: # Average
5494 color = (color + ((left + up) >> 1)) & 0xff
5495 elif filter_type == 4: # Paeth
5496 a = left
5497 b = up
5498 c = 0
5499
5500 if x > 2 and y > 0:
5501 c = _get_pixel(basex - stride - 3)
5502
5503 p = a + b - c
5504
5505 pa = abs(p - a)
5506 pb = abs(p - b)
5507 pc = abs(p - c)
5508
5509 if pa <= pb and pa <= pc:
5510 color = (color + a) & 0xff
5511 elif pb <= pc:
5512 color = (color + b) & 0xff
5513 else:
5514 color = (color + c) & 0xff
5515
5516 current_row.append(color)
5517
5518 return width, height, pixels
5519
5520
5521 def write_xattr(path, key, value):
5522 # This mess below finds the best xattr tool for the job
5523 try:
5524 # try the pyxattr module...
5525 import xattr
5526
5527 if hasattr(xattr, 'set'): # pyxattr
5528 # Unicode arguments are not supported in python-pyxattr until
5529 # version 0.5.0
5530 # See https://github.com/ytdl-org/youtube-dl/issues/5498
5531 pyxattr_required_version = '0.5.0'
5532 if version_tuple(xattr.__version__) < version_tuple(pyxattr_required_version):
5533 # TODO: fallback to CLI tools
5534 raise XAttrUnavailableError(
5535 'python-pyxattr is detected but is too old. '
5536 'youtube-dl requires %s or above while your version is %s. '
5537 'Falling back to other xattr implementations' % (
5538 pyxattr_required_version, xattr.__version__))
5539
5540 setxattr = xattr.set
5541 else: # xattr
5542 setxattr = xattr.setxattr
5543
5544 try:
5545 setxattr(path, key, value)
5546 except EnvironmentError as e:
5547 raise XAttrMetadataError(e.errno, e.strerror)
5548
5549 except ImportError:
5550 if compat_os_name == 'nt':
5551 # Write xattrs to NTFS Alternate Data Streams:
5552 # http://en.wikipedia.org/wiki/NTFS#Alternate_data_streams_.28ADS.29
5553 assert ':' not in key
5554 assert os.path.exists(path)
5555
5556 ads_fn = path + ':' + key
5557 try:
5558 with open(ads_fn, 'wb') as f:
5559 f.write(value)
5560 except EnvironmentError as e:
5561 raise XAttrMetadataError(e.errno, e.strerror)
5562 else:
5563 user_has_setfattr = check_executable('setfattr', ['--version'])
5564 user_has_xattr = check_executable('xattr', ['-h'])
5565
5566 if user_has_setfattr or user_has_xattr:
5567
5568 value = value.decode('utf-8')
5569 if user_has_setfattr:
5570 executable = 'setfattr'
5571 opts = ['-n', key, '-v', value]
5572 elif user_has_xattr:
5573 executable = 'xattr'
5574 opts = ['-w', key, value]
5575
5576 cmd = ([encodeFilename(executable, True)]
5577 + [encodeArgument(o) for o in opts]
5578 + [encodeFilename(path, True)])
5579
5580 try:
5581 p = subprocess.Popen(
5582 cmd, stdout=subprocess.PIPE, stderr=subprocess.PIPE, stdin=subprocess.PIPE)
5583 except EnvironmentError as e:
5584 raise XAttrMetadataError(e.errno, e.strerror)
5585 stdout, stderr = p.communicate()
5586 stderr = stderr.decode('utf-8', 'replace')
5587 if p.returncode != 0:
5588 raise XAttrMetadataError(p.returncode, stderr)
5589
5590 else:
5591 # On Unix, and can't find pyxattr, setfattr, or xattr.
5592 if sys.platform.startswith('linux'):
5593 raise XAttrUnavailableError(
5594 "Couldn't find a tool to set the xattrs. "
5595 "Install either the python 'pyxattr' or 'xattr' "
5596 "modules, or the GNU 'attr' package "
5597 "(which contains the 'setfattr' tool).")
5598 else:
5599 raise XAttrUnavailableError(
5600 "Couldn't find a tool to set the xattrs. "
5601 "Install either the python 'xattr' module, "
5602 "or the 'xattr' binary.")
5603
5604
5605 def random_birthday(year_field, month_field, day_field):
5606 start_date = datetime.date(1950, 1, 1)
5607 end_date = datetime.date(1995, 12, 31)
5608 offset = random.randint(0, (end_date - start_date).days)
5609 random_date = start_date + datetime.timedelta(offset)
5610 return {
5611 year_field: str(random_date.year),
5612 month_field: str(random_date.month),
5613 day_field: str(random_date.day),
5614 }