]> jfr.im git - yt-dlp.git/blame_incremental - yt_dlp/utils.py
[EmbedThumbnail] Add compat-option `embed-thumbnail-atomicparsley`
[yt-dlp.git] / yt_dlp / utils.py
... / ...
CommitLineData
1#!/usr/bin/env python3
2# coding: utf-8
3
4from __future__ import unicode_literals
5
6import base64
7import binascii
8import calendar
9import codecs
10import collections
11import contextlib
12import ctypes
13import datetime
14import email.utils
15import email.header
16import errno
17import functools
18import gzip
19import imp
20import io
21import itertools
22import json
23import locale
24import math
25import operator
26import os
27import platform
28import random
29import re
30import socket
31import ssl
32import subprocess
33import sys
34import tempfile
35import time
36import traceback
37import xml.etree.ElementTree
38import zlib
39
40from .compat import (
41 compat_HTMLParseError,
42 compat_HTMLParser,
43 compat_HTTPError,
44 compat_basestring,
45 compat_chr,
46 compat_cookiejar,
47 compat_ctypes_WINFUNCTYPE,
48 compat_etree_fromstring,
49 compat_expanduser,
50 compat_html_entities,
51 compat_html_entities_html5,
52 compat_http_client,
53 compat_integer_types,
54 compat_numeric_types,
55 compat_kwargs,
56 compat_os_name,
57 compat_parse_qs,
58 compat_shlex_quote,
59 compat_str,
60 compat_struct_pack,
61 compat_struct_unpack,
62 compat_urllib_error,
63 compat_urllib_parse,
64 compat_urllib_parse_urlencode,
65 compat_urllib_parse_urlparse,
66 compat_urllib_parse_urlunparse,
67 compat_urllib_parse_quote,
68 compat_urllib_parse_quote_plus,
69 compat_urllib_parse_unquote_plus,
70 compat_urllib_request,
71 compat_urlparse,
72 compat_xpath,
73)
74
75from .socks import (
76 ProxyType,
77 sockssocket,
78)
79
80
81def register_socks_protocols():
82 # "Register" SOCKS protocols
83 # In Python < 2.6.5, urlsplit() suffers from bug https://bugs.python.org/issue7904
84 # URLs with protocols not in urlparse.uses_netloc are not handled correctly
85 for scheme in ('socks', 'socks4', 'socks4a', 'socks5'):
86 if scheme not in compat_urlparse.uses_netloc:
87 compat_urlparse.uses_netloc.append(scheme)
88
89
90# This is not clearly defined otherwise
91compiled_regex_type = type(re.compile(''))
92
93
94def random_user_agent():
95 _USER_AGENT_TPL = 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/%s Safari/537.36'
96 _CHROME_VERSIONS = (
97 '74.0.3729.129',
98 '76.0.3780.3',
99 '76.0.3780.2',
100 '74.0.3729.128',
101 '76.0.3780.1',
102 '76.0.3780.0',
103 '75.0.3770.15',
104 '74.0.3729.127',
105 '74.0.3729.126',
106 '76.0.3779.1',
107 '76.0.3779.0',
108 '75.0.3770.14',
109 '74.0.3729.125',
110 '76.0.3778.1',
111 '76.0.3778.0',
112 '75.0.3770.13',
113 '74.0.3729.124',
114 '74.0.3729.123',
115 '73.0.3683.121',
116 '76.0.3777.1',
117 '76.0.3777.0',
118 '75.0.3770.12',
119 '74.0.3729.122',
120 '76.0.3776.4',
121 '75.0.3770.11',
122 '74.0.3729.121',
123 '76.0.3776.3',
124 '76.0.3776.2',
125 '73.0.3683.120',
126 '74.0.3729.120',
127 '74.0.3729.119',
128 '74.0.3729.118',
129 '76.0.3776.1',
130 '76.0.3776.0',
131 '76.0.3775.5',
132 '75.0.3770.10',
133 '74.0.3729.117',
134 '76.0.3775.4',
135 '76.0.3775.3',
136 '74.0.3729.116',
137 '75.0.3770.9',
138 '76.0.3775.2',
139 '76.0.3775.1',
140 '76.0.3775.0',
141 '75.0.3770.8',
142 '74.0.3729.115',
143 '74.0.3729.114',
144 '76.0.3774.1',
145 '76.0.3774.0',
146 '75.0.3770.7',
147 '74.0.3729.113',
148 '74.0.3729.112',
149 '74.0.3729.111',
150 '76.0.3773.1',
151 '76.0.3773.0',
152 '75.0.3770.6',
153 '74.0.3729.110',
154 '74.0.3729.109',
155 '76.0.3772.1',
156 '76.0.3772.0',
157 '75.0.3770.5',
158 '74.0.3729.108',
159 '74.0.3729.107',
160 '76.0.3771.1',
161 '76.0.3771.0',
162 '75.0.3770.4',
163 '74.0.3729.106',
164 '74.0.3729.105',
165 '75.0.3770.3',
166 '74.0.3729.104',
167 '74.0.3729.103',
168 '74.0.3729.102',
169 '75.0.3770.2',
170 '74.0.3729.101',
171 '75.0.3770.1',
172 '75.0.3770.0',
173 '74.0.3729.100',
174 '75.0.3769.5',
175 '75.0.3769.4',
176 '74.0.3729.99',
177 '75.0.3769.3',
178 '75.0.3769.2',
179 '75.0.3768.6',
180 '74.0.3729.98',
181 '75.0.3769.1',
182 '75.0.3769.0',
183 '74.0.3729.97',
184 '73.0.3683.119',
185 '73.0.3683.118',
186 '74.0.3729.96',
187 '75.0.3768.5',
188 '75.0.3768.4',
189 '75.0.3768.3',
190 '75.0.3768.2',
191 '74.0.3729.95',
192 '74.0.3729.94',
193 '75.0.3768.1',
194 '75.0.3768.0',
195 '74.0.3729.93',
196 '74.0.3729.92',
197 '73.0.3683.117',
198 '74.0.3729.91',
199 '75.0.3766.3',
200 '74.0.3729.90',
201 '75.0.3767.2',
202 '75.0.3767.1',
203 '75.0.3767.0',
204 '74.0.3729.89',
205 '73.0.3683.116',
206 '75.0.3766.2',
207 '74.0.3729.88',
208 '75.0.3766.1',
209 '75.0.3766.0',
210 '74.0.3729.87',
211 '73.0.3683.115',
212 '74.0.3729.86',
213 '75.0.3765.1',
214 '75.0.3765.0',
215 '74.0.3729.85',
216 '73.0.3683.114',
217 '74.0.3729.84',
218 '75.0.3764.1',
219 '75.0.3764.0',
220 '74.0.3729.83',
221 '73.0.3683.113',
222 '75.0.3763.2',
223 '75.0.3761.4',
224 '74.0.3729.82',
225 '75.0.3763.1',
226 '75.0.3763.0',
227 '74.0.3729.81',
228 '73.0.3683.112',
229 '75.0.3762.1',
230 '75.0.3762.0',
231 '74.0.3729.80',
232 '75.0.3761.3',
233 '74.0.3729.79',
234 '73.0.3683.111',
235 '75.0.3761.2',
236 '74.0.3729.78',
237 '74.0.3729.77',
238 '75.0.3761.1',
239 '75.0.3761.0',
240 '73.0.3683.110',
241 '74.0.3729.76',
242 '74.0.3729.75',
243 '75.0.3760.0',
244 '74.0.3729.74',
245 '75.0.3759.8',
246 '75.0.3759.7',
247 '75.0.3759.6',
248 '74.0.3729.73',
249 '75.0.3759.5',
250 '74.0.3729.72',
251 '73.0.3683.109',
252 '75.0.3759.4',
253 '75.0.3759.3',
254 '74.0.3729.71',
255 '75.0.3759.2',
256 '74.0.3729.70',
257 '73.0.3683.108',
258 '74.0.3729.69',
259 '75.0.3759.1',
260 '75.0.3759.0',
261 '74.0.3729.68',
262 '73.0.3683.107',
263 '74.0.3729.67',
264 '75.0.3758.1',
265 '75.0.3758.0',
266 '74.0.3729.66',
267 '73.0.3683.106',
268 '74.0.3729.65',
269 '75.0.3757.1',
270 '75.0.3757.0',
271 '74.0.3729.64',
272 '73.0.3683.105',
273 '74.0.3729.63',
274 '75.0.3756.1',
275 '75.0.3756.0',
276 '74.0.3729.62',
277 '73.0.3683.104',
278 '75.0.3755.3',
279 '75.0.3755.2',
280 '73.0.3683.103',
281 '75.0.3755.1',
282 '75.0.3755.0',
283 '74.0.3729.61',
284 '73.0.3683.102',
285 '74.0.3729.60',
286 '75.0.3754.2',
287 '74.0.3729.59',
288 '75.0.3753.4',
289 '74.0.3729.58',
290 '75.0.3754.1',
291 '75.0.3754.0',
292 '74.0.3729.57',
293 '73.0.3683.101',
294 '75.0.3753.3',
295 '75.0.3752.2',
296 '75.0.3753.2',
297 '74.0.3729.56',
298 '75.0.3753.1',
299 '75.0.3753.0',
300 '74.0.3729.55',
301 '73.0.3683.100',
302 '74.0.3729.54',
303 '75.0.3752.1',
304 '75.0.3752.0',
305 '74.0.3729.53',
306 '73.0.3683.99',
307 '74.0.3729.52',
308 '75.0.3751.1',
309 '75.0.3751.0',
310 '74.0.3729.51',
311 '73.0.3683.98',
312 '74.0.3729.50',
313 '75.0.3750.0',
314 '74.0.3729.49',
315 '74.0.3729.48',
316 '74.0.3729.47',
317 '75.0.3749.3',
318 '74.0.3729.46',
319 '73.0.3683.97',
320 '75.0.3749.2',
321 '74.0.3729.45',
322 '75.0.3749.1',
323 '75.0.3749.0',
324 '74.0.3729.44',
325 '73.0.3683.96',
326 '74.0.3729.43',
327 '74.0.3729.42',
328 '75.0.3748.1',
329 '75.0.3748.0',
330 '74.0.3729.41',
331 '75.0.3747.1',
332 '73.0.3683.95',
333 '75.0.3746.4',
334 '74.0.3729.40',
335 '74.0.3729.39',
336 '75.0.3747.0',
337 '75.0.3746.3',
338 '75.0.3746.2',
339 '74.0.3729.38',
340 '75.0.3746.1',
341 '75.0.3746.0',
342 '74.0.3729.37',
343 '73.0.3683.94',
344 '75.0.3745.5',
345 '75.0.3745.4',
346 '75.0.3745.3',
347 '75.0.3745.2',
348 '74.0.3729.36',
349 '75.0.3745.1',
350 '75.0.3745.0',
351 '75.0.3744.2',
352 '74.0.3729.35',
353 '73.0.3683.93',
354 '74.0.3729.34',
355 '75.0.3744.1',
356 '75.0.3744.0',
357 '74.0.3729.33',
358 '73.0.3683.92',
359 '74.0.3729.32',
360 '74.0.3729.31',
361 '73.0.3683.91',
362 '75.0.3741.2',
363 '75.0.3740.5',
364 '74.0.3729.30',
365 '75.0.3741.1',
366 '75.0.3741.0',
367 '74.0.3729.29',
368 '75.0.3740.4',
369 '73.0.3683.90',
370 '74.0.3729.28',
371 '75.0.3740.3',
372 '73.0.3683.89',
373 '75.0.3740.2',
374 '74.0.3729.27',
375 '75.0.3740.1',
376 '75.0.3740.0',
377 '74.0.3729.26',
378 '73.0.3683.88',
379 '73.0.3683.87',
380 '74.0.3729.25',
381 '75.0.3739.1',
382 '75.0.3739.0',
383 '73.0.3683.86',
384 '74.0.3729.24',
385 '73.0.3683.85',
386 '75.0.3738.4',
387 '75.0.3738.3',
388 '75.0.3738.2',
389 '75.0.3738.1',
390 '75.0.3738.0',
391 '74.0.3729.23',
392 '73.0.3683.84',
393 '74.0.3729.22',
394 '74.0.3729.21',
395 '75.0.3737.1',
396 '75.0.3737.0',
397 '74.0.3729.20',
398 '73.0.3683.83',
399 '74.0.3729.19',
400 '75.0.3736.1',
401 '75.0.3736.0',
402 '74.0.3729.18',
403 '73.0.3683.82',
404 '74.0.3729.17',
405 '75.0.3735.1',
406 '75.0.3735.0',
407 '74.0.3729.16',
408 '73.0.3683.81',
409 '75.0.3734.1',
410 '75.0.3734.0',
411 '74.0.3729.15',
412 '73.0.3683.80',
413 '74.0.3729.14',
414 '75.0.3733.1',
415 '75.0.3733.0',
416 '75.0.3732.1',
417 '74.0.3729.13',
418 '74.0.3729.12',
419 '73.0.3683.79',
420 '74.0.3729.11',
421 '75.0.3732.0',
422 '74.0.3729.10',
423 '73.0.3683.78',
424 '74.0.3729.9',
425 '74.0.3729.8',
426 '74.0.3729.7',
427 '75.0.3731.3',
428 '75.0.3731.2',
429 '75.0.3731.0',
430 '74.0.3729.6',
431 '73.0.3683.77',
432 '73.0.3683.76',
433 '75.0.3730.5',
434 '75.0.3730.4',
435 '73.0.3683.75',
436 '74.0.3729.5',
437 '73.0.3683.74',
438 '75.0.3730.3',
439 '75.0.3730.2',
440 '74.0.3729.4',
441 '73.0.3683.73',
442 '73.0.3683.72',
443 '75.0.3730.1',
444 '75.0.3730.0',
445 '74.0.3729.3',
446 '73.0.3683.71',
447 '74.0.3729.2',
448 '73.0.3683.70',
449 '74.0.3729.1',
450 '74.0.3729.0',
451 '74.0.3726.4',
452 '73.0.3683.69',
453 '74.0.3726.3',
454 '74.0.3728.0',
455 '74.0.3726.2',
456 '73.0.3683.68',
457 '74.0.3726.1',
458 '74.0.3726.0',
459 '74.0.3725.4',
460 '73.0.3683.67',
461 '73.0.3683.66',
462 '74.0.3725.3',
463 '74.0.3725.2',
464 '74.0.3725.1',
465 '74.0.3724.8',
466 '74.0.3725.0',
467 '73.0.3683.65',
468 '74.0.3724.7',
469 '74.0.3724.6',
470 '74.0.3724.5',
471 '74.0.3724.4',
472 '74.0.3724.3',
473 '74.0.3724.2',
474 '74.0.3724.1',
475 '74.0.3724.0',
476 '73.0.3683.64',
477 '74.0.3723.1',
478 '74.0.3723.0',
479 '73.0.3683.63',
480 '74.0.3722.1',
481 '74.0.3722.0',
482 '73.0.3683.62',
483 '74.0.3718.9',
484 '74.0.3702.3',
485 '74.0.3721.3',
486 '74.0.3721.2',
487 '74.0.3721.1',
488 '74.0.3721.0',
489 '74.0.3720.6',
490 '73.0.3683.61',
491 '72.0.3626.122',
492 '73.0.3683.60',
493 '74.0.3720.5',
494 '72.0.3626.121',
495 '74.0.3718.8',
496 '74.0.3720.4',
497 '74.0.3720.3',
498 '74.0.3718.7',
499 '74.0.3720.2',
500 '74.0.3720.1',
501 '74.0.3720.0',
502 '74.0.3718.6',
503 '74.0.3719.5',
504 '73.0.3683.59',
505 '74.0.3718.5',
506 '74.0.3718.4',
507 '74.0.3719.4',
508 '74.0.3719.3',
509 '74.0.3719.2',
510 '74.0.3719.1',
511 '73.0.3683.58',
512 '74.0.3719.0',
513 '73.0.3683.57',
514 '73.0.3683.56',
515 '74.0.3718.3',
516 '73.0.3683.55',
517 '74.0.3718.2',
518 '74.0.3718.1',
519 '74.0.3718.0',
520 '73.0.3683.54',
521 '74.0.3717.2',
522 '73.0.3683.53',
523 '74.0.3717.1',
524 '74.0.3717.0',
525 '73.0.3683.52',
526 '74.0.3716.1',
527 '74.0.3716.0',
528 '73.0.3683.51',
529 '74.0.3715.1',
530 '74.0.3715.0',
531 '73.0.3683.50',
532 '74.0.3711.2',
533 '74.0.3714.2',
534 '74.0.3713.3',
535 '74.0.3714.1',
536 '74.0.3714.0',
537 '73.0.3683.49',
538 '74.0.3713.1',
539 '74.0.3713.0',
540 '72.0.3626.120',
541 '73.0.3683.48',
542 '74.0.3712.2',
543 '74.0.3712.1',
544 '74.0.3712.0',
545 '73.0.3683.47',
546 '72.0.3626.119',
547 '73.0.3683.46',
548 '74.0.3710.2',
549 '72.0.3626.118',
550 '74.0.3711.1',
551 '74.0.3711.0',
552 '73.0.3683.45',
553 '72.0.3626.117',
554 '74.0.3710.1',
555 '74.0.3710.0',
556 '73.0.3683.44',
557 '72.0.3626.116',
558 '74.0.3709.1',
559 '74.0.3709.0',
560 '74.0.3704.9',
561 '73.0.3683.43',
562 '72.0.3626.115',
563 '74.0.3704.8',
564 '74.0.3704.7',
565 '74.0.3708.0',
566 '74.0.3706.7',
567 '74.0.3704.6',
568 '73.0.3683.42',
569 '72.0.3626.114',
570 '74.0.3706.6',
571 '72.0.3626.113',
572 '74.0.3704.5',
573 '74.0.3706.5',
574 '74.0.3706.4',
575 '74.0.3706.3',
576 '74.0.3706.2',
577 '74.0.3706.1',
578 '74.0.3706.0',
579 '73.0.3683.41',
580 '72.0.3626.112',
581 '74.0.3705.1',
582 '74.0.3705.0',
583 '73.0.3683.40',
584 '72.0.3626.111',
585 '73.0.3683.39',
586 '74.0.3704.4',
587 '73.0.3683.38',
588 '74.0.3704.3',
589 '74.0.3704.2',
590 '74.0.3704.1',
591 '74.0.3704.0',
592 '73.0.3683.37',
593 '72.0.3626.110',
594 '72.0.3626.109',
595 '74.0.3703.3',
596 '74.0.3703.2',
597 '73.0.3683.36',
598 '74.0.3703.1',
599 '74.0.3703.0',
600 '73.0.3683.35',
601 '72.0.3626.108',
602 '74.0.3702.2',
603 '74.0.3699.3',
604 '74.0.3702.1',
605 '74.0.3702.0',
606 '73.0.3683.34',
607 '72.0.3626.107',
608 '73.0.3683.33',
609 '74.0.3701.1',
610 '74.0.3701.0',
611 '73.0.3683.32',
612 '73.0.3683.31',
613 '72.0.3626.105',
614 '74.0.3700.1',
615 '74.0.3700.0',
616 '73.0.3683.29',
617 '72.0.3626.103',
618 '74.0.3699.2',
619 '74.0.3699.1',
620 '74.0.3699.0',
621 '73.0.3683.28',
622 '72.0.3626.102',
623 '73.0.3683.27',
624 '73.0.3683.26',
625 '74.0.3698.0',
626 '74.0.3696.2',
627 '72.0.3626.101',
628 '73.0.3683.25',
629 '74.0.3696.1',
630 '74.0.3696.0',
631 '74.0.3694.8',
632 '72.0.3626.100',
633 '74.0.3694.7',
634 '74.0.3694.6',
635 '74.0.3694.5',
636 '74.0.3694.4',
637 '72.0.3626.99',
638 '72.0.3626.98',
639 '74.0.3694.3',
640 '73.0.3683.24',
641 '72.0.3626.97',
642 '72.0.3626.96',
643 '72.0.3626.95',
644 '73.0.3683.23',
645 '72.0.3626.94',
646 '73.0.3683.22',
647 '73.0.3683.21',
648 '72.0.3626.93',
649 '74.0.3694.2',
650 '72.0.3626.92',
651 '74.0.3694.1',
652 '74.0.3694.0',
653 '74.0.3693.6',
654 '73.0.3683.20',
655 '72.0.3626.91',
656 '74.0.3693.5',
657 '74.0.3693.4',
658 '74.0.3693.3',
659 '74.0.3693.2',
660 '73.0.3683.19',
661 '74.0.3693.1',
662 '74.0.3693.0',
663 '73.0.3683.18',
664 '72.0.3626.90',
665 '74.0.3692.1',
666 '74.0.3692.0',
667 '73.0.3683.17',
668 '72.0.3626.89',
669 '74.0.3687.3',
670 '74.0.3691.1',
671 '74.0.3691.0',
672 '73.0.3683.16',
673 '72.0.3626.88',
674 '72.0.3626.87',
675 '73.0.3683.15',
676 '74.0.3690.1',
677 '74.0.3690.0',
678 '73.0.3683.14',
679 '72.0.3626.86',
680 '73.0.3683.13',
681 '73.0.3683.12',
682 '74.0.3689.1',
683 '74.0.3689.0',
684 '73.0.3683.11',
685 '72.0.3626.85',
686 '73.0.3683.10',
687 '72.0.3626.84',
688 '73.0.3683.9',
689 '74.0.3688.1',
690 '74.0.3688.0',
691 '73.0.3683.8',
692 '72.0.3626.83',
693 '74.0.3687.2',
694 '74.0.3687.1',
695 '74.0.3687.0',
696 '73.0.3683.7',
697 '72.0.3626.82',
698 '74.0.3686.4',
699 '72.0.3626.81',
700 '74.0.3686.3',
701 '74.0.3686.2',
702 '74.0.3686.1',
703 '74.0.3686.0',
704 '73.0.3683.6',
705 '72.0.3626.80',
706 '74.0.3685.1',
707 '74.0.3685.0',
708 '73.0.3683.5',
709 '72.0.3626.79',
710 '74.0.3684.1',
711 '74.0.3684.0',
712 '73.0.3683.4',
713 '72.0.3626.78',
714 '72.0.3626.77',
715 '73.0.3683.3',
716 '73.0.3683.2',
717 '72.0.3626.76',
718 '73.0.3683.1',
719 '73.0.3683.0',
720 '72.0.3626.75',
721 '71.0.3578.141',
722 '73.0.3682.1',
723 '73.0.3682.0',
724 '72.0.3626.74',
725 '71.0.3578.140',
726 '73.0.3681.4',
727 '73.0.3681.3',
728 '73.0.3681.2',
729 '73.0.3681.1',
730 '73.0.3681.0',
731 '72.0.3626.73',
732 '71.0.3578.139',
733 '72.0.3626.72',
734 '72.0.3626.71',
735 '73.0.3680.1',
736 '73.0.3680.0',
737 '72.0.3626.70',
738 '71.0.3578.138',
739 '73.0.3678.2',
740 '73.0.3679.1',
741 '73.0.3679.0',
742 '72.0.3626.69',
743 '71.0.3578.137',
744 '73.0.3678.1',
745 '73.0.3678.0',
746 '71.0.3578.136',
747 '73.0.3677.1',
748 '73.0.3677.0',
749 '72.0.3626.68',
750 '72.0.3626.67',
751 '71.0.3578.135',
752 '73.0.3676.1',
753 '73.0.3676.0',
754 '73.0.3674.2',
755 '72.0.3626.66',
756 '71.0.3578.134',
757 '73.0.3674.1',
758 '73.0.3674.0',
759 '72.0.3626.65',
760 '71.0.3578.133',
761 '73.0.3673.2',
762 '73.0.3673.1',
763 '73.0.3673.0',
764 '72.0.3626.64',
765 '71.0.3578.132',
766 '72.0.3626.63',
767 '72.0.3626.62',
768 '72.0.3626.61',
769 '72.0.3626.60',
770 '73.0.3672.1',
771 '73.0.3672.0',
772 '72.0.3626.59',
773 '71.0.3578.131',
774 '73.0.3671.3',
775 '73.0.3671.2',
776 '73.0.3671.1',
777 '73.0.3671.0',
778 '72.0.3626.58',
779 '71.0.3578.130',
780 '73.0.3670.1',
781 '73.0.3670.0',
782 '72.0.3626.57',
783 '71.0.3578.129',
784 '73.0.3669.1',
785 '73.0.3669.0',
786 '72.0.3626.56',
787 '71.0.3578.128',
788 '73.0.3668.2',
789 '73.0.3668.1',
790 '73.0.3668.0',
791 '72.0.3626.55',
792 '71.0.3578.127',
793 '73.0.3667.2',
794 '73.0.3667.1',
795 '73.0.3667.0',
796 '72.0.3626.54',
797 '71.0.3578.126',
798 '73.0.3666.1',
799 '73.0.3666.0',
800 '72.0.3626.53',
801 '71.0.3578.125',
802 '73.0.3665.4',
803 '73.0.3665.3',
804 '72.0.3626.52',
805 '73.0.3665.2',
806 '73.0.3664.4',
807 '73.0.3665.1',
808 '73.0.3665.0',
809 '72.0.3626.51',
810 '71.0.3578.124',
811 '72.0.3626.50',
812 '73.0.3664.3',
813 '73.0.3664.2',
814 '73.0.3664.1',
815 '73.0.3664.0',
816 '73.0.3663.2',
817 '72.0.3626.49',
818 '71.0.3578.123',
819 '73.0.3663.1',
820 '73.0.3663.0',
821 '72.0.3626.48',
822 '71.0.3578.122',
823 '73.0.3662.1',
824 '73.0.3662.0',
825 '72.0.3626.47',
826 '71.0.3578.121',
827 '73.0.3661.1',
828 '72.0.3626.46',
829 '73.0.3661.0',
830 '72.0.3626.45',
831 '71.0.3578.120',
832 '73.0.3660.2',
833 '73.0.3660.1',
834 '73.0.3660.0',
835 '72.0.3626.44',
836 '71.0.3578.119',
837 '73.0.3659.1',
838 '73.0.3659.0',
839 '72.0.3626.43',
840 '71.0.3578.118',
841 '73.0.3658.1',
842 '73.0.3658.0',
843 '72.0.3626.42',
844 '71.0.3578.117',
845 '73.0.3657.1',
846 '73.0.3657.0',
847 '72.0.3626.41',
848 '71.0.3578.116',
849 '73.0.3656.1',
850 '73.0.3656.0',
851 '72.0.3626.40',
852 '71.0.3578.115',
853 '73.0.3655.1',
854 '73.0.3655.0',
855 '72.0.3626.39',
856 '71.0.3578.114',
857 '73.0.3654.1',
858 '73.0.3654.0',
859 '72.0.3626.38',
860 '71.0.3578.113',
861 '73.0.3653.1',
862 '73.0.3653.0',
863 '72.0.3626.37',
864 '71.0.3578.112',
865 '73.0.3652.1',
866 '73.0.3652.0',
867 '72.0.3626.36',
868 '71.0.3578.111',
869 '73.0.3651.1',
870 '73.0.3651.0',
871 '72.0.3626.35',
872 '71.0.3578.110',
873 '73.0.3650.1',
874 '73.0.3650.0',
875 '72.0.3626.34',
876 '71.0.3578.109',
877 '73.0.3649.1',
878 '73.0.3649.0',
879 '72.0.3626.33',
880 '71.0.3578.108',
881 '73.0.3648.2',
882 '73.0.3648.1',
883 '73.0.3648.0',
884 '72.0.3626.32',
885 '71.0.3578.107',
886 '73.0.3647.2',
887 '73.0.3647.1',
888 '73.0.3647.0',
889 '72.0.3626.31',
890 '71.0.3578.106',
891 '73.0.3635.3',
892 '73.0.3646.2',
893 '73.0.3646.1',
894 '73.0.3646.0',
895 '72.0.3626.30',
896 '71.0.3578.105',
897 '72.0.3626.29',
898 '73.0.3645.2',
899 '73.0.3645.1',
900 '73.0.3645.0',
901 '72.0.3626.28',
902 '71.0.3578.104',
903 '72.0.3626.27',
904 '72.0.3626.26',
905 '72.0.3626.25',
906 '72.0.3626.24',
907 '73.0.3644.0',
908 '73.0.3643.2',
909 '72.0.3626.23',
910 '71.0.3578.103',
911 '73.0.3643.1',
912 '73.0.3643.0',
913 '72.0.3626.22',
914 '71.0.3578.102',
915 '73.0.3642.1',
916 '73.0.3642.0',
917 '72.0.3626.21',
918 '71.0.3578.101',
919 '73.0.3641.1',
920 '73.0.3641.0',
921 '72.0.3626.20',
922 '71.0.3578.100',
923 '72.0.3626.19',
924 '73.0.3640.1',
925 '73.0.3640.0',
926 '72.0.3626.18',
927 '73.0.3639.1',
928 '71.0.3578.99',
929 '73.0.3639.0',
930 '72.0.3626.17',
931 '73.0.3638.2',
932 '72.0.3626.16',
933 '73.0.3638.1',
934 '73.0.3638.0',
935 '72.0.3626.15',
936 '71.0.3578.98',
937 '73.0.3635.2',
938 '71.0.3578.97',
939 '73.0.3637.1',
940 '73.0.3637.0',
941 '72.0.3626.14',
942 '71.0.3578.96',
943 '71.0.3578.95',
944 '72.0.3626.13',
945 '71.0.3578.94',
946 '73.0.3636.2',
947 '71.0.3578.93',
948 '73.0.3636.1',
949 '73.0.3636.0',
950 '72.0.3626.12',
951 '71.0.3578.92',
952 '73.0.3635.1',
953 '73.0.3635.0',
954 '72.0.3626.11',
955 '71.0.3578.91',
956 '73.0.3634.2',
957 '73.0.3634.1',
958 '73.0.3634.0',
959 '72.0.3626.10',
960 '71.0.3578.90',
961 '71.0.3578.89',
962 '73.0.3633.2',
963 '73.0.3633.1',
964 '73.0.3633.0',
965 '72.0.3610.4',
966 '72.0.3626.9',
967 '71.0.3578.88',
968 '73.0.3632.5',
969 '73.0.3632.4',
970 '73.0.3632.3',
971 '73.0.3632.2',
972 '73.0.3632.1',
973 '73.0.3632.0',
974 '72.0.3626.8',
975 '71.0.3578.87',
976 '73.0.3631.2',
977 '73.0.3631.1',
978 '73.0.3631.0',
979 '72.0.3626.7',
980 '71.0.3578.86',
981 '72.0.3626.6',
982 '73.0.3630.1',
983 '73.0.3630.0',
984 '72.0.3626.5',
985 '71.0.3578.85',
986 '72.0.3626.4',
987 '73.0.3628.3',
988 '73.0.3628.2',
989 '73.0.3629.1',
990 '73.0.3629.0',
991 '72.0.3626.3',
992 '71.0.3578.84',
993 '73.0.3628.1',
994 '73.0.3628.0',
995 '71.0.3578.83',
996 '73.0.3627.1',
997 '73.0.3627.0',
998 '72.0.3626.2',
999 '71.0.3578.82',
1000 '71.0.3578.81',
1001 '71.0.3578.80',
1002 '72.0.3626.1',
1003 '72.0.3626.0',
1004 '71.0.3578.79',
1005 '70.0.3538.124',
1006 '71.0.3578.78',
1007 '72.0.3623.4',
1008 '72.0.3625.2',
1009 '72.0.3625.1',
1010 '72.0.3625.0',
1011 '71.0.3578.77',
1012 '70.0.3538.123',
1013 '72.0.3624.4',
1014 '72.0.3624.3',
1015 '72.0.3624.2',
1016 '71.0.3578.76',
1017 '72.0.3624.1',
1018 '72.0.3624.0',
1019 '72.0.3623.3',
1020 '71.0.3578.75',
1021 '70.0.3538.122',
1022 '71.0.3578.74',
1023 '72.0.3623.2',
1024 '72.0.3610.3',
1025 '72.0.3623.1',
1026 '72.0.3623.0',
1027 '72.0.3622.3',
1028 '72.0.3622.2',
1029 '71.0.3578.73',
1030 '70.0.3538.121',
1031 '72.0.3622.1',
1032 '72.0.3622.0',
1033 '71.0.3578.72',
1034 '70.0.3538.120',
1035 '72.0.3621.1',
1036 '72.0.3621.0',
1037 '71.0.3578.71',
1038 '70.0.3538.119',
1039 '72.0.3620.1',
1040 '72.0.3620.0',
1041 '71.0.3578.70',
1042 '70.0.3538.118',
1043 '71.0.3578.69',
1044 '72.0.3619.1',
1045 '72.0.3619.0',
1046 '71.0.3578.68',
1047 '70.0.3538.117',
1048 '71.0.3578.67',
1049 '72.0.3618.1',
1050 '72.0.3618.0',
1051 '71.0.3578.66',
1052 '70.0.3538.116',
1053 '72.0.3617.1',
1054 '72.0.3617.0',
1055 '71.0.3578.65',
1056 '70.0.3538.115',
1057 '72.0.3602.3',
1058 '71.0.3578.64',
1059 '72.0.3616.1',
1060 '72.0.3616.0',
1061 '71.0.3578.63',
1062 '70.0.3538.114',
1063 '71.0.3578.62',
1064 '72.0.3615.1',
1065 '72.0.3615.0',
1066 '71.0.3578.61',
1067 '70.0.3538.113',
1068 '72.0.3614.1',
1069 '72.0.3614.0',
1070 '71.0.3578.60',
1071 '70.0.3538.112',
1072 '72.0.3613.1',
1073 '72.0.3613.0',
1074 '71.0.3578.59',
1075 '70.0.3538.111',
1076 '72.0.3612.2',
1077 '72.0.3612.1',
1078 '72.0.3612.0',
1079 '70.0.3538.110',
1080 '71.0.3578.58',
1081 '70.0.3538.109',
1082 '72.0.3611.2',
1083 '72.0.3611.1',
1084 '72.0.3611.0',
1085 '71.0.3578.57',
1086 '70.0.3538.108',
1087 '72.0.3610.2',
1088 '71.0.3578.56',
1089 '71.0.3578.55',
1090 '72.0.3610.1',
1091 '72.0.3610.0',
1092 '71.0.3578.54',
1093 '70.0.3538.107',
1094 '71.0.3578.53',
1095 '72.0.3609.3',
1096 '71.0.3578.52',
1097 '72.0.3609.2',
1098 '71.0.3578.51',
1099 '72.0.3608.5',
1100 '72.0.3609.1',
1101 '72.0.3609.0',
1102 '71.0.3578.50',
1103 '70.0.3538.106',
1104 '72.0.3608.4',
1105 '72.0.3608.3',
1106 '72.0.3608.2',
1107 '71.0.3578.49',
1108 '72.0.3608.1',
1109 '72.0.3608.0',
1110 '70.0.3538.105',
1111 '71.0.3578.48',
1112 '72.0.3607.1',
1113 '72.0.3607.0',
1114 '71.0.3578.47',
1115 '70.0.3538.104',
1116 '72.0.3606.2',
1117 '72.0.3606.1',
1118 '72.0.3606.0',
1119 '71.0.3578.46',
1120 '70.0.3538.103',
1121 '70.0.3538.102',
1122 '72.0.3605.3',
1123 '72.0.3605.2',
1124 '72.0.3605.1',
1125 '72.0.3605.0',
1126 '71.0.3578.45',
1127 '70.0.3538.101',
1128 '71.0.3578.44',
1129 '71.0.3578.43',
1130 '70.0.3538.100',
1131 '70.0.3538.99',
1132 '71.0.3578.42',
1133 '72.0.3604.1',
1134 '72.0.3604.0',
1135 '71.0.3578.41',
1136 '70.0.3538.98',
1137 '71.0.3578.40',
1138 '72.0.3603.2',
1139 '72.0.3603.1',
1140 '72.0.3603.0',
1141 '71.0.3578.39',
1142 '70.0.3538.97',
1143 '72.0.3602.2',
1144 '71.0.3578.38',
1145 '71.0.3578.37',
1146 '72.0.3602.1',
1147 '72.0.3602.0',
1148 '71.0.3578.36',
1149 '70.0.3538.96',
1150 '72.0.3601.1',
1151 '72.0.3601.0',
1152 '71.0.3578.35',
1153 '70.0.3538.95',
1154 '72.0.3600.1',
1155 '72.0.3600.0',
1156 '71.0.3578.34',
1157 '70.0.3538.94',
1158 '72.0.3599.3',
1159 '72.0.3599.2',
1160 '72.0.3599.1',
1161 '72.0.3599.0',
1162 '71.0.3578.33',
1163 '70.0.3538.93',
1164 '72.0.3598.1',
1165 '72.0.3598.0',
1166 '71.0.3578.32',
1167 '70.0.3538.87',
1168 '72.0.3597.1',
1169 '72.0.3597.0',
1170 '72.0.3596.2',
1171 '71.0.3578.31',
1172 '70.0.3538.86',
1173 '71.0.3578.30',
1174 '71.0.3578.29',
1175 '72.0.3596.1',
1176 '72.0.3596.0',
1177 '71.0.3578.28',
1178 '70.0.3538.85',
1179 '72.0.3595.2',
1180 '72.0.3591.3',
1181 '72.0.3595.1',
1182 '72.0.3595.0',
1183 '71.0.3578.27',
1184 '70.0.3538.84',
1185 '72.0.3594.1',
1186 '72.0.3594.0',
1187 '71.0.3578.26',
1188 '70.0.3538.83',
1189 '72.0.3593.2',
1190 '72.0.3593.1',
1191 '72.0.3593.0',
1192 '71.0.3578.25',
1193 '70.0.3538.82',
1194 '72.0.3589.3',
1195 '72.0.3592.2',
1196 '72.0.3592.1',
1197 '72.0.3592.0',
1198 '71.0.3578.24',
1199 '72.0.3589.2',
1200 '70.0.3538.81',
1201 '70.0.3538.80',
1202 '72.0.3591.2',
1203 '72.0.3591.1',
1204 '72.0.3591.0',
1205 '71.0.3578.23',
1206 '70.0.3538.79',
1207 '71.0.3578.22',
1208 '72.0.3590.1',
1209 '72.0.3590.0',
1210 '71.0.3578.21',
1211 '70.0.3538.78',
1212 '70.0.3538.77',
1213 '72.0.3589.1',
1214 '72.0.3589.0',
1215 '71.0.3578.20',
1216 '70.0.3538.76',
1217 '71.0.3578.19',
1218 '70.0.3538.75',
1219 '72.0.3588.1',
1220 '72.0.3588.0',
1221 '71.0.3578.18',
1222 '70.0.3538.74',
1223 '72.0.3586.2',
1224 '72.0.3587.0',
1225 '71.0.3578.17',
1226 '70.0.3538.73',
1227 '72.0.3586.1',
1228 '72.0.3586.0',
1229 '71.0.3578.16',
1230 '70.0.3538.72',
1231 '72.0.3585.1',
1232 '72.0.3585.0',
1233 '71.0.3578.15',
1234 '70.0.3538.71',
1235 '71.0.3578.14',
1236 '72.0.3584.1',
1237 '72.0.3584.0',
1238 '71.0.3578.13',
1239 '70.0.3538.70',
1240 '72.0.3583.2',
1241 '71.0.3578.12',
1242 '72.0.3583.1',
1243 '72.0.3583.0',
1244 '71.0.3578.11',
1245 '70.0.3538.69',
1246 '71.0.3578.10',
1247 '72.0.3582.0',
1248 '72.0.3581.4',
1249 '71.0.3578.9',
1250 '70.0.3538.67',
1251 '72.0.3581.3',
1252 '72.0.3581.2',
1253 '72.0.3581.1',
1254 '72.0.3581.0',
1255 '71.0.3578.8',
1256 '70.0.3538.66',
1257 '72.0.3580.1',
1258 '72.0.3580.0',
1259 '71.0.3578.7',
1260 '70.0.3538.65',
1261 '71.0.3578.6',
1262 '72.0.3579.1',
1263 '72.0.3579.0',
1264 '71.0.3578.5',
1265 '70.0.3538.64',
1266 '71.0.3578.4',
1267 '71.0.3578.3',
1268 '71.0.3578.2',
1269 '71.0.3578.1',
1270 '71.0.3578.0',
1271 '70.0.3538.63',
1272 '69.0.3497.128',
1273 '70.0.3538.62',
1274 '70.0.3538.61',
1275 '70.0.3538.60',
1276 '70.0.3538.59',
1277 '71.0.3577.1',
1278 '71.0.3577.0',
1279 '70.0.3538.58',
1280 '69.0.3497.127',
1281 '71.0.3576.2',
1282 '71.0.3576.1',
1283 '71.0.3576.0',
1284 '70.0.3538.57',
1285 '70.0.3538.56',
1286 '71.0.3575.2',
1287 '70.0.3538.55',
1288 '69.0.3497.126',
1289 '70.0.3538.54',
1290 '71.0.3575.1',
1291 '71.0.3575.0',
1292 '71.0.3574.1',
1293 '71.0.3574.0',
1294 '70.0.3538.53',
1295 '69.0.3497.125',
1296 '70.0.3538.52',
1297 '71.0.3573.1',
1298 '71.0.3573.0',
1299 '70.0.3538.51',
1300 '69.0.3497.124',
1301 '71.0.3572.1',
1302 '71.0.3572.0',
1303 '70.0.3538.50',
1304 '69.0.3497.123',
1305 '71.0.3571.2',
1306 '70.0.3538.49',
1307 '69.0.3497.122',
1308 '71.0.3571.1',
1309 '71.0.3571.0',
1310 '70.0.3538.48',
1311 '69.0.3497.121',
1312 '71.0.3570.1',
1313 '71.0.3570.0',
1314 '70.0.3538.47',
1315 '69.0.3497.120',
1316 '71.0.3568.2',
1317 '71.0.3569.1',
1318 '71.0.3569.0',
1319 '70.0.3538.46',
1320 '69.0.3497.119',
1321 '70.0.3538.45',
1322 '71.0.3568.1',
1323 '71.0.3568.0',
1324 '70.0.3538.44',
1325 '69.0.3497.118',
1326 '70.0.3538.43',
1327 '70.0.3538.42',
1328 '71.0.3567.1',
1329 '71.0.3567.0',
1330 '70.0.3538.41',
1331 '69.0.3497.117',
1332 '71.0.3566.1',
1333 '71.0.3566.0',
1334 '70.0.3538.40',
1335 '69.0.3497.116',
1336 '71.0.3565.1',
1337 '71.0.3565.0',
1338 '70.0.3538.39',
1339 '69.0.3497.115',
1340 '71.0.3564.1',
1341 '71.0.3564.0',
1342 '70.0.3538.38',
1343 '69.0.3497.114',
1344 '71.0.3563.0',
1345 '71.0.3562.2',
1346 '70.0.3538.37',
1347 '69.0.3497.113',
1348 '70.0.3538.36',
1349 '70.0.3538.35',
1350 '71.0.3562.1',
1351 '71.0.3562.0',
1352 '70.0.3538.34',
1353 '69.0.3497.112',
1354 '70.0.3538.33',
1355 '71.0.3561.1',
1356 '71.0.3561.0',
1357 '70.0.3538.32',
1358 '69.0.3497.111',
1359 '71.0.3559.6',
1360 '71.0.3560.1',
1361 '71.0.3560.0',
1362 '71.0.3559.5',
1363 '71.0.3559.4',
1364 '70.0.3538.31',
1365 '69.0.3497.110',
1366 '71.0.3559.3',
1367 '70.0.3538.30',
1368 '69.0.3497.109',
1369 '71.0.3559.2',
1370 '71.0.3559.1',
1371 '71.0.3559.0',
1372 '70.0.3538.29',
1373 '69.0.3497.108',
1374 '71.0.3558.2',
1375 '71.0.3558.1',
1376 '71.0.3558.0',
1377 '70.0.3538.28',
1378 '69.0.3497.107',
1379 '71.0.3557.2',
1380 '71.0.3557.1',
1381 '71.0.3557.0',
1382 '70.0.3538.27',
1383 '69.0.3497.106',
1384 '71.0.3554.4',
1385 '70.0.3538.26',
1386 '71.0.3556.1',
1387 '71.0.3556.0',
1388 '70.0.3538.25',
1389 '71.0.3554.3',
1390 '69.0.3497.105',
1391 '71.0.3554.2',
1392 '70.0.3538.24',
1393 '69.0.3497.104',
1394 '71.0.3555.2',
1395 '70.0.3538.23',
1396 '71.0.3555.1',
1397 '71.0.3555.0',
1398 '70.0.3538.22',
1399 '69.0.3497.103',
1400 '71.0.3554.1',
1401 '71.0.3554.0',
1402 '70.0.3538.21',
1403 '69.0.3497.102',
1404 '71.0.3553.3',
1405 '70.0.3538.20',
1406 '69.0.3497.101',
1407 '71.0.3553.2',
1408 '69.0.3497.100',
1409 '71.0.3553.1',
1410 '71.0.3553.0',
1411 '70.0.3538.19',
1412 '69.0.3497.99',
1413 '69.0.3497.98',
1414 '69.0.3497.97',
1415 '71.0.3552.6',
1416 '71.0.3552.5',
1417 '71.0.3552.4',
1418 '71.0.3552.3',
1419 '71.0.3552.2',
1420 '71.0.3552.1',
1421 '71.0.3552.0',
1422 '70.0.3538.18',
1423 '69.0.3497.96',
1424 '71.0.3551.3',
1425 '71.0.3551.2',
1426 '71.0.3551.1',
1427 '71.0.3551.0',
1428 '70.0.3538.17',
1429 '69.0.3497.95',
1430 '71.0.3550.3',
1431 '71.0.3550.2',
1432 '71.0.3550.1',
1433 '71.0.3550.0',
1434 '70.0.3538.16',
1435 '69.0.3497.94',
1436 '71.0.3549.1',
1437 '71.0.3549.0',
1438 '70.0.3538.15',
1439 '69.0.3497.93',
1440 '69.0.3497.92',
1441 '71.0.3548.1',
1442 '71.0.3548.0',
1443 '70.0.3538.14',
1444 '69.0.3497.91',
1445 '71.0.3547.1',
1446 '71.0.3547.0',
1447 '70.0.3538.13',
1448 '69.0.3497.90',
1449 '71.0.3546.2',
1450 '69.0.3497.89',
1451 '71.0.3546.1',
1452 '71.0.3546.0',
1453 '70.0.3538.12',
1454 '69.0.3497.88',
1455 '71.0.3545.4',
1456 '71.0.3545.3',
1457 '71.0.3545.2',
1458 '71.0.3545.1',
1459 '71.0.3545.0',
1460 '70.0.3538.11',
1461 '69.0.3497.87',
1462 '71.0.3544.5',
1463 '71.0.3544.4',
1464 '71.0.3544.3',
1465 '71.0.3544.2',
1466 '71.0.3544.1',
1467 '71.0.3544.0',
1468 '69.0.3497.86',
1469 '70.0.3538.10',
1470 '69.0.3497.85',
1471 '70.0.3538.9',
1472 '69.0.3497.84',
1473 '71.0.3543.4',
1474 '70.0.3538.8',
1475 '71.0.3543.3',
1476 '71.0.3543.2',
1477 '71.0.3543.1',
1478 '71.0.3543.0',
1479 '70.0.3538.7',
1480 '69.0.3497.83',
1481 '71.0.3542.2',
1482 '71.0.3542.1',
1483 '71.0.3542.0',
1484 '70.0.3538.6',
1485 '69.0.3497.82',
1486 '69.0.3497.81',
1487 '71.0.3541.1',
1488 '71.0.3541.0',
1489 '70.0.3538.5',
1490 '69.0.3497.80',
1491 '71.0.3540.1',
1492 '71.0.3540.0',
1493 '70.0.3538.4',
1494 '69.0.3497.79',
1495 '70.0.3538.3',
1496 '71.0.3539.1',
1497 '71.0.3539.0',
1498 '69.0.3497.78',
1499 '68.0.3440.134',
1500 '69.0.3497.77',
1501 '70.0.3538.2',
1502 '70.0.3538.1',
1503 '70.0.3538.0',
1504 '69.0.3497.76',
1505 '68.0.3440.133',
1506 '69.0.3497.75',
1507 '70.0.3537.2',
1508 '70.0.3537.1',
1509 '70.0.3537.0',
1510 '69.0.3497.74',
1511 '68.0.3440.132',
1512 '70.0.3536.0',
1513 '70.0.3535.5',
1514 '70.0.3535.4',
1515 '70.0.3535.3',
1516 '69.0.3497.73',
1517 '68.0.3440.131',
1518 '70.0.3532.8',
1519 '70.0.3532.7',
1520 '69.0.3497.72',
1521 '69.0.3497.71',
1522 '70.0.3535.2',
1523 '70.0.3535.1',
1524 '70.0.3535.0',
1525 '69.0.3497.70',
1526 '68.0.3440.130',
1527 '69.0.3497.69',
1528 '68.0.3440.129',
1529 '70.0.3534.4',
1530 '70.0.3534.3',
1531 '70.0.3534.2',
1532 '70.0.3534.1',
1533 '70.0.3534.0',
1534 '69.0.3497.68',
1535 '68.0.3440.128',
1536 '70.0.3533.2',
1537 '70.0.3533.1',
1538 '70.0.3533.0',
1539 '69.0.3497.67',
1540 '68.0.3440.127',
1541 '70.0.3532.6',
1542 '70.0.3532.5',
1543 '70.0.3532.4',
1544 '69.0.3497.66',
1545 '68.0.3440.126',
1546 '70.0.3532.3',
1547 '70.0.3532.2',
1548 '70.0.3532.1',
1549 '69.0.3497.60',
1550 '69.0.3497.65',
1551 '69.0.3497.64',
1552 '70.0.3532.0',
1553 '70.0.3531.0',
1554 '70.0.3530.4',
1555 '70.0.3530.3',
1556 '70.0.3530.2',
1557 '69.0.3497.58',
1558 '68.0.3440.125',
1559 '69.0.3497.57',
1560 '69.0.3497.56',
1561 '69.0.3497.55',
1562 '69.0.3497.54',
1563 '70.0.3530.1',
1564 '70.0.3530.0',
1565 '69.0.3497.53',
1566 '68.0.3440.124',
1567 '69.0.3497.52',
1568 '70.0.3529.3',
1569 '70.0.3529.2',
1570 '70.0.3529.1',
1571 '70.0.3529.0',
1572 '69.0.3497.51',
1573 '70.0.3528.4',
1574 '68.0.3440.123',
1575 '70.0.3528.3',
1576 '70.0.3528.2',
1577 '70.0.3528.1',
1578 '70.0.3528.0',
1579 '69.0.3497.50',
1580 '68.0.3440.122',
1581 '70.0.3527.1',
1582 '70.0.3527.0',
1583 '69.0.3497.49',
1584 '68.0.3440.121',
1585 '70.0.3526.1',
1586 '70.0.3526.0',
1587 '68.0.3440.120',
1588 '69.0.3497.48',
1589 '69.0.3497.47',
1590 '68.0.3440.119',
1591 '68.0.3440.118',
1592 '70.0.3525.5',
1593 '70.0.3525.4',
1594 '70.0.3525.3',
1595 '68.0.3440.117',
1596 '69.0.3497.46',
1597 '70.0.3525.2',
1598 '70.0.3525.1',
1599 '70.0.3525.0',
1600 '69.0.3497.45',
1601 '68.0.3440.116',
1602 '70.0.3524.4',
1603 '70.0.3524.3',
1604 '69.0.3497.44',
1605 '70.0.3524.2',
1606 '70.0.3524.1',
1607 '70.0.3524.0',
1608 '70.0.3523.2',
1609 '69.0.3497.43',
1610 '68.0.3440.115',
1611 '70.0.3505.9',
1612 '69.0.3497.42',
1613 '70.0.3505.8',
1614 '70.0.3523.1',
1615 '70.0.3523.0',
1616 '69.0.3497.41',
1617 '68.0.3440.114',
1618 '70.0.3505.7',
1619 '69.0.3497.40',
1620 '70.0.3522.1',
1621 '70.0.3522.0',
1622 '70.0.3521.2',
1623 '69.0.3497.39',
1624 '68.0.3440.113',
1625 '70.0.3505.6',
1626 '70.0.3521.1',
1627 '70.0.3521.0',
1628 '69.0.3497.38',
1629 '68.0.3440.112',
1630 '70.0.3520.1',
1631 '70.0.3520.0',
1632 '69.0.3497.37',
1633 '68.0.3440.111',
1634 '70.0.3519.3',
1635 '70.0.3519.2',
1636 '70.0.3519.1',
1637 '70.0.3519.0',
1638 '69.0.3497.36',
1639 '68.0.3440.110',
1640 '70.0.3518.1',
1641 '70.0.3518.0',
1642 '69.0.3497.35',
1643 '69.0.3497.34',
1644 '68.0.3440.109',
1645 '70.0.3517.1',
1646 '70.0.3517.0',
1647 '69.0.3497.33',
1648 '68.0.3440.108',
1649 '69.0.3497.32',
1650 '70.0.3516.3',
1651 '70.0.3516.2',
1652 '70.0.3516.1',
1653 '70.0.3516.0',
1654 '69.0.3497.31',
1655 '68.0.3440.107',
1656 '70.0.3515.4',
1657 '68.0.3440.106',
1658 '70.0.3515.3',
1659 '70.0.3515.2',
1660 '70.0.3515.1',
1661 '70.0.3515.0',
1662 '69.0.3497.30',
1663 '68.0.3440.105',
1664 '68.0.3440.104',
1665 '70.0.3514.2',
1666 '70.0.3514.1',
1667 '70.0.3514.0',
1668 '69.0.3497.29',
1669 '68.0.3440.103',
1670 '70.0.3513.1',
1671 '70.0.3513.0',
1672 '69.0.3497.28',
1673 )
1674 return _USER_AGENT_TPL % random.choice(_CHROME_VERSIONS)
1675
1676
1677std_headers = {
1678 'User-Agent': random_user_agent(),
1679 'Accept-Charset': 'ISO-8859-1,utf-8;q=0.7,*;q=0.7',
1680 'Accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8',
1681 'Accept-Encoding': 'gzip, deflate',
1682 'Accept-Language': 'en-us,en;q=0.5',
1683}
1684
1685
1686USER_AGENTS = {
1687 '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',
1688}
1689
1690
1691NO_DEFAULT = object()
1692
1693ENGLISH_MONTH_NAMES = [
1694 'January', 'February', 'March', 'April', 'May', 'June',
1695 'July', 'August', 'September', 'October', 'November', 'December']
1696
1697MONTH_NAMES = {
1698 'en': ENGLISH_MONTH_NAMES,
1699 'fr': [
1700 'janvier', 'février', 'mars', 'avril', 'mai', 'juin',
1701 'juillet', 'août', 'septembre', 'octobre', 'novembre', 'décembre'],
1702}
1703
1704KNOWN_EXTENSIONS = (
1705 'mp4', 'm4a', 'm4p', 'm4b', 'm4r', 'm4v', 'aac',
1706 'flv', 'f4v', 'f4a', 'f4b',
1707 'webm', 'ogg', 'ogv', 'oga', 'ogx', 'spx', 'opus',
1708 'mkv', 'mka', 'mk3d',
1709 'avi', 'divx',
1710 'mov',
1711 'asf', 'wmv', 'wma',
1712 '3gp', '3g2',
1713 'mp3',
1714 'flac',
1715 'ape',
1716 'wav',
1717 'f4f', 'f4m', 'm3u8', 'smil')
1718
1719# needed for sanitizing filenames in restricted mode
1720ACCENT_CHARS = dict(zip('ÂÃÄÀÁÅÆÇÈÉÊËÌÍÎÏÐÑÒÓÔÕÖŐØŒÙÚÛÜŰÝÞßàáâãäåæçèéêëìíîïðñòóôõöőøœùúûüűýþÿ',
1721 itertools.chain('AAAAAA', ['AE'], 'CEEEEIIIIDNOOOOOOO', ['OE'], 'UUUUUY', ['TH', 'ss'],
1722 'aaaaaa', ['ae'], 'ceeeeiiiionooooooo', ['oe'], 'uuuuuy', ['th'], 'y')))
1723
1724DATE_FORMATS = (
1725 '%d %B %Y',
1726 '%d %b %Y',
1727 '%B %d %Y',
1728 '%B %dst %Y',
1729 '%B %dnd %Y',
1730 '%B %drd %Y',
1731 '%B %dth %Y',
1732 '%b %d %Y',
1733 '%b %dst %Y',
1734 '%b %dnd %Y',
1735 '%b %drd %Y',
1736 '%b %dth %Y',
1737 '%b %dst %Y %I:%M',
1738 '%b %dnd %Y %I:%M',
1739 '%b %drd %Y %I:%M',
1740 '%b %dth %Y %I:%M',
1741 '%Y %m %d',
1742 '%Y-%m-%d',
1743 '%Y/%m/%d',
1744 '%Y/%m/%d %H:%M',
1745 '%Y/%m/%d %H:%M:%S',
1746 '%Y-%m-%d %H:%M',
1747 '%Y-%m-%d %H:%M:%S',
1748 '%Y-%m-%d %H:%M:%S.%f',
1749 '%Y-%m-%d %H:%M:%S:%f',
1750 '%d.%m.%Y %H:%M',
1751 '%d.%m.%Y %H.%M',
1752 '%Y-%m-%dT%H:%M:%SZ',
1753 '%Y-%m-%dT%H:%M:%S.%fZ',
1754 '%Y-%m-%dT%H:%M:%S.%f0Z',
1755 '%Y-%m-%dT%H:%M:%S',
1756 '%Y-%m-%dT%H:%M:%S.%f',
1757 '%Y-%m-%dT%H:%M',
1758 '%b %d %Y at %H:%M',
1759 '%b %d %Y at %H:%M:%S',
1760 '%B %d %Y at %H:%M',
1761 '%B %d %Y at %H:%M:%S',
1762)
1763
1764DATE_FORMATS_DAY_FIRST = list(DATE_FORMATS)
1765DATE_FORMATS_DAY_FIRST.extend([
1766 '%d-%m-%Y',
1767 '%d.%m.%Y',
1768 '%d.%m.%y',
1769 '%d/%m/%Y',
1770 '%d/%m/%y',
1771 '%d/%m/%Y %H:%M:%S',
1772])
1773
1774DATE_FORMATS_MONTH_FIRST = list(DATE_FORMATS)
1775DATE_FORMATS_MONTH_FIRST.extend([
1776 '%m-%d-%Y',
1777 '%m.%d.%Y',
1778 '%m/%d/%Y',
1779 '%m/%d/%y',
1780 '%m/%d/%Y %H:%M:%S',
1781])
1782
1783PACKED_CODES_RE = r"}\('(.+)',(\d+),(\d+),'([^']+)'\.split\('\|'\)"
1784JSON_LD_RE = r'(?is)<script[^>]+type=(["\']?)application/ld\+json\1[^>]*>(?P<json_ld>.+?)</script>'
1785
1786
1787def preferredencoding():
1788 """Get preferred encoding.
1789
1790 Returns the best encoding scheme for the system, based on
1791 locale.getpreferredencoding() and some further tweaks.
1792 """
1793 try:
1794 pref = locale.getpreferredencoding()
1795 'TEST'.encode(pref)
1796 except Exception:
1797 pref = 'UTF-8'
1798
1799 return pref
1800
1801
1802def write_json_file(obj, fn):
1803 """ Encode obj as JSON and write it to fn, atomically if possible """
1804
1805 fn = encodeFilename(fn)
1806 if sys.version_info < (3, 0) and sys.platform != 'win32':
1807 encoding = get_filesystem_encoding()
1808 # os.path.basename returns a bytes object, but NamedTemporaryFile
1809 # will fail if the filename contains non ascii characters unless we
1810 # use a unicode object
1811 path_basename = lambda f: os.path.basename(fn).decode(encoding)
1812 # the same for os.path.dirname
1813 path_dirname = lambda f: os.path.dirname(fn).decode(encoding)
1814 else:
1815 path_basename = os.path.basename
1816 path_dirname = os.path.dirname
1817
1818 args = {
1819 'suffix': '.tmp',
1820 'prefix': path_basename(fn) + '.',
1821 'dir': path_dirname(fn),
1822 'delete': False,
1823 }
1824
1825 # In Python 2.x, json.dump expects a bytestream.
1826 # In Python 3.x, it writes to a character stream
1827 if sys.version_info < (3, 0):
1828 args['mode'] = 'wb'
1829 else:
1830 args.update({
1831 'mode': 'w',
1832 'encoding': 'utf-8',
1833 })
1834
1835 tf = tempfile.NamedTemporaryFile(**compat_kwargs(args))
1836
1837 try:
1838 with tf:
1839 json.dump(obj, tf, default=repr)
1840 if sys.platform == 'win32':
1841 # Need to remove existing file on Windows, else os.rename raises
1842 # WindowsError or FileExistsError.
1843 try:
1844 os.unlink(fn)
1845 except OSError:
1846 pass
1847 try:
1848 mask = os.umask(0)
1849 os.umask(mask)
1850 os.chmod(tf.name, 0o666 & ~mask)
1851 except OSError:
1852 pass
1853 os.rename(tf.name, fn)
1854 except Exception:
1855 try:
1856 os.remove(tf.name)
1857 except OSError:
1858 pass
1859 raise
1860
1861
1862if sys.version_info >= (2, 7):
1863 def find_xpath_attr(node, xpath, key, val=None):
1864 """ Find the xpath xpath[@key=val] """
1865 assert re.match(r'^[a-zA-Z_-]+$', key)
1866 expr = xpath + ('[@%s]' % key if val is None else "[@%s='%s']" % (key, val))
1867 return node.find(expr)
1868else:
1869 def find_xpath_attr(node, xpath, key, val=None):
1870 for f in node.findall(compat_xpath(xpath)):
1871 if key not in f.attrib:
1872 continue
1873 if val is None or f.attrib.get(key) == val:
1874 return f
1875 return None
1876
1877# On python2.6 the xml.etree.ElementTree.Element methods don't support
1878# the namespace parameter
1879
1880
1881def xpath_with_ns(path, ns_map):
1882 components = [c.split(':') for c in path.split('/')]
1883 replaced = []
1884 for c in components:
1885 if len(c) == 1:
1886 replaced.append(c[0])
1887 else:
1888 ns, tag = c
1889 replaced.append('{%s}%s' % (ns_map[ns], tag))
1890 return '/'.join(replaced)
1891
1892
1893def xpath_element(node, xpath, name=None, fatal=False, default=NO_DEFAULT):
1894 def _find_xpath(xpath):
1895 return node.find(compat_xpath(xpath))
1896
1897 if isinstance(xpath, (str, compat_str)):
1898 n = _find_xpath(xpath)
1899 else:
1900 for xp in xpath:
1901 n = _find_xpath(xp)
1902 if n is not None:
1903 break
1904
1905 if n 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' % name)
1911 else:
1912 return None
1913 return n
1914
1915
1916def xpath_text(node, xpath, name=None, fatal=False, default=NO_DEFAULT):
1917 n = xpath_element(node, xpath, name, fatal=fatal, default=default)
1918 if n is None or n == default:
1919 return n
1920 if n.text is None:
1921 if default is not NO_DEFAULT:
1922 return default
1923 elif fatal:
1924 name = xpath if name is None else name
1925 raise ExtractorError('Could not find XML element\'s text %s' % name)
1926 else:
1927 return None
1928 return n.text
1929
1930
1931def xpath_attr(node, xpath, key, name=None, fatal=False, default=NO_DEFAULT):
1932 n = find_xpath_attr(node, xpath, key)
1933 if n is None:
1934 if default is not NO_DEFAULT:
1935 return default
1936 elif fatal:
1937 name = '%s[@%s]' % (xpath, key) if name is None else name
1938 raise ExtractorError('Could not find XML attribute %s' % name)
1939 else:
1940 return None
1941 return n.attrib[key]
1942
1943
1944def get_element_by_id(id, html):
1945 """Return the content of the tag with the specified ID in the passed HTML document"""
1946 return get_element_by_attribute('id', id, html)
1947
1948
1949def get_element_by_class(class_name, html):
1950 """Return the content of the first tag with the specified class in the passed HTML document"""
1951 retval = get_elements_by_class(class_name, html)
1952 return retval[0] if retval else None
1953
1954
1955def get_element_by_attribute(attribute, value, html, escape_value=True):
1956 retval = get_elements_by_attribute(attribute, value, html, escape_value)
1957 return retval[0] if retval else None
1958
1959
1960def get_elements_by_class(class_name, html):
1961 """Return the content of all tags with the specified class in the passed HTML document as a list"""
1962 return get_elements_by_attribute(
1963 'class', r'[^\'"]*\b%s\b[^\'"]*' % re.escape(class_name),
1964 html, escape_value=False)
1965
1966
1967def get_elements_by_attribute(attribute, value, html, escape_value=True):
1968 """Return the content of the tag with the specified attribute in the passed HTML document"""
1969
1970 value = re.escape(value) if escape_value else value
1971
1972 retlist = []
1973 for m in re.finditer(r'''(?xs)
1974 <([a-zA-Z0-9:._-]+)
1975 (?:\s+[a-zA-Z0-9:._-]+(?:=[a-zA-Z0-9:._-]*|="[^"]*"|='[^']*'|))*?
1976 \s+%s=['"]?%s['"]?
1977 (?:\s+[a-zA-Z0-9:._-]+(?:=[a-zA-Z0-9:._-]*|="[^"]*"|='[^']*'|))*?
1978 \s*>
1979 (?P<content>.*?)
1980 </\1>
1981 ''' % (re.escape(attribute), value), html):
1982 res = m.group('content')
1983
1984 if res.startswith('"') or res.startswith("'"):
1985 res = res[1:-1]
1986
1987 retlist.append(unescapeHTML(res))
1988
1989 return retlist
1990
1991
1992class HTMLAttributeParser(compat_HTMLParser):
1993 """Trivial HTML parser to gather the attributes for a single element"""
1994
1995 def __init__(self):
1996 self.attrs = {}
1997 compat_HTMLParser.__init__(self)
1998
1999 def handle_starttag(self, tag, attrs):
2000 self.attrs = dict(attrs)
2001
2002
2003def extract_attributes(html_element):
2004 """Given a string for an HTML element such as
2005 <el
2006 a="foo" B="bar" c="&98;az" d=boz
2007 empty= noval entity="&amp;"
2008 sq='"' dq="'"
2009 >
2010 Decode and return a dictionary of attributes.
2011 {
2012 'a': 'foo', 'b': 'bar', c: 'baz', d: 'boz',
2013 'empty': '', 'noval': None, 'entity': '&',
2014 'sq': '"', 'dq': '\''
2015 }.
2016 NB HTMLParser is stricter in Python 2.6 & 3.2 than in later versions,
2017 but the cases in the unit test will work for all of 2.6, 2.7, 3.2-3.5.
2018 """
2019 parser = HTMLAttributeParser()
2020 try:
2021 parser.feed(html_element)
2022 parser.close()
2023 # Older Python may throw HTMLParseError in case of malformed HTML
2024 except compat_HTMLParseError:
2025 pass
2026 return parser.attrs
2027
2028
2029def clean_html(html):
2030 """Clean an HTML snippet into a readable string"""
2031
2032 if html is None: # Convenience for sanitizing descriptions etc.
2033 return html
2034
2035 # Newline vs <br />
2036 html = html.replace('\n', ' ')
2037 html = re.sub(r'(?u)\s*<\s*br\s*/?\s*>\s*', '\n', html)
2038 html = re.sub(r'(?u)<\s*/\s*p\s*>\s*<\s*p[^>]*>', '\n', html)
2039 # Strip html tags
2040 html = re.sub('<.*?>', '', html)
2041 # Replace html entities
2042 html = unescapeHTML(html)
2043 return html.strip()
2044
2045
2046def sanitize_open(filename, open_mode):
2047 """Try to open the given filename, and slightly tweak it if this fails.
2048
2049 Attempts to open the given filename. If this fails, it tries to change
2050 the filename slightly, step by step, until it's either able to open it
2051 or it fails and raises a final exception, like the standard open()
2052 function.
2053
2054 It returns the tuple (stream, definitive_file_name).
2055 """
2056 try:
2057 if filename == '-':
2058 if sys.platform == 'win32':
2059 import msvcrt
2060 msvcrt.setmode(sys.stdout.fileno(), os.O_BINARY)
2061 return (sys.stdout.buffer if hasattr(sys.stdout, 'buffer') else sys.stdout, filename)
2062 stream = open(encodeFilename(filename), open_mode)
2063 return (stream, filename)
2064 except (IOError, OSError) as err:
2065 if err.errno in (errno.EACCES,):
2066 raise
2067
2068 # In case of error, try to remove win32 forbidden chars
2069 alt_filename = sanitize_path(filename)
2070 if alt_filename == filename:
2071 raise
2072 else:
2073 # An exception here should be caught in the caller
2074 stream = open(encodeFilename(alt_filename), open_mode)
2075 return (stream, alt_filename)
2076
2077
2078def timeconvert(timestr):
2079 """Convert RFC 2822 defined time string into system timestamp"""
2080 timestamp = None
2081 timetuple = email.utils.parsedate_tz(timestr)
2082 if timetuple is not None:
2083 timestamp = email.utils.mktime_tz(timetuple)
2084 return timestamp
2085
2086
2087def sanitize_filename(s, restricted=False, is_id=False):
2088 """Sanitizes a string so it could be used as part of a filename.
2089 If restricted is set, use a stricter subset of allowed characters.
2090 Set is_id if this is not an arbitrary string, but an ID that should be kept
2091 if possible.
2092 """
2093 def replace_insane(char):
2094 if restricted and char in ACCENT_CHARS:
2095 return ACCENT_CHARS[char]
2096 if char == '?' or ord(char) < 32 or ord(char) == 127:
2097 return ''
2098 elif char == '"':
2099 return '' if restricted else '\''
2100 elif char == ':':
2101 return '_-' if restricted else ' -'
2102 elif char in '\\/|*<>':
2103 return '_'
2104 if restricted and (char in '!&\'()[]{}$;`^,#' or char.isspace()):
2105 return '_'
2106 if restricted and ord(char) > 127:
2107 return '_'
2108 return char
2109
2110 if s == '':
2111 return ''
2112 # Handle timestamps
2113 s = re.sub(r'[0-9]+(?::[0-9]+)+', lambda m: m.group(0).replace(':', '_'), s)
2114 result = ''.join(map(replace_insane, s))
2115 if not is_id:
2116 while '__' in result:
2117 result = result.replace('__', '_')
2118 result = result.strip('_')
2119 # Common case of "Foreign band name - English song title"
2120 if restricted and result.startswith('-_'):
2121 result = result[2:]
2122 if result.startswith('-'):
2123 result = '_' + result[len('-'):]
2124 result = result.lstrip('.')
2125 if not result:
2126 result = '_'
2127 return result
2128
2129
2130def sanitize_path(s, force=False):
2131 """Sanitizes and normalizes path on Windows"""
2132 if sys.platform == 'win32':
2133 force = False
2134 drive_or_unc, _ = os.path.splitdrive(s)
2135 if sys.version_info < (2, 7) and not drive_or_unc:
2136 drive_or_unc, _ = os.path.splitunc(s)
2137 elif force:
2138 drive_or_unc = ''
2139 else:
2140 return s
2141
2142 norm_path = os.path.normpath(remove_start(s, drive_or_unc)).split(os.path.sep)
2143 if drive_or_unc:
2144 norm_path.pop(0)
2145 sanitized_path = [
2146 path_part if path_part in ['.', '..'] else re.sub(r'(?:[/<>:"\|\\?\*]|[\s.]$)', '#', path_part)
2147 for path_part in norm_path]
2148 if drive_or_unc:
2149 sanitized_path.insert(0, drive_or_unc + os.path.sep)
2150 elif force and s[0] == os.path.sep:
2151 sanitized_path.insert(0, os.path.sep)
2152 return os.path.join(*sanitized_path)
2153
2154
2155def sanitize_url(url):
2156 # Prepend protocol-less URLs with `http:` scheme in order to mitigate
2157 # the number of unwanted failures due to missing protocol
2158 if url.startswith('//'):
2159 return 'http:%s' % url
2160 # Fix some common typos seen so far
2161 COMMON_TYPOS = (
2162 # https://github.com/ytdl-org/youtube-dl/issues/15649
2163 (r'^httpss://', r'https://'),
2164 # https://bx1.be/lives/direct-tv/
2165 (r'^rmtp([es]?)://', r'rtmp\1://'),
2166 )
2167 for mistake, fixup in COMMON_TYPOS:
2168 if re.match(mistake, url):
2169 return re.sub(mistake, fixup, url)
2170 return url
2171
2172
2173def extract_basic_auth(url):
2174 parts = compat_urlparse.urlsplit(url)
2175 if parts.username is None:
2176 return url, None
2177 url = compat_urlparse.urlunsplit(parts._replace(netloc=(
2178 parts.hostname if parts.port is None
2179 else '%s:%d' % (parts.hostname, parts.port))))
2180 auth_payload = base64.b64encode(
2181 ('%s:%s' % (parts.username, parts.password or '')).encode('utf-8'))
2182 return url, 'Basic ' + auth_payload.decode('utf-8')
2183
2184
2185def sanitized_Request(url, *args, **kwargs):
2186 url, auth_header = extract_basic_auth(escape_url(sanitize_url(url)))
2187 if auth_header is not None:
2188 headers = args[1] if len(args) >= 2 else kwargs.setdefault('headers', {})
2189 headers['Authorization'] = auth_header
2190 return compat_urllib_request.Request(url, *args, **kwargs)
2191
2192
2193def expand_path(s):
2194 """Expand shell variables and ~"""
2195 return os.path.expandvars(compat_expanduser(s))
2196
2197
2198def orderedSet(iterable):
2199 """ Remove all duplicates from the input iterable """
2200 res = []
2201 for el in iterable:
2202 if el not in res:
2203 res.append(el)
2204 return res
2205
2206
2207def _htmlentity_transform(entity_with_semicolon):
2208 """Transforms an HTML entity to a character."""
2209 entity = entity_with_semicolon[:-1]
2210
2211 # Known non-numeric HTML entity
2212 if entity in compat_html_entities.name2codepoint:
2213 return compat_chr(compat_html_entities.name2codepoint[entity])
2214
2215 # TODO: HTML5 allows entities without a semicolon. For example,
2216 # '&Eacuteric' should be decoded as 'Éric'.
2217 if entity_with_semicolon in compat_html_entities_html5:
2218 return compat_html_entities_html5[entity_with_semicolon]
2219
2220 mobj = re.match(r'#(x[0-9a-fA-F]+|[0-9]+)', entity)
2221 if mobj is not None:
2222 numstr = mobj.group(1)
2223 if numstr.startswith('x'):
2224 base = 16
2225 numstr = '0%s' % numstr
2226 else:
2227 base = 10
2228 # See https://github.com/ytdl-org/youtube-dl/issues/7518
2229 try:
2230 return compat_chr(int(numstr, base))
2231 except ValueError:
2232 pass
2233
2234 # Unknown entity in name, return its literal representation
2235 return '&%s;' % entity
2236
2237
2238def unescapeHTML(s):
2239 if s is None:
2240 return None
2241 assert type(s) == compat_str
2242
2243 return re.sub(
2244 r'&([^&;]+;)', lambda m: _htmlentity_transform(m.group(1)), s)
2245
2246
2247def escapeHTML(text):
2248 return (
2249 text
2250 .replace('&', '&amp;')
2251 .replace('<', '&lt;')
2252 .replace('>', '&gt;')
2253 .replace('"', '&quot;')
2254 .replace("'", '&#39;')
2255 )
2256
2257
2258def process_communicate_or_kill(p, *args, **kwargs):
2259 try:
2260 return p.communicate(*args, **kwargs)
2261 except BaseException: # Including KeyboardInterrupt
2262 p.kill()
2263 p.wait()
2264 raise
2265
2266
2267def get_subprocess_encoding():
2268 if sys.platform == 'win32' and sys.getwindowsversion()[0] >= 5:
2269 # For subprocess calls, encode with locale encoding
2270 # Refer to http://stackoverflow.com/a/9951851/35070
2271 encoding = preferredencoding()
2272 else:
2273 encoding = sys.getfilesystemencoding()
2274 if encoding is None:
2275 encoding = 'utf-8'
2276 return encoding
2277
2278
2279def encodeFilename(s, for_subprocess=False):
2280 """
2281 @param s The name of the file
2282 """
2283
2284 assert type(s) == compat_str
2285
2286 # Python 3 has a Unicode API
2287 if sys.version_info >= (3, 0):
2288 return s
2289
2290 # Pass '' directly to use Unicode APIs on Windows 2000 and up
2291 # (Detecting Windows NT 4 is tricky because 'major >= 4' would
2292 # match Windows 9x series as well. Besides, NT 4 is obsolete.)
2293 if not for_subprocess and sys.platform == 'win32' and sys.getwindowsversion()[0] >= 5:
2294 return s
2295
2296 # Jython assumes filenames are Unicode strings though reported as Python 2.x compatible
2297 if sys.platform.startswith('java'):
2298 return s
2299
2300 return s.encode(get_subprocess_encoding(), 'ignore')
2301
2302
2303def decodeFilename(b, for_subprocess=False):
2304
2305 if sys.version_info >= (3, 0):
2306 return b
2307
2308 if not isinstance(b, bytes):
2309 return b
2310
2311 return b.decode(get_subprocess_encoding(), 'ignore')
2312
2313
2314def encodeArgument(s):
2315 if not isinstance(s, compat_str):
2316 # Legacy code that uses byte strings
2317 # Uncomment the following line after fixing all post processors
2318 # assert False, 'Internal error: %r should be of type %r, is %r' % (s, compat_str, type(s))
2319 s = s.decode('ascii')
2320 return encodeFilename(s, True)
2321
2322
2323def decodeArgument(b):
2324 return decodeFilename(b, True)
2325
2326
2327def decodeOption(optval):
2328 if optval is None:
2329 return optval
2330 if isinstance(optval, bytes):
2331 optval = optval.decode(preferredencoding())
2332
2333 assert isinstance(optval, compat_str)
2334 return optval
2335
2336
2337def formatSeconds(secs, delim=':', msec=False):
2338 if secs > 3600:
2339 ret = '%d%s%02d%s%02d' % (secs // 3600, delim, (secs % 3600) // 60, delim, secs % 60)
2340 elif secs > 60:
2341 ret = '%d%s%02d' % (secs // 60, delim, secs % 60)
2342 else:
2343 ret = '%d' % secs
2344 return '%s.%03d' % (ret, secs % 1) if msec else ret
2345
2346
2347def make_HTTPS_handler(params, **kwargs):
2348 opts_no_check_certificate = params.get('nocheckcertificate', False)
2349 if hasattr(ssl, 'create_default_context'): # Python >= 3.4 or 2.7.9
2350 context = ssl.create_default_context(ssl.Purpose.SERVER_AUTH)
2351 if opts_no_check_certificate:
2352 context.check_hostname = False
2353 context.verify_mode = ssl.CERT_NONE
2354 try:
2355 return YoutubeDLHTTPSHandler(params, context=context, **kwargs)
2356 except TypeError:
2357 # Python 2.7.8
2358 # (create_default_context present but HTTPSHandler has no context=)
2359 pass
2360
2361 if sys.version_info < (3, 2):
2362 return YoutubeDLHTTPSHandler(params, **kwargs)
2363 else: # Python < 3.4
2364 context = ssl.SSLContext(ssl.PROTOCOL_TLSv1)
2365 context.verify_mode = (ssl.CERT_NONE
2366 if opts_no_check_certificate
2367 else ssl.CERT_REQUIRED)
2368 context.set_default_verify_paths()
2369 return YoutubeDLHTTPSHandler(params, context=context, **kwargs)
2370
2371
2372def bug_reports_message(before=';'):
2373 if ytdl_is_updateable():
2374 update_cmd = 'type yt-dlp -U to update'
2375 else:
2376 update_cmd = 'see https://github.com/yt-dlp/yt-dlp on how to update'
2377 msg = 'please report this issue on https://github.com/yt-dlp/yt-dlp .'
2378 msg += ' Make sure you are using the latest version; %s.' % update_cmd
2379 msg += ' Be sure to call yt-dlp with the --verbose flag and include its complete output.'
2380
2381 before = before.rstrip()
2382 if not before or before.endswith(('.', '!', '?')):
2383 msg = msg[0].title() + msg[1:]
2384
2385 return (before + ' ' if before else '') + msg
2386
2387
2388class YoutubeDLError(Exception):
2389 """Base exception for YoutubeDL errors."""
2390 pass
2391
2392
2393network_exceptions = [compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error]
2394if hasattr(ssl, 'CertificateError'):
2395 network_exceptions.append(ssl.CertificateError)
2396network_exceptions = tuple(network_exceptions)
2397
2398
2399class ExtractorError(YoutubeDLError):
2400 """Error during info extraction."""
2401
2402 def __init__(self, msg, tb=None, expected=False, cause=None, video_id=None):
2403 """ tb, if given, is the original traceback (so that it can be printed out).
2404 If expected is set, this is a normal error message and most likely not a bug in yt-dlp.
2405 """
2406
2407 if sys.exc_info()[0] in network_exceptions:
2408 expected = True
2409 if video_id is not None:
2410 msg = video_id + ': ' + msg
2411 if cause:
2412 msg += ' (caused by %r)' % cause
2413 if not expected:
2414 msg += bug_reports_message()
2415 super(ExtractorError, self).__init__(msg)
2416
2417 self.traceback = tb
2418 self.exc_info = sys.exc_info() # preserve original exception
2419 self.cause = cause
2420 self.video_id = video_id
2421
2422 def format_traceback(self):
2423 if self.traceback is None:
2424 return None
2425 return ''.join(traceback.format_tb(self.traceback))
2426
2427
2428class UnsupportedError(ExtractorError):
2429 def __init__(self, url):
2430 super(UnsupportedError, self).__init__(
2431 'Unsupported URL: %s' % url, expected=True)
2432 self.url = url
2433
2434
2435class RegexNotFoundError(ExtractorError):
2436 """Error when a regex didn't match"""
2437 pass
2438
2439
2440class GeoRestrictedError(ExtractorError):
2441 """Geographic restriction Error exception.
2442
2443 This exception may be thrown when a video is not available from your
2444 geographic location due to geographic restrictions imposed by a website.
2445 """
2446
2447 def __init__(self, msg, countries=None):
2448 super(GeoRestrictedError, self).__init__(msg, expected=True)
2449 self.msg = msg
2450 self.countries = countries
2451
2452
2453class DownloadError(YoutubeDLError):
2454 """Download Error exception.
2455
2456 This exception may be thrown by FileDownloader objects if they are not
2457 configured to continue on errors. They will contain the appropriate
2458 error message.
2459 """
2460
2461 def __init__(self, msg, exc_info=None):
2462 """ exc_info, if given, is the original exception that caused the trouble (as returned by sys.exc_info()). """
2463 super(DownloadError, self).__init__(msg)
2464 self.exc_info = exc_info
2465
2466
2467class EntryNotInPlaylist(YoutubeDLError):
2468 """Entry not in playlist exception.
2469
2470 This exception will be thrown by YoutubeDL when a requested entry
2471 is not found in the playlist info_dict
2472 """
2473 pass
2474
2475
2476class SameFileError(YoutubeDLError):
2477 """Same File exception.
2478
2479 This exception will be thrown by FileDownloader objects if they detect
2480 multiple files would have to be downloaded to the same file on disk.
2481 """
2482 pass
2483
2484
2485class PostProcessingError(YoutubeDLError):
2486 """Post Processing exception.
2487
2488 This exception may be raised by PostProcessor's .run() method to
2489 indicate an error in the postprocessing task.
2490 """
2491
2492 def __init__(self, msg):
2493 super(PostProcessingError, self).__init__(msg)
2494 self.msg = msg
2495
2496
2497class ExistingVideoReached(YoutubeDLError):
2498 """ --max-downloads limit has been reached. """
2499 pass
2500
2501
2502class RejectedVideoReached(YoutubeDLError):
2503 """ --max-downloads limit has been reached. """
2504 pass
2505
2506
2507class MaxDownloadsReached(YoutubeDLError):
2508 """ --max-downloads limit has been reached. """
2509 pass
2510
2511
2512class UnavailableVideoError(YoutubeDLError):
2513 """Unavailable Format exception.
2514
2515 This exception will be thrown when a video is requested
2516 in a format that is not available for that video.
2517 """
2518 pass
2519
2520
2521class ContentTooShortError(YoutubeDLError):
2522 """Content Too Short exception.
2523
2524 This exception may be raised by FileDownloader objects when a file they
2525 download is too small for what the server announced first, indicating
2526 the connection was probably interrupted.
2527 """
2528
2529 def __init__(self, downloaded, expected):
2530 super(ContentTooShortError, self).__init__(
2531 'Downloaded {0} bytes, expected {1} bytes'.format(downloaded, expected)
2532 )
2533 # Both in bytes
2534 self.downloaded = downloaded
2535 self.expected = expected
2536
2537
2538class XAttrMetadataError(YoutubeDLError):
2539 def __init__(self, code=None, msg='Unknown error'):
2540 super(XAttrMetadataError, self).__init__(msg)
2541 self.code = code
2542 self.msg = msg
2543
2544 # Parsing code and msg
2545 if (self.code in (errno.ENOSPC, errno.EDQUOT)
2546 or 'No space left' in self.msg or 'Disk quota exceeded' in self.msg):
2547 self.reason = 'NO_SPACE'
2548 elif self.code == errno.E2BIG or 'Argument list too long' in self.msg:
2549 self.reason = 'VALUE_TOO_LONG'
2550 else:
2551 self.reason = 'NOT_SUPPORTED'
2552
2553
2554class XAttrUnavailableError(YoutubeDLError):
2555 pass
2556
2557
2558def _create_http_connection(ydl_handler, http_class, is_https, *args, **kwargs):
2559 # Working around python 2 bug (see http://bugs.python.org/issue17849) by limiting
2560 # expected HTTP responses to meet HTTP/1.0 or later (see also
2561 # https://github.com/ytdl-org/youtube-dl/issues/6727)
2562 if sys.version_info < (3, 0):
2563 kwargs['strict'] = True
2564 hc = http_class(*args, **compat_kwargs(kwargs))
2565 source_address = ydl_handler._params.get('source_address')
2566
2567 if source_address is not None:
2568 # This is to workaround _create_connection() from socket where it will try all
2569 # address data from getaddrinfo() including IPv6. This filters the result from
2570 # getaddrinfo() based on the source_address value.
2571 # This is based on the cpython socket.create_connection() function.
2572 # https://github.com/python/cpython/blob/master/Lib/socket.py#L691
2573 def _create_connection(address, timeout=socket._GLOBAL_DEFAULT_TIMEOUT, source_address=None):
2574 host, port = address
2575 err = None
2576 addrs = socket.getaddrinfo(host, port, 0, socket.SOCK_STREAM)
2577 af = socket.AF_INET if '.' in source_address[0] else socket.AF_INET6
2578 ip_addrs = [addr for addr in addrs if addr[0] == af]
2579 if addrs and not ip_addrs:
2580 ip_version = 'v4' if af == socket.AF_INET else 'v6'
2581 raise socket.error(
2582 "No remote IP%s addresses available for connect, can't use '%s' as source address"
2583 % (ip_version, source_address[0]))
2584 for res in ip_addrs:
2585 af, socktype, proto, canonname, sa = res
2586 sock = None
2587 try:
2588 sock = socket.socket(af, socktype, proto)
2589 if timeout is not socket._GLOBAL_DEFAULT_TIMEOUT:
2590 sock.settimeout(timeout)
2591 sock.bind(source_address)
2592 sock.connect(sa)
2593 err = None # Explicitly break reference cycle
2594 return sock
2595 except socket.error as _:
2596 err = _
2597 if sock is not None:
2598 sock.close()
2599 if err is not None:
2600 raise err
2601 else:
2602 raise socket.error('getaddrinfo returns an empty list')
2603 if hasattr(hc, '_create_connection'):
2604 hc._create_connection = _create_connection
2605 sa = (source_address, 0)
2606 if hasattr(hc, 'source_address'): # Python 2.7+
2607 hc.source_address = sa
2608 else: # Python 2.6
2609 def _hc_connect(self, *args, **kwargs):
2610 sock = _create_connection(
2611 (self.host, self.port), self.timeout, sa)
2612 if is_https:
2613 self.sock = ssl.wrap_socket(
2614 sock, self.key_file, self.cert_file,
2615 ssl_version=ssl.PROTOCOL_TLSv1)
2616 else:
2617 self.sock = sock
2618 hc.connect = functools.partial(_hc_connect, hc)
2619
2620 return hc
2621
2622
2623def handle_youtubedl_headers(headers):
2624 filtered_headers = headers
2625
2626 if 'Youtubedl-no-compression' in filtered_headers:
2627 filtered_headers = dict((k, v) for k, v in filtered_headers.items() if k.lower() != 'accept-encoding')
2628 del filtered_headers['Youtubedl-no-compression']
2629
2630 return filtered_headers
2631
2632
2633class YoutubeDLHandler(compat_urllib_request.HTTPHandler):
2634 """Handler for HTTP requests and responses.
2635
2636 This class, when installed with an OpenerDirector, automatically adds
2637 the standard headers to every HTTP request and handles gzipped and
2638 deflated responses from web servers. If compression is to be avoided in
2639 a particular request, the original request in the program code only has
2640 to include the HTTP header "Youtubedl-no-compression", which will be
2641 removed before making the real request.
2642
2643 Part of this code was copied from:
2644
2645 http://techknack.net/python-urllib2-handlers/
2646
2647 Andrew Rowls, the author of that code, agreed to release it to the
2648 public domain.
2649 """
2650
2651 def __init__(self, params, *args, **kwargs):
2652 compat_urllib_request.HTTPHandler.__init__(self, *args, **kwargs)
2653 self._params = params
2654
2655 def http_open(self, req):
2656 conn_class = compat_http_client.HTTPConnection
2657
2658 socks_proxy = req.headers.get('Ytdl-socks-proxy')
2659 if socks_proxy:
2660 conn_class = make_socks_conn_class(conn_class, socks_proxy)
2661 del req.headers['Ytdl-socks-proxy']
2662
2663 return self.do_open(functools.partial(
2664 _create_http_connection, self, conn_class, False),
2665 req)
2666
2667 @staticmethod
2668 def deflate(data):
2669 if not data:
2670 return data
2671 try:
2672 return zlib.decompress(data, -zlib.MAX_WBITS)
2673 except zlib.error:
2674 return zlib.decompress(data)
2675
2676 def http_request(self, req):
2677 # According to RFC 3986, URLs can not contain non-ASCII characters, however this is not
2678 # always respected by websites, some tend to give out URLs with non percent-encoded
2679 # non-ASCII characters (see telemb.py, ard.py [#3412])
2680 # urllib chokes on URLs with non-ASCII characters (see http://bugs.python.org/issue3991)
2681 # To work around aforementioned issue we will replace request's original URL with
2682 # percent-encoded one
2683 # Since redirects are also affected (e.g. http://www.southpark.de/alle-episoden/s18e09)
2684 # the code of this workaround has been moved here from YoutubeDL.urlopen()
2685 url = req.get_full_url()
2686 url_escaped = escape_url(url)
2687
2688 # Substitute URL if any change after escaping
2689 if url != url_escaped:
2690 req = update_Request(req, url=url_escaped)
2691
2692 for h, v in std_headers.items():
2693 # Capitalize is needed because of Python bug 2275: http://bugs.python.org/issue2275
2694 # The dict keys are capitalized because of this bug by urllib
2695 if h.capitalize() not in req.headers:
2696 req.add_header(h, v)
2697
2698 req.headers = handle_youtubedl_headers(req.headers)
2699
2700 if sys.version_info < (2, 7) and '#' in req.get_full_url():
2701 # Python 2.6 is brain-dead when it comes to fragments
2702 req._Request__original = req._Request__original.partition('#')[0]
2703 req._Request__r_type = req._Request__r_type.partition('#')[0]
2704
2705 return req
2706
2707 def http_response(self, req, resp):
2708 old_resp = resp
2709 # gzip
2710 if resp.headers.get('Content-encoding', '') == 'gzip':
2711 content = resp.read()
2712 gz = gzip.GzipFile(fileobj=io.BytesIO(content), mode='rb')
2713 try:
2714 uncompressed = io.BytesIO(gz.read())
2715 except IOError as original_ioerror:
2716 # There may be junk add the end of the file
2717 # See http://stackoverflow.com/q/4928560/35070 for details
2718 for i in range(1, 1024):
2719 try:
2720 gz = gzip.GzipFile(fileobj=io.BytesIO(content[:-i]), mode='rb')
2721 uncompressed = io.BytesIO(gz.read())
2722 except IOError:
2723 continue
2724 break
2725 else:
2726 raise original_ioerror
2727 resp = compat_urllib_request.addinfourl(uncompressed, old_resp.headers, old_resp.url, old_resp.code)
2728 resp.msg = old_resp.msg
2729 del resp.headers['Content-encoding']
2730 # deflate
2731 if resp.headers.get('Content-encoding', '') == 'deflate':
2732 gz = io.BytesIO(self.deflate(resp.read()))
2733 resp = compat_urllib_request.addinfourl(gz, old_resp.headers, old_resp.url, old_resp.code)
2734 resp.msg = old_resp.msg
2735 del resp.headers['Content-encoding']
2736 # Percent-encode redirect URL of Location HTTP header to satisfy RFC 3986 (see
2737 # https://github.com/ytdl-org/youtube-dl/issues/6457).
2738 if 300 <= resp.code < 400:
2739 location = resp.headers.get('Location')
2740 if location:
2741 # As of RFC 2616 default charset is iso-8859-1 that is respected by python 3
2742 if sys.version_info >= (3, 0):
2743 location = location.encode('iso-8859-1').decode('utf-8')
2744 else:
2745 location = location.decode('utf-8')
2746 location_escaped = escape_url(location)
2747 if location != location_escaped:
2748 del resp.headers['Location']
2749 if sys.version_info < (3, 0):
2750 location_escaped = location_escaped.encode('utf-8')
2751 resp.headers['Location'] = location_escaped
2752 return resp
2753
2754 https_request = http_request
2755 https_response = http_response
2756
2757
2758def make_socks_conn_class(base_class, socks_proxy):
2759 assert issubclass(base_class, (
2760 compat_http_client.HTTPConnection, compat_http_client.HTTPSConnection))
2761
2762 url_components = compat_urlparse.urlparse(socks_proxy)
2763 if url_components.scheme.lower() == 'socks5':
2764 socks_type = ProxyType.SOCKS5
2765 elif url_components.scheme.lower() in ('socks', 'socks4'):
2766 socks_type = ProxyType.SOCKS4
2767 elif url_components.scheme.lower() == 'socks4a':
2768 socks_type = ProxyType.SOCKS4A
2769
2770 def unquote_if_non_empty(s):
2771 if not s:
2772 return s
2773 return compat_urllib_parse_unquote_plus(s)
2774
2775 proxy_args = (
2776 socks_type,
2777 url_components.hostname, url_components.port or 1080,
2778 True, # Remote DNS
2779 unquote_if_non_empty(url_components.username),
2780 unquote_if_non_empty(url_components.password),
2781 )
2782
2783 class SocksConnection(base_class):
2784 def connect(self):
2785 self.sock = sockssocket()
2786 self.sock.setproxy(*proxy_args)
2787 if type(self.timeout) in (int, float):
2788 self.sock.settimeout(self.timeout)
2789 self.sock.connect((self.host, self.port))
2790
2791 if isinstance(self, compat_http_client.HTTPSConnection):
2792 if hasattr(self, '_context'): # Python > 2.6
2793 self.sock = self._context.wrap_socket(
2794 self.sock, server_hostname=self.host)
2795 else:
2796 self.sock = ssl.wrap_socket(self.sock)
2797
2798 return SocksConnection
2799
2800
2801class YoutubeDLHTTPSHandler(compat_urllib_request.HTTPSHandler):
2802 def __init__(self, params, https_conn_class=None, *args, **kwargs):
2803 compat_urllib_request.HTTPSHandler.__init__(self, *args, **kwargs)
2804 self._https_conn_class = https_conn_class or compat_http_client.HTTPSConnection
2805 self._params = params
2806
2807 def https_open(self, req):
2808 kwargs = {}
2809 conn_class = self._https_conn_class
2810
2811 if hasattr(self, '_context'): # python > 2.6
2812 kwargs['context'] = self._context
2813 if hasattr(self, '_check_hostname'): # python 3.x
2814 kwargs['check_hostname'] = self._check_hostname
2815
2816 socks_proxy = req.headers.get('Ytdl-socks-proxy')
2817 if socks_proxy:
2818 conn_class = make_socks_conn_class(conn_class, socks_proxy)
2819 del req.headers['Ytdl-socks-proxy']
2820
2821 return self.do_open(functools.partial(
2822 _create_http_connection, self, conn_class, True),
2823 req, **kwargs)
2824
2825
2826class YoutubeDLCookieJar(compat_cookiejar.MozillaCookieJar):
2827 """
2828 See [1] for cookie file format.
2829
2830 1. https://curl.haxx.se/docs/http-cookies.html
2831 """
2832 _HTTPONLY_PREFIX = '#HttpOnly_'
2833 _ENTRY_LEN = 7
2834 _HEADER = '''# Netscape HTTP Cookie File
2835# This file is generated by yt-dlp. Do not edit.
2836
2837'''
2838 _CookieFileEntry = collections.namedtuple(
2839 'CookieFileEntry',
2840 ('domain_name', 'include_subdomains', 'path', 'https_only', 'expires_at', 'name', 'value'))
2841
2842 def save(self, filename=None, ignore_discard=False, ignore_expires=False):
2843 """
2844 Save cookies to a file.
2845
2846 Most of the code is taken from CPython 3.8 and slightly adapted
2847 to support cookie files with UTF-8 in both python 2 and 3.
2848 """
2849 if filename is None:
2850 if self.filename is not None:
2851 filename = self.filename
2852 else:
2853 raise ValueError(compat_cookiejar.MISSING_FILENAME_TEXT)
2854
2855 # Store session cookies with `expires` set to 0 instead of an empty
2856 # string
2857 for cookie in self:
2858 if cookie.expires is None:
2859 cookie.expires = 0
2860
2861 with io.open(filename, 'w', encoding='utf-8') as f:
2862 f.write(self._HEADER)
2863 now = time.time()
2864 for cookie in self:
2865 if not ignore_discard and cookie.discard:
2866 continue
2867 if not ignore_expires and cookie.is_expired(now):
2868 continue
2869 if cookie.secure:
2870 secure = 'TRUE'
2871 else:
2872 secure = 'FALSE'
2873 if cookie.domain.startswith('.'):
2874 initial_dot = 'TRUE'
2875 else:
2876 initial_dot = 'FALSE'
2877 if cookie.expires is not None:
2878 expires = compat_str(cookie.expires)
2879 else:
2880 expires = ''
2881 if cookie.value is None:
2882 # cookies.txt regards 'Set-Cookie: foo' as a cookie
2883 # with no name, whereas http.cookiejar regards it as a
2884 # cookie with no value.
2885 name = ''
2886 value = cookie.name
2887 else:
2888 name = cookie.name
2889 value = cookie.value
2890 f.write(
2891 '\t'.join([cookie.domain, initial_dot, cookie.path,
2892 secure, expires, name, value]) + '\n')
2893
2894 def load(self, filename=None, ignore_discard=False, ignore_expires=False):
2895 """Load cookies from a file."""
2896 if filename is None:
2897 if self.filename is not None:
2898 filename = self.filename
2899 else:
2900 raise ValueError(compat_cookiejar.MISSING_FILENAME_TEXT)
2901
2902 def prepare_line(line):
2903 if line.startswith(self._HTTPONLY_PREFIX):
2904 line = line[len(self._HTTPONLY_PREFIX):]
2905 # comments and empty lines are fine
2906 if line.startswith('#') or not line.strip():
2907 return line
2908 cookie_list = line.split('\t')
2909 if len(cookie_list) != self._ENTRY_LEN:
2910 raise compat_cookiejar.LoadError('invalid length %d' % len(cookie_list))
2911 cookie = self._CookieFileEntry(*cookie_list)
2912 if cookie.expires_at and not cookie.expires_at.isdigit():
2913 raise compat_cookiejar.LoadError('invalid expires at %s' % cookie.expires_at)
2914 return line
2915
2916 cf = io.StringIO()
2917 with io.open(filename, encoding='utf-8') as f:
2918 for line in f:
2919 try:
2920 cf.write(prepare_line(line))
2921 except compat_cookiejar.LoadError as e:
2922 write_string(
2923 'WARNING: skipping cookie file entry due to %s: %r\n'
2924 % (e, line), sys.stderr)
2925 continue
2926 cf.seek(0)
2927 self._really_load(cf, filename, ignore_discard, ignore_expires)
2928 # Session cookies are denoted by either `expires` field set to
2929 # an empty string or 0. MozillaCookieJar only recognizes the former
2930 # (see [1]). So we need force the latter to be recognized as session
2931 # cookies on our own.
2932 # Session cookies may be important for cookies-based authentication,
2933 # e.g. usually, when user does not check 'Remember me' check box while
2934 # logging in on a site, some important cookies are stored as session
2935 # cookies so that not recognizing them will result in failed login.
2936 # 1. https://bugs.python.org/issue17164
2937 for cookie in self:
2938 # Treat `expires=0` cookies as session cookies
2939 if cookie.expires == 0:
2940 cookie.expires = None
2941 cookie.discard = True
2942
2943
2944class YoutubeDLCookieProcessor(compat_urllib_request.HTTPCookieProcessor):
2945 def __init__(self, cookiejar=None):
2946 compat_urllib_request.HTTPCookieProcessor.__init__(self, cookiejar)
2947
2948 def http_response(self, request, response):
2949 # Python 2 will choke on next HTTP request in row if there are non-ASCII
2950 # characters in Set-Cookie HTTP header of last response (see
2951 # https://github.com/ytdl-org/youtube-dl/issues/6769).
2952 # In order to at least prevent crashing we will percent encode Set-Cookie
2953 # header before HTTPCookieProcessor starts processing it.
2954 # if sys.version_info < (3, 0) and response.headers:
2955 # for set_cookie_header in ('Set-Cookie', 'Set-Cookie2'):
2956 # set_cookie = response.headers.get(set_cookie_header)
2957 # if set_cookie:
2958 # set_cookie_escaped = compat_urllib_parse.quote(set_cookie, b"%/;:@&=+$,!~*'()?#[] ")
2959 # if set_cookie != set_cookie_escaped:
2960 # del response.headers[set_cookie_header]
2961 # response.headers[set_cookie_header] = set_cookie_escaped
2962 return compat_urllib_request.HTTPCookieProcessor.http_response(self, request, response)
2963
2964 https_request = compat_urllib_request.HTTPCookieProcessor.http_request
2965 https_response = http_response
2966
2967
2968class YoutubeDLRedirectHandler(compat_urllib_request.HTTPRedirectHandler):
2969 """YoutubeDL redirect handler
2970
2971 The code is based on HTTPRedirectHandler implementation from CPython [1].
2972
2973 This redirect handler solves two issues:
2974 - ensures redirect URL is always unicode under python 2
2975 - introduces support for experimental HTTP response status code
2976 308 Permanent Redirect [2] used by some sites [3]
2977
2978 1. https://github.com/python/cpython/blob/master/Lib/urllib/request.py
2979 2. https://developer.mozilla.org/en-US/docs/Web/HTTP/Status/308
2980 3. https://github.com/ytdl-org/youtube-dl/issues/28768
2981 """
2982
2983 http_error_301 = http_error_303 = http_error_307 = http_error_308 = compat_urllib_request.HTTPRedirectHandler.http_error_302
2984
2985 def redirect_request(self, req, fp, code, msg, headers, newurl):
2986 """Return a Request or None in response to a redirect.
2987
2988 This is called by the http_error_30x methods when a
2989 redirection response is received. If a redirection should
2990 take place, return a new Request to allow http_error_30x to
2991 perform the redirect. Otherwise, raise HTTPError if no-one
2992 else should try to handle this url. Return None if you can't
2993 but another Handler might.
2994 """
2995 m = req.get_method()
2996 if (not (code in (301, 302, 303, 307, 308) and m in ("GET", "HEAD")
2997 or code in (301, 302, 303) and m == "POST")):
2998 raise compat_HTTPError(req.full_url, code, msg, headers, fp)
2999 # Strictly (according to RFC 2616), 301 or 302 in response to
3000 # a POST MUST NOT cause a redirection without confirmation
3001 # from the user (of urllib.request, in this case). In practice,
3002 # essentially all clients do redirect in this case, so we do
3003 # the same.
3004
3005 # On python 2 urlh.geturl() may sometimes return redirect URL
3006 # as byte string instead of unicode. This workaround allows
3007 # to force it always return unicode.
3008 if sys.version_info[0] < 3:
3009 newurl = compat_str(newurl)
3010
3011 # Be conciliant with URIs containing a space. This is mainly
3012 # redundant with the more complete encoding done in http_error_302(),
3013 # but it is kept for compatibility with other callers.
3014 newurl = newurl.replace(' ', '%20')
3015
3016 CONTENT_HEADERS = ("content-length", "content-type")
3017 # NB: don't use dict comprehension for python 2.6 compatibility
3018 newheaders = dict((k, v) for k, v in req.headers.items()
3019 if k.lower() not in CONTENT_HEADERS)
3020 return compat_urllib_request.Request(
3021 newurl, headers=newheaders, origin_req_host=req.origin_req_host,
3022 unverifiable=True)
3023
3024
3025def extract_timezone(date_str):
3026 m = re.search(
3027 r'^.{8,}?(?P<tz>Z$| ?(?P<sign>\+|-)(?P<hours>[0-9]{2}):?(?P<minutes>[0-9]{2})$)',
3028 date_str)
3029 if not m:
3030 timezone = datetime.timedelta()
3031 else:
3032 date_str = date_str[:-len(m.group('tz'))]
3033 if not m.group('sign'):
3034 timezone = datetime.timedelta()
3035 else:
3036 sign = 1 if m.group('sign') == '+' else -1
3037 timezone = datetime.timedelta(
3038 hours=sign * int(m.group('hours')),
3039 minutes=sign * int(m.group('minutes')))
3040 return timezone, date_str
3041
3042
3043def parse_iso8601(date_str, delimiter='T', timezone=None):
3044 """ Return a UNIX timestamp from the given date """
3045
3046 if date_str is None:
3047 return None
3048
3049 date_str = re.sub(r'\.[0-9]+', '', date_str)
3050
3051 if timezone is None:
3052 timezone, date_str = extract_timezone(date_str)
3053
3054 try:
3055 date_format = '%Y-%m-%d{0}%H:%M:%S'.format(delimiter)
3056 dt = datetime.datetime.strptime(date_str, date_format) - timezone
3057 return calendar.timegm(dt.timetuple())
3058 except ValueError:
3059 pass
3060
3061
3062def date_formats(day_first=True):
3063 return DATE_FORMATS_DAY_FIRST if day_first else DATE_FORMATS_MONTH_FIRST
3064
3065
3066def unified_strdate(date_str, day_first=True):
3067 """Return a string with the date in the format YYYYMMDD"""
3068
3069 if date_str is None:
3070 return None
3071 upload_date = None
3072 # Replace commas
3073 date_str = date_str.replace(',', ' ')
3074 # Remove AM/PM + timezone
3075 date_str = re.sub(r'(?i)\s*(?:AM|PM)(?:\s+[A-Z]+)?', '', date_str)
3076 _, date_str = extract_timezone(date_str)
3077
3078 for expression in date_formats(day_first):
3079 try:
3080 upload_date = datetime.datetime.strptime(date_str, expression).strftime('%Y%m%d')
3081 except ValueError:
3082 pass
3083 if upload_date is None:
3084 timetuple = email.utils.parsedate_tz(date_str)
3085 if timetuple:
3086 try:
3087 upload_date = datetime.datetime(*timetuple[:6]).strftime('%Y%m%d')
3088 except ValueError:
3089 pass
3090 if upload_date is not None:
3091 return compat_str(upload_date)
3092
3093
3094def unified_timestamp(date_str, day_first=True):
3095 if date_str is None:
3096 return None
3097
3098 date_str = re.sub(r'[,|]', '', date_str)
3099
3100 pm_delta = 12 if re.search(r'(?i)PM', date_str) else 0
3101 timezone, date_str = extract_timezone(date_str)
3102
3103 # Remove AM/PM + timezone
3104 date_str = re.sub(r'(?i)\s*(?:AM|PM)(?:\s+[A-Z]+)?', '', date_str)
3105
3106 # Remove unrecognized timezones from ISO 8601 alike timestamps
3107 m = re.search(r'\d{1,2}:\d{1,2}(?:\.\d+)?(?P<tz>\s*[A-Z]+)$', date_str)
3108 if m:
3109 date_str = date_str[:-len(m.group('tz'))]
3110
3111 # Python only supports microseconds, so remove nanoseconds
3112 m = re.search(r'^([0-9]{4,}-[0-9]{1,2}-[0-9]{1,2}T[0-9]{1,2}:[0-9]{1,2}:[0-9]{1,2}\.[0-9]{6})[0-9]+$', date_str)
3113 if m:
3114 date_str = m.group(1)
3115
3116 for expression in date_formats(day_first):
3117 try:
3118 dt = datetime.datetime.strptime(date_str, expression) - timezone + datetime.timedelta(hours=pm_delta)
3119 return calendar.timegm(dt.timetuple())
3120 except ValueError:
3121 pass
3122 timetuple = email.utils.parsedate_tz(date_str)
3123 if timetuple:
3124 return calendar.timegm(timetuple) + pm_delta * 3600
3125
3126
3127def determine_ext(url, default_ext='unknown_video'):
3128 if url is None or '.' not in url:
3129 return default_ext
3130 guess = url.partition('?')[0].rpartition('.')[2]
3131 if re.match(r'^[A-Za-z0-9]+$', guess):
3132 return guess
3133 # Try extract ext from URLs like http://example.com/foo/bar.mp4/?download
3134 elif guess.rstrip('/') in KNOWN_EXTENSIONS:
3135 return guess.rstrip('/')
3136 else:
3137 return default_ext
3138
3139
3140def subtitles_filename(filename, sub_lang, sub_format, expected_real_ext=None):
3141 return replace_extension(filename, sub_lang + '.' + sub_format, expected_real_ext)
3142
3143
3144def datetime_from_str(date_str, precision='auto', format='%Y%m%d'):
3145 """
3146 Return a datetime object from a string in the format YYYYMMDD or
3147 (now|today|date)[+-][0-9](microsecond|second|minute|hour|day|week|month|year)(s)?
3148
3149 format: string date format used to return datetime object from
3150 precision: round the time portion of a datetime object.
3151 auto|microsecond|second|minute|hour|day.
3152 auto: round to the unit provided in date_str (if applicable).
3153 """
3154 auto_precision = False
3155 if precision == 'auto':
3156 auto_precision = True
3157 precision = 'microsecond'
3158 today = datetime_round(datetime.datetime.now(), precision)
3159 if date_str in ('now', 'today'):
3160 return today
3161 if date_str == 'yesterday':
3162 return today - datetime.timedelta(days=1)
3163 match = re.match(
3164 r'(?P<start>.+)(?P<sign>[+-])(?P<time>\d+)(?P<unit>microsecond|second|minute|hour|day|week|month|year)(s)?',
3165 date_str)
3166 if match is not None:
3167 start_time = datetime_from_str(match.group('start'), precision, format)
3168 time = int(match.group('time')) * (-1 if match.group('sign') == '-' else 1)
3169 unit = match.group('unit')
3170 if unit == 'month' or unit == 'year':
3171 new_date = datetime_add_months(start_time, time * 12 if unit == 'year' else time)
3172 unit = 'day'
3173 else:
3174 if unit == 'week':
3175 unit = 'day'
3176 time *= 7
3177 delta = datetime.timedelta(**{unit + 's': time})
3178 new_date = start_time + delta
3179 if auto_precision:
3180 return datetime_round(new_date, unit)
3181 return new_date
3182
3183 return datetime_round(datetime.datetime.strptime(date_str, format), precision)
3184
3185
3186def date_from_str(date_str, format='%Y%m%d'):
3187 """
3188 Return a datetime object from a string in the format YYYYMMDD or
3189 (now|today|date)[+-][0-9](microsecond|second|minute|hour|day|week|month|year)(s)?
3190
3191 format: string date format used to return datetime object from
3192 """
3193 return datetime_from_str(date_str, precision='microsecond', format=format).date()
3194
3195
3196def datetime_add_months(dt, months):
3197 """Increment/Decrement a datetime object by months."""
3198 month = dt.month + months - 1
3199 year = dt.year + month // 12
3200 month = month % 12 + 1
3201 day = min(dt.day, calendar.monthrange(year, month)[1])
3202 return dt.replace(year, month, day)
3203
3204
3205def datetime_round(dt, precision='day'):
3206 """
3207 Round a datetime object's time to a specific precision
3208 """
3209 if precision == 'microsecond':
3210 return dt
3211
3212 unit_seconds = {
3213 'day': 86400,
3214 'hour': 3600,
3215 'minute': 60,
3216 'second': 1,
3217 }
3218 roundto = lambda x, n: ((x + n / 2) // n) * n
3219 timestamp = calendar.timegm(dt.timetuple())
3220 return datetime.datetime.utcfromtimestamp(roundto(timestamp, unit_seconds[precision]))
3221
3222
3223def hyphenate_date(date_str):
3224 """
3225 Convert a date in 'YYYYMMDD' format to 'YYYY-MM-DD' format"""
3226 match = re.match(r'^(\d\d\d\d)(\d\d)(\d\d)$', date_str)
3227 if match is not None:
3228 return '-'.join(match.groups())
3229 else:
3230 return date_str
3231
3232
3233class DateRange(object):
3234 """Represents a time interval between two dates"""
3235
3236 def __init__(self, start=None, end=None):
3237 """start and end must be strings in the format accepted by date"""
3238 if start is not None:
3239 self.start = date_from_str(start)
3240 else:
3241 self.start = datetime.datetime.min.date()
3242 if end is not None:
3243 self.end = date_from_str(end)
3244 else:
3245 self.end = datetime.datetime.max.date()
3246 if self.start > self.end:
3247 raise ValueError('Date range: "%s" , the start date must be before the end date' % self)
3248
3249 @classmethod
3250 def day(cls, day):
3251 """Returns a range that only contains the given day"""
3252 return cls(day, day)
3253
3254 def __contains__(self, date):
3255 """Check if the date is in the range"""
3256 if not isinstance(date, datetime.date):
3257 date = date_from_str(date)
3258 return self.start <= date <= self.end
3259
3260 def __str__(self):
3261 return '%s - %s' % (self.start.isoformat(), self.end.isoformat())
3262
3263
3264def platform_name():
3265 """ Returns the platform name as a compat_str """
3266 res = platform.platform()
3267 if isinstance(res, bytes):
3268 res = res.decode(preferredencoding())
3269
3270 assert isinstance(res, compat_str)
3271 return res
3272
3273
3274def _windows_write_string(s, out):
3275 """ Returns True if the string was written using special methods,
3276 False if it has yet to be written out."""
3277 # Adapted from http://stackoverflow.com/a/3259271/35070
3278
3279 import ctypes
3280 import ctypes.wintypes
3281
3282 WIN_OUTPUT_IDS = {
3283 1: -11,
3284 2: -12,
3285 }
3286
3287 try:
3288 fileno = out.fileno()
3289 except AttributeError:
3290 # If the output stream doesn't have a fileno, it's virtual
3291 return False
3292 except io.UnsupportedOperation:
3293 # Some strange Windows pseudo files?
3294 return False
3295 if fileno not in WIN_OUTPUT_IDS:
3296 return False
3297
3298 GetStdHandle = compat_ctypes_WINFUNCTYPE(
3299 ctypes.wintypes.HANDLE, ctypes.wintypes.DWORD)(
3300 ('GetStdHandle', ctypes.windll.kernel32))
3301 h = GetStdHandle(WIN_OUTPUT_IDS[fileno])
3302
3303 WriteConsoleW = compat_ctypes_WINFUNCTYPE(
3304 ctypes.wintypes.BOOL, ctypes.wintypes.HANDLE, ctypes.wintypes.LPWSTR,
3305 ctypes.wintypes.DWORD, ctypes.POINTER(ctypes.wintypes.DWORD),
3306 ctypes.wintypes.LPVOID)(('WriteConsoleW', ctypes.windll.kernel32))
3307 written = ctypes.wintypes.DWORD(0)
3308
3309 GetFileType = compat_ctypes_WINFUNCTYPE(ctypes.wintypes.DWORD, ctypes.wintypes.DWORD)(('GetFileType', ctypes.windll.kernel32))
3310 FILE_TYPE_CHAR = 0x0002
3311 FILE_TYPE_REMOTE = 0x8000
3312 GetConsoleMode = compat_ctypes_WINFUNCTYPE(
3313 ctypes.wintypes.BOOL, ctypes.wintypes.HANDLE,
3314 ctypes.POINTER(ctypes.wintypes.DWORD))(
3315 ('GetConsoleMode', ctypes.windll.kernel32))
3316 INVALID_HANDLE_VALUE = ctypes.wintypes.DWORD(-1).value
3317
3318 def not_a_console(handle):
3319 if handle == INVALID_HANDLE_VALUE or handle is None:
3320 return True
3321 return ((GetFileType(handle) & ~FILE_TYPE_REMOTE) != FILE_TYPE_CHAR
3322 or GetConsoleMode(handle, ctypes.byref(ctypes.wintypes.DWORD())) == 0)
3323
3324 if not_a_console(h):
3325 return False
3326
3327 def next_nonbmp_pos(s):
3328 try:
3329 return next(i for i, c in enumerate(s) if ord(c) > 0xffff)
3330 except StopIteration:
3331 return len(s)
3332
3333 while s:
3334 count = min(next_nonbmp_pos(s), 1024)
3335
3336 ret = WriteConsoleW(
3337 h, s, count if count else 2, ctypes.byref(written), None)
3338 if ret == 0:
3339 raise OSError('Failed to write string')
3340 if not count: # We just wrote a non-BMP character
3341 assert written.value == 2
3342 s = s[1:]
3343 else:
3344 assert written.value > 0
3345 s = s[written.value:]
3346 return True
3347
3348
3349def write_string(s, out=None, encoding=None):
3350 if out is None:
3351 out = sys.stderr
3352 assert type(s) == compat_str
3353
3354 if sys.platform == 'win32' and encoding is None and hasattr(out, 'fileno'):
3355 if _windows_write_string(s, out):
3356 return
3357
3358 if ('b' in getattr(out, 'mode', '')
3359 or sys.version_info[0] < 3): # Python 2 lies about mode of sys.stderr
3360 byt = s.encode(encoding or preferredencoding(), 'ignore')
3361 out.write(byt)
3362 elif hasattr(out, 'buffer'):
3363 enc = encoding or getattr(out, 'encoding', None) or preferredencoding()
3364 byt = s.encode(enc, 'ignore')
3365 out.buffer.write(byt)
3366 else:
3367 out.write(s)
3368 out.flush()
3369
3370
3371def bytes_to_intlist(bs):
3372 if not bs:
3373 return []
3374 if isinstance(bs[0], int): # Python 3
3375 return list(bs)
3376 else:
3377 return [ord(c) for c in bs]
3378
3379
3380def intlist_to_bytes(xs):
3381 if not xs:
3382 return b''
3383 return compat_struct_pack('%dB' % len(xs), *xs)
3384
3385
3386# Cross-platform file locking
3387if sys.platform == 'win32':
3388 import ctypes.wintypes
3389 import msvcrt
3390
3391 class OVERLAPPED(ctypes.Structure):
3392 _fields_ = [
3393 ('Internal', ctypes.wintypes.LPVOID),
3394 ('InternalHigh', ctypes.wintypes.LPVOID),
3395 ('Offset', ctypes.wintypes.DWORD),
3396 ('OffsetHigh', ctypes.wintypes.DWORD),
3397 ('hEvent', ctypes.wintypes.HANDLE),
3398 ]
3399
3400 kernel32 = ctypes.windll.kernel32
3401 LockFileEx = kernel32.LockFileEx
3402 LockFileEx.argtypes = [
3403 ctypes.wintypes.HANDLE, # hFile
3404 ctypes.wintypes.DWORD, # dwFlags
3405 ctypes.wintypes.DWORD, # dwReserved
3406 ctypes.wintypes.DWORD, # nNumberOfBytesToLockLow
3407 ctypes.wintypes.DWORD, # nNumberOfBytesToLockHigh
3408 ctypes.POINTER(OVERLAPPED) # Overlapped
3409 ]
3410 LockFileEx.restype = ctypes.wintypes.BOOL
3411 UnlockFileEx = kernel32.UnlockFileEx
3412 UnlockFileEx.argtypes = [
3413 ctypes.wintypes.HANDLE, # hFile
3414 ctypes.wintypes.DWORD, # dwReserved
3415 ctypes.wintypes.DWORD, # nNumberOfBytesToLockLow
3416 ctypes.wintypes.DWORD, # nNumberOfBytesToLockHigh
3417 ctypes.POINTER(OVERLAPPED) # Overlapped
3418 ]
3419 UnlockFileEx.restype = ctypes.wintypes.BOOL
3420 whole_low = 0xffffffff
3421 whole_high = 0x7fffffff
3422
3423 def _lock_file(f, exclusive):
3424 overlapped = OVERLAPPED()
3425 overlapped.Offset = 0
3426 overlapped.OffsetHigh = 0
3427 overlapped.hEvent = 0
3428 f._lock_file_overlapped_p = ctypes.pointer(overlapped)
3429 handle = msvcrt.get_osfhandle(f.fileno())
3430 if not LockFileEx(handle, 0x2 if exclusive else 0x0, 0,
3431 whole_low, whole_high, f._lock_file_overlapped_p):
3432 raise OSError('Locking file failed: %r' % ctypes.FormatError())
3433
3434 def _unlock_file(f):
3435 assert f._lock_file_overlapped_p
3436 handle = msvcrt.get_osfhandle(f.fileno())
3437 if not UnlockFileEx(handle, 0,
3438 whole_low, whole_high, f._lock_file_overlapped_p):
3439 raise OSError('Unlocking file failed: %r' % ctypes.FormatError())
3440
3441else:
3442 # Some platforms, such as Jython, is missing fcntl
3443 try:
3444 import fcntl
3445
3446 def _lock_file(f, exclusive):
3447 fcntl.flock(f, fcntl.LOCK_EX if exclusive else fcntl.LOCK_SH)
3448
3449 def _unlock_file(f):
3450 fcntl.flock(f, fcntl.LOCK_UN)
3451 except ImportError:
3452 UNSUPPORTED_MSG = 'file locking is not supported on this platform'
3453
3454 def _lock_file(f, exclusive):
3455 raise IOError(UNSUPPORTED_MSG)
3456
3457 def _unlock_file(f):
3458 raise IOError(UNSUPPORTED_MSG)
3459
3460
3461class locked_file(object):
3462 def __init__(self, filename, mode, encoding=None):
3463 assert mode in ['r', 'a', 'w']
3464 self.f = io.open(filename, mode, encoding=encoding)
3465 self.mode = mode
3466
3467 def __enter__(self):
3468 exclusive = self.mode != 'r'
3469 try:
3470 _lock_file(self.f, exclusive)
3471 except IOError:
3472 self.f.close()
3473 raise
3474 return self
3475
3476 def __exit__(self, etype, value, traceback):
3477 try:
3478 _unlock_file(self.f)
3479 finally:
3480 self.f.close()
3481
3482 def __iter__(self):
3483 return iter(self.f)
3484
3485 def write(self, *args):
3486 return self.f.write(*args)
3487
3488 def read(self, *args):
3489 return self.f.read(*args)
3490
3491
3492def get_filesystem_encoding():
3493 encoding = sys.getfilesystemencoding()
3494 return encoding if encoding is not None else 'utf-8'
3495
3496
3497def shell_quote(args):
3498 quoted_args = []
3499 encoding = get_filesystem_encoding()
3500 for a in args:
3501 if isinstance(a, bytes):
3502 # We may get a filename encoded with 'encodeFilename'
3503 a = a.decode(encoding)
3504 quoted_args.append(compat_shlex_quote(a))
3505 return ' '.join(quoted_args)
3506
3507
3508def smuggle_url(url, data):
3509 """ Pass additional data in a URL for internal use. """
3510
3511 url, idata = unsmuggle_url(url, {})
3512 data.update(idata)
3513 sdata = compat_urllib_parse_urlencode(
3514 {'__youtubedl_smuggle': json.dumps(data)})
3515 return url + '#' + sdata
3516
3517
3518def unsmuggle_url(smug_url, default=None):
3519 if '#__youtubedl_smuggle' not in smug_url:
3520 return smug_url, default
3521 url, _, sdata = smug_url.rpartition('#')
3522 jsond = compat_parse_qs(sdata)['__youtubedl_smuggle'][0]
3523 data = json.loads(jsond)
3524 return url, data
3525
3526
3527def format_bytes(bytes):
3528 if bytes is None:
3529 return 'N/A'
3530 if type(bytes) is str:
3531 bytes = float(bytes)
3532 if bytes == 0.0:
3533 exponent = 0
3534 else:
3535 exponent = int(math.log(bytes, 1024.0))
3536 suffix = ['B', 'KiB', 'MiB', 'GiB', 'TiB', 'PiB', 'EiB', 'ZiB', 'YiB'][exponent]
3537 converted = float(bytes) / float(1024 ** exponent)
3538 return '%.2f%s' % (converted, suffix)
3539
3540
3541def lookup_unit_table(unit_table, s):
3542 units_re = '|'.join(re.escape(u) for u in unit_table)
3543 m = re.match(
3544 r'(?P<num>[0-9]+(?:[,.][0-9]*)?)\s*(?P<unit>%s)\b' % units_re, s)
3545 if not m:
3546 return None
3547 num_str = m.group('num').replace(',', '.')
3548 mult = unit_table[m.group('unit')]
3549 return int(float(num_str) * mult)
3550
3551
3552def parse_filesize(s):
3553 if s is None:
3554 return None
3555
3556 # The lower-case forms are of course incorrect and unofficial,
3557 # but we support those too
3558 _UNIT_TABLE = {
3559 'B': 1,
3560 'b': 1,
3561 'bytes': 1,
3562 'KiB': 1024,
3563 'KB': 1000,
3564 'kB': 1024,
3565 'Kb': 1000,
3566 'kb': 1000,
3567 'kilobytes': 1000,
3568 'kibibytes': 1024,
3569 'MiB': 1024 ** 2,
3570 'MB': 1000 ** 2,
3571 'mB': 1024 ** 2,
3572 'Mb': 1000 ** 2,
3573 'mb': 1000 ** 2,
3574 'megabytes': 1000 ** 2,
3575 'mebibytes': 1024 ** 2,
3576 'GiB': 1024 ** 3,
3577 'GB': 1000 ** 3,
3578 'gB': 1024 ** 3,
3579 'Gb': 1000 ** 3,
3580 'gb': 1000 ** 3,
3581 'gigabytes': 1000 ** 3,
3582 'gibibytes': 1024 ** 3,
3583 'TiB': 1024 ** 4,
3584 'TB': 1000 ** 4,
3585 'tB': 1024 ** 4,
3586 'Tb': 1000 ** 4,
3587 'tb': 1000 ** 4,
3588 'terabytes': 1000 ** 4,
3589 'tebibytes': 1024 ** 4,
3590 'PiB': 1024 ** 5,
3591 'PB': 1000 ** 5,
3592 'pB': 1024 ** 5,
3593 'Pb': 1000 ** 5,
3594 'pb': 1000 ** 5,
3595 'petabytes': 1000 ** 5,
3596 'pebibytes': 1024 ** 5,
3597 'EiB': 1024 ** 6,
3598 'EB': 1000 ** 6,
3599 'eB': 1024 ** 6,
3600 'Eb': 1000 ** 6,
3601 'eb': 1000 ** 6,
3602 'exabytes': 1000 ** 6,
3603 'exbibytes': 1024 ** 6,
3604 'ZiB': 1024 ** 7,
3605 'ZB': 1000 ** 7,
3606 'zB': 1024 ** 7,
3607 'Zb': 1000 ** 7,
3608 'zb': 1000 ** 7,
3609 'zettabytes': 1000 ** 7,
3610 'zebibytes': 1024 ** 7,
3611 'YiB': 1024 ** 8,
3612 'YB': 1000 ** 8,
3613 'yB': 1024 ** 8,
3614 'Yb': 1000 ** 8,
3615 'yb': 1000 ** 8,
3616 'yottabytes': 1000 ** 8,
3617 'yobibytes': 1024 ** 8,
3618 }
3619
3620 return lookup_unit_table(_UNIT_TABLE, s)
3621
3622
3623def parse_count(s):
3624 if s is None:
3625 return None
3626
3627 s = s.strip()
3628
3629 if re.match(r'^[\d,.]+$', s):
3630 return str_to_int(s)
3631
3632 _UNIT_TABLE = {
3633 'k': 1000,
3634 'K': 1000,
3635 'm': 1000 ** 2,
3636 'M': 1000 ** 2,
3637 'kk': 1000 ** 2,
3638 'KK': 1000 ** 2,
3639 }
3640
3641 return lookup_unit_table(_UNIT_TABLE, s)
3642
3643
3644def parse_resolution(s):
3645 if s is None:
3646 return {}
3647
3648 mobj = re.search(r'\b(?P<w>\d+)\s*[xX×]\s*(?P<h>\d+)\b', s)
3649 if mobj:
3650 return {
3651 'width': int(mobj.group('w')),
3652 'height': int(mobj.group('h')),
3653 }
3654
3655 mobj = re.search(r'\b(\d+)[pPiI]\b', s)
3656 if mobj:
3657 return {'height': int(mobj.group(1))}
3658
3659 mobj = re.search(r'\b([48])[kK]\b', s)
3660 if mobj:
3661 return {'height': int(mobj.group(1)) * 540}
3662
3663 return {}
3664
3665
3666def parse_bitrate(s):
3667 if not isinstance(s, compat_str):
3668 return
3669 mobj = re.search(r'\b(\d+)\s*kbps', s)
3670 if mobj:
3671 return int(mobj.group(1))
3672
3673
3674def month_by_name(name, lang='en'):
3675 """ Return the number of a month by (locale-independently) English name """
3676
3677 month_names = MONTH_NAMES.get(lang, MONTH_NAMES['en'])
3678
3679 try:
3680 return month_names.index(name) + 1
3681 except ValueError:
3682 return None
3683
3684
3685def month_by_abbreviation(abbrev):
3686 """ Return the number of a month by (locale-independently) English
3687 abbreviations """
3688
3689 try:
3690 return [s[:3] for s in ENGLISH_MONTH_NAMES].index(abbrev) + 1
3691 except ValueError:
3692 return None
3693
3694
3695def fix_xml_ampersands(xml_str):
3696 """Replace all the '&' by '&amp;' in XML"""
3697 return re.sub(
3698 r'&(?!amp;|lt;|gt;|apos;|quot;|#x[0-9a-fA-F]{,4};|#[0-9]{,4};)',
3699 '&amp;',
3700 xml_str)
3701
3702
3703def setproctitle(title):
3704 assert isinstance(title, compat_str)
3705
3706 # ctypes in Jython is not complete
3707 # http://bugs.jython.org/issue2148
3708 if sys.platform.startswith('java'):
3709 return
3710
3711 try:
3712 libc = ctypes.cdll.LoadLibrary('libc.so.6')
3713 except OSError:
3714 return
3715 except TypeError:
3716 # LoadLibrary in Windows Python 2.7.13 only expects
3717 # a bytestring, but since unicode_literals turns
3718 # every string into a unicode string, it fails.
3719 return
3720 title_bytes = title.encode('utf-8')
3721 buf = ctypes.create_string_buffer(len(title_bytes))
3722 buf.value = title_bytes
3723 try:
3724 libc.prctl(15, buf, 0, 0, 0)
3725 except AttributeError:
3726 return # Strange libc, just skip this
3727
3728
3729def remove_start(s, start):
3730 return s[len(start):] if s is not None and s.startswith(start) else s
3731
3732
3733def remove_end(s, end):
3734 return s[:-len(end)] if s is not None and s.endswith(end) else s
3735
3736
3737def remove_quotes(s):
3738 if s is None or len(s) < 2:
3739 return s
3740 for quote in ('"', "'", ):
3741 if s[0] == quote and s[-1] == quote:
3742 return s[1:-1]
3743 return s
3744
3745
3746def get_domain(url):
3747 domain = re.match(r'(?:https?:\/\/)?(?:www\.)?(?P<domain>[^\n\/]+\.[^\n\/]+)(?:\/(.*))?', url)
3748 return domain.group('domain') if domain else None
3749
3750
3751def url_basename(url):
3752 path = compat_urlparse.urlparse(url).path
3753 return path.strip('/').split('/')[-1]
3754
3755
3756def base_url(url):
3757 return re.match(r'https?://[^?#&]+/', url).group()
3758
3759
3760def urljoin(base, path):
3761 if isinstance(path, bytes):
3762 path = path.decode('utf-8')
3763 if not isinstance(path, compat_str) or not path:
3764 return None
3765 if re.match(r'^(?:[a-zA-Z][a-zA-Z0-9+-.]*:)?//', path):
3766 return path
3767 if isinstance(base, bytes):
3768 base = base.decode('utf-8')
3769 if not isinstance(base, compat_str) or not re.match(
3770 r'^(?:https?:)?//', base):
3771 return None
3772 return compat_urlparse.urljoin(base, path)
3773
3774
3775class HEADRequest(compat_urllib_request.Request):
3776 def get_method(self):
3777 return 'HEAD'
3778
3779
3780class PUTRequest(compat_urllib_request.Request):
3781 def get_method(self):
3782 return 'PUT'
3783
3784
3785def int_or_none(v, scale=1, default=None, get_attr=None, invscale=1):
3786 if get_attr:
3787 if v is not None:
3788 v = getattr(v, get_attr, None)
3789 if v == '':
3790 v = None
3791 if v is None:
3792 return default
3793 try:
3794 return int(v) * invscale // scale
3795 except (ValueError, TypeError):
3796 return default
3797
3798
3799def str_or_none(v, default=None):
3800 return default if v is None else compat_str(v)
3801
3802
3803def str_to_int(int_str):
3804 """ A more relaxed version of int_or_none """
3805 if isinstance(int_str, compat_integer_types):
3806 return int_str
3807 elif isinstance(int_str, compat_str):
3808 int_str = re.sub(r'[,\.\+]', '', int_str)
3809 return int_or_none(int_str)
3810
3811
3812def float_or_none(v, scale=1, invscale=1, default=None):
3813 if v is None:
3814 return default
3815 try:
3816 return float(v) * invscale / scale
3817 except (ValueError, TypeError):
3818 return default
3819
3820
3821def bool_or_none(v, default=None):
3822 return v if isinstance(v, bool) else default
3823
3824
3825def strip_or_none(v, default=None):
3826 return v.strip() if isinstance(v, compat_str) else default
3827
3828
3829def url_or_none(url):
3830 if not url or not isinstance(url, compat_str):
3831 return None
3832 url = url.strip()
3833 return url if re.match(r'^(?:(?:https?|rt(?:m(?:pt?[es]?|fp)|sp[su]?)|mms|ftps?):)?//', url) else None
3834
3835
3836def strftime_or_none(timestamp, date_format, default=None):
3837 datetime_object = None
3838 try:
3839 if isinstance(timestamp, compat_numeric_types): # unix timestamp
3840 datetime_object = datetime.datetime.utcfromtimestamp(timestamp)
3841 elif isinstance(timestamp, compat_str): # assume YYYYMMDD
3842 datetime_object = datetime.datetime.strptime(timestamp, '%Y%m%d')
3843 return datetime_object.strftime(date_format)
3844 except (ValueError, TypeError, AttributeError):
3845 return default
3846
3847
3848def parse_duration(s):
3849 if not isinstance(s, compat_basestring):
3850 return None
3851
3852 s = s.strip()
3853
3854 days, hours, mins, secs, ms = [None] * 5
3855 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)
3856 if m:
3857 days, hours, mins, secs, ms = m.groups()
3858 else:
3859 m = re.match(
3860 r'''(?ix)(?:P?
3861 (?:
3862 [0-9]+\s*y(?:ears?)?\s*
3863 )?
3864 (?:
3865 [0-9]+\s*m(?:onths?)?\s*
3866 )?
3867 (?:
3868 [0-9]+\s*w(?:eeks?)?\s*
3869 )?
3870 (?:
3871 (?P<days>[0-9]+)\s*d(?:ays?)?\s*
3872 )?
3873 T)?
3874 (?:
3875 (?P<hours>[0-9]+)\s*h(?:ours?)?\s*
3876 )?
3877 (?:
3878 (?P<mins>[0-9]+)\s*m(?:in(?:ute)?s?)?\s*
3879 )?
3880 (?:
3881 (?P<secs>[0-9]+)(?P<ms>\.[0-9]+)?\s*s(?:ec(?:ond)?s?)?\s*
3882 )?Z?$''', s)
3883 if m:
3884 days, hours, mins, secs, ms = m.groups()
3885 else:
3886 m = re.match(r'(?i)(?:(?P<hours>[0-9.]+)\s*(?:hours?)|(?P<mins>[0-9.]+)\s*(?:mins?\.?|minutes?)\s*)Z?$', s)
3887 if m:
3888 hours, mins = m.groups()
3889 else:
3890 return None
3891
3892 duration = 0
3893 if secs:
3894 duration += float(secs)
3895 if mins:
3896 duration += float(mins) * 60
3897 if hours:
3898 duration += float(hours) * 60 * 60
3899 if days:
3900 duration += float(days) * 24 * 60 * 60
3901 if ms:
3902 duration += float(ms)
3903 return duration
3904
3905
3906def prepend_extension(filename, ext, expected_real_ext=None):
3907 name, real_ext = os.path.splitext(filename)
3908 return (
3909 '{0}.{1}{2}'.format(name, ext, real_ext)
3910 if not expected_real_ext or real_ext[1:] == expected_real_ext
3911 else '{0}.{1}'.format(filename, ext))
3912
3913
3914def replace_extension(filename, ext, expected_real_ext=None):
3915 name, real_ext = os.path.splitext(filename)
3916 return '{0}.{1}'.format(
3917 name if not expected_real_ext or real_ext[1:] == expected_real_ext else filename,
3918 ext)
3919
3920
3921def check_executable(exe, args=[]):
3922 """ Checks if the given binary is installed somewhere in PATH, and returns its name.
3923 args can be a list of arguments for a short output (like -version) """
3924 try:
3925 process_communicate_or_kill(subprocess.Popen(
3926 [exe] + args, stdout=subprocess.PIPE, stderr=subprocess.PIPE))
3927 except OSError:
3928 return False
3929 return exe
3930
3931
3932def get_exe_version(exe, args=['--version'],
3933 version_re=None, unrecognized='present'):
3934 """ Returns the version of the specified executable,
3935 or False if the executable is not present """
3936 try:
3937 # STDIN should be redirected too. On UNIX-like systems, ffmpeg triggers
3938 # SIGTTOU if yt-dlp is run in the background.
3939 # See https://github.com/ytdl-org/youtube-dl/issues/955#issuecomment-209789656
3940 out, _ = process_communicate_or_kill(subprocess.Popen(
3941 [encodeArgument(exe)] + args,
3942 stdin=subprocess.PIPE,
3943 stdout=subprocess.PIPE, stderr=subprocess.STDOUT))
3944 except OSError:
3945 return False
3946 if isinstance(out, bytes): # Python 2.x
3947 out = out.decode('ascii', 'ignore')
3948 return detect_exe_version(out, version_re, unrecognized)
3949
3950
3951def detect_exe_version(output, version_re=None, unrecognized='present'):
3952 assert isinstance(output, compat_str)
3953 if version_re is None:
3954 version_re = r'version\s+([-0-9._a-zA-Z]+)'
3955 m = re.search(version_re, output)
3956 if m:
3957 return m.group(1)
3958 else:
3959 return unrecognized
3960
3961
3962class LazyList(collections.Sequence):
3963 ''' Lazy immutable list from an iterable
3964 Note that slices of a LazyList are lists and not LazyList'''
3965
3966 def __init__(self, iterable):
3967 self.__iterable = iter(iterable)
3968 self.__cache = []
3969 self.__reversed = False
3970
3971 def __iter__(self):
3972 if self.__reversed:
3973 # We need to consume the entire iterable to iterate in reverse
3974 yield from self.exhaust()[::-1]
3975 return
3976 yield from self.__cache
3977 for item in self.__iterable:
3978 self.__cache.append(item)
3979 yield item
3980
3981 def exhaust(self):
3982 ''' Evaluate the entire iterable '''
3983 self.__cache.extend(self.__iterable)
3984 return self.__cache
3985
3986 @staticmethod
3987 def _reverse_index(x):
3988 return -(x + 1)
3989
3990 def __getitem__(self, idx):
3991 if isinstance(idx, slice):
3992 step = idx.step or 1
3993 start = idx.start if idx.start is not None else 0 if step > 0 else -1
3994 stop = idx.stop if idx.stop is not None else -1 if step > 0 else 0
3995 if self.__reversed:
3996 start, stop, step = map(self._reverse_index, (start, stop, step))
3997 idx = slice(start, stop, step)
3998 elif isinstance(idx, int):
3999 if self.__reversed:
4000 idx = self._reverse_index(idx)
4001 start = stop = idx
4002 else:
4003 raise TypeError('indices must be integers or slices')
4004 if start < 0 or stop < 0:
4005 # We need to consume the entire iterable to be able to slice from the end
4006 # Obviously, never use this with infinite iterables
4007 return self.exhaust()[idx]
4008
4009 n = max(start, stop) - len(self.__cache) + 1
4010 if n > 0:
4011 self.__cache.extend(itertools.islice(self.__iterable, n))
4012 return self.__cache[idx]
4013
4014 def __bool__(self):
4015 try:
4016 self[-1] if self.__reversed else self[0]
4017 except IndexError:
4018 return False
4019 return True
4020
4021 def __len__(self):
4022 self.exhaust()
4023 return len(self.__cache)
4024
4025 def __reversed__(self):
4026 self.__reversed = not self.__reversed
4027 return self
4028
4029 def __repr__(self):
4030 # repr and str should mimic a list. So we exhaust the iterable
4031 return repr(self.exhaust())
4032
4033 def __str__(self):
4034 return repr(self.exhaust())
4035
4036
4037class PagedList(object):
4038 def __len__(self):
4039 # This is only useful for tests
4040 return len(self.getslice())
4041
4042 def getslice(self, start, end):
4043 raise NotImplementedError('This method must be implemented by subclasses')
4044
4045 def __getitem__(self, idx):
4046 if not isinstance(idx, int) or idx < 0:
4047 raise TypeError('indices must be non-negative integers')
4048 entries = self.getslice(idx, idx + 1)
4049 return entries[0] if entries else None
4050
4051
4052class OnDemandPagedList(PagedList):
4053 def __init__(self, pagefunc, pagesize, use_cache=True):
4054 self._pagefunc = pagefunc
4055 self._pagesize = pagesize
4056 self._use_cache = use_cache
4057 if use_cache:
4058 self._cache = {}
4059
4060 def getslice(self, start=0, end=None):
4061 res = []
4062 for pagenum in itertools.count(start // self._pagesize):
4063 firstid = pagenum * self._pagesize
4064 nextfirstid = pagenum * self._pagesize + self._pagesize
4065 if start >= nextfirstid:
4066 continue
4067
4068 page_results = None
4069 if self._use_cache:
4070 page_results = self._cache.get(pagenum)
4071 if page_results is None:
4072 page_results = list(self._pagefunc(pagenum))
4073 if self._use_cache:
4074 self._cache[pagenum] = page_results
4075
4076 startv = (
4077 start % self._pagesize
4078 if firstid <= start < nextfirstid
4079 else 0)
4080
4081 endv = (
4082 ((end - 1) % self._pagesize) + 1
4083 if (end is not None and firstid <= end <= nextfirstid)
4084 else None)
4085
4086 if startv != 0 or endv is not None:
4087 page_results = page_results[startv:endv]
4088 res.extend(page_results)
4089
4090 # A little optimization - if current page is not "full", ie. does
4091 # not contain page_size videos then we can assume that this page
4092 # is the last one - there are no more ids on further pages -
4093 # i.e. no need to query again.
4094 if len(page_results) + startv < self._pagesize:
4095 break
4096
4097 # If we got the whole page, but the next page is not interesting,
4098 # break out early as well
4099 if end == nextfirstid:
4100 break
4101 return res
4102
4103
4104class InAdvancePagedList(PagedList):
4105 def __init__(self, pagefunc, pagecount, pagesize):
4106 self._pagefunc = pagefunc
4107 self._pagecount = pagecount
4108 self._pagesize = pagesize
4109
4110 def getslice(self, start=0, end=None):
4111 res = []
4112 start_page = start // self._pagesize
4113 end_page = (
4114 self._pagecount if end is None else (end // self._pagesize + 1))
4115 skip_elems = start - start_page * self._pagesize
4116 only_more = None if end is None else end - start
4117 for pagenum in range(start_page, end_page):
4118 page = list(self._pagefunc(pagenum))
4119 if skip_elems:
4120 page = page[skip_elems:]
4121 skip_elems = None
4122 if only_more is not None:
4123 if len(page) < only_more:
4124 only_more -= len(page)
4125 else:
4126 page = page[:only_more]
4127 res.extend(page)
4128 break
4129 res.extend(page)
4130 return res
4131
4132
4133def uppercase_escape(s):
4134 unicode_escape = codecs.getdecoder('unicode_escape')
4135 return re.sub(
4136 r'\\U[0-9a-fA-F]{8}',
4137 lambda m: unicode_escape(m.group(0))[0],
4138 s)
4139
4140
4141def lowercase_escape(s):
4142 unicode_escape = codecs.getdecoder('unicode_escape')
4143 return re.sub(
4144 r'\\u[0-9a-fA-F]{4}',
4145 lambda m: unicode_escape(m.group(0))[0],
4146 s)
4147
4148
4149def escape_rfc3986(s):
4150 """Escape non-ASCII characters as suggested by RFC 3986"""
4151 if sys.version_info < (3, 0) and isinstance(s, compat_str):
4152 s = s.encode('utf-8')
4153 return compat_urllib_parse.quote(s, b"%/;:@&=+$,!~*'()?#[]")
4154
4155
4156def escape_url(url):
4157 """Escape URL as suggested by RFC 3986"""
4158 url_parsed = compat_urllib_parse_urlparse(url)
4159 return url_parsed._replace(
4160 netloc=url_parsed.netloc.encode('idna').decode('ascii'),
4161 path=escape_rfc3986(url_parsed.path),
4162 params=escape_rfc3986(url_parsed.params),
4163 query=escape_rfc3986(url_parsed.query),
4164 fragment=escape_rfc3986(url_parsed.fragment)
4165 ).geturl()
4166
4167
4168def read_batch_urls(batch_fd):
4169 def fixup(url):
4170 if not isinstance(url, compat_str):
4171 url = url.decode('utf-8', 'replace')
4172 BOM_UTF8 = ('\xef\xbb\xbf', '\ufeff')
4173 for bom in BOM_UTF8:
4174 if url.startswith(bom):
4175 url = url[len(bom):]
4176 url = url.lstrip()
4177 if not url or url.startswith(('#', ';', ']')):
4178 return False
4179 # "#" cannot be stripped out since it is part of the URI
4180 # However, it can be safely stipped out if follwing a whitespace
4181 return re.split(r'\s#', url, 1)[0].rstrip()
4182
4183 with contextlib.closing(batch_fd) as fd:
4184 return [url for url in map(fixup, fd) if url]
4185
4186
4187def urlencode_postdata(*args, **kargs):
4188 return compat_urllib_parse_urlencode(*args, **kargs).encode('ascii')
4189
4190
4191def update_url_query(url, query):
4192 if not query:
4193 return url
4194 parsed_url = compat_urlparse.urlparse(url)
4195 qs = compat_parse_qs(parsed_url.query)
4196 qs.update(query)
4197 return compat_urlparse.urlunparse(parsed_url._replace(
4198 query=compat_urllib_parse_urlencode(qs, True)))
4199
4200
4201def update_Request(req, url=None, data=None, headers={}, query={}):
4202 req_headers = req.headers.copy()
4203 req_headers.update(headers)
4204 req_data = data or req.data
4205 req_url = update_url_query(url or req.get_full_url(), query)
4206 req_get_method = req.get_method()
4207 if req_get_method == 'HEAD':
4208 req_type = HEADRequest
4209 elif req_get_method == 'PUT':
4210 req_type = PUTRequest
4211 else:
4212 req_type = compat_urllib_request.Request
4213 new_req = req_type(
4214 req_url, data=req_data, headers=req_headers,
4215 origin_req_host=req.origin_req_host, unverifiable=req.unverifiable)
4216 if hasattr(req, 'timeout'):
4217 new_req.timeout = req.timeout
4218 return new_req
4219
4220
4221def _multipart_encode_impl(data, boundary):
4222 content_type = 'multipart/form-data; boundary=%s' % boundary
4223
4224 out = b''
4225 for k, v in data.items():
4226 out += b'--' + boundary.encode('ascii') + b'\r\n'
4227 if isinstance(k, compat_str):
4228 k = k.encode('utf-8')
4229 if isinstance(v, compat_str):
4230 v = v.encode('utf-8')
4231 # RFC 2047 requires non-ASCII field names to be encoded, while RFC 7578
4232 # suggests sending UTF-8 directly. Firefox sends UTF-8, too
4233 content = b'Content-Disposition: form-data; name="' + k + b'"\r\n\r\n' + v + b'\r\n'
4234 if boundary.encode('ascii') in content:
4235 raise ValueError('Boundary overlaps with data')
4236 out += content
4237
4238 out += b'--' + boundary.encode('ascii') + b'--\r\n'
4239
4240 return out, content_type
4241
4242
4243def multipart_encode(data, boundary=None):
4244 '''
4245 Encode a dict to RFC 7578-compliant form-data
4246
4247 data:
4248 A dict where keys and values can be either Unicode or bytes-like
4249 objects.
4250 boundary:
4251 If specified a Unicode object, it's used as the boundary. Otherwise
4252 a random boundary is generated.
4253
4254 Reference: https://tools.ietf.org/html/rfc7578
4255 '''
4256 has_specified_boundary = boundary is not None
4257
4258 while True:
4259 if boundary is None:
4260 boundary = '---------------' + str(random.randrange(0x0fffffff, 0xffffffff))
4261
4262 try:
4263 out, content_type = _multipart_encode_impl(data, boundary)
4264 break
4265 except ValueError:
4266 if has_specified_boundary:
4267 raise
4268 boundary = None
4269
4270 return out, content_type
4271
4272
4273def dict_get(d, key_or_keys, default=None, skip_false_values=True):
4274 if isinstance(key_or_keys, (list, tuple)):
4275 for key in key_or_keys:
4276 if key not in d or d[key] is None or skip_false_values and not d[key]:
4277 continue
4278 return d[key]
4279 return default
4280 return d.get(key_or_keys, default)
4281
4282
4283def try_get(src, getter, expected_type=None):
4284 if not isinstance(getter, (list, tuple)):
4285 getter = [getter]
4286 for get in getter:
4287 try:
4288 v = get(src)
4289 except (AttributeError, KeyError, TypeError, IndexError):
4290 pass
4291 else:
4292 if expected_type is None or isinstance(v, expected_type):
4293 return v
4294
4295
4296def merge_dicts(*dicts):
4297 merged = {}
4298 for a_dict in dicts:
4299 for k, v in a_dict.items():
4300 if v is None:
4301 continue
4302 if (k not in merged
4303 or (isinstance(v, compat_str) and v
4304 and isinstance(merged[k], compat_str)
4305 and not merged[k])):
4306 merged[k] = v
4307 return merged
4308
4309
4310def encode_compat_str(string, encoding=preferredencoding(), errors='strict'):
4311 return string if isinstance(string, compat_str) else compat_str(string, encoding, errors)
4312
4313
4314US_RATINGS = {
4315 'G': 0,
4316 'PG': 10,
4317 'PG-13': 13,
4318 'R': 16,
4319 'NC': 18,
4320}
4321
4322
4323TV_PARENTAL_GUIDELINES = {
4324 'TV-Y': 0,
4325 'TV-Y7': 7,
4326 'TV-G': 0,
4327 'TV-PG': 0,
4328 'TV-14': 14,
4329 'TV-MA': 17,
4330}
4331
4332
4333def parse_age_limit(s):
4334 if type(s) == int:
4335 return s if 0 <= s <= 21 else None
4336 if not isinstance(s, compat_basestring):
4337 return None
4338 m = re.match(r'^(?P<age>\d{1,2})\+?$', s)
4339 if m:
4340 return int(m.group('age'))
4341 s = s.upper()
4342 if s in US_RATINGS:
4343 return US_RATINGS[s]
4344 m = re.match(r'^TV[_-]?(%s)$' % '|'.join(k[3:] for k in TV_PARENTAL_GUIDELINES), s)
4345 if m:
4346 return TV_PARENTAL_GUIDELINES['TV-' + m.group(1)]
4347 return None
4348
4349
4350def strip_jsonp(code):
4351 return re.sub(
4352 r'''(?sx)^
4353 (?:window\.)?(?P<func_name>[a-zA-Z0-9_.$]*)
4354 (?:\s*&&\s*(?P=func_name))?
4355 \s*\(\s*(?P<callback_data>.*)\);?
4356 \s*?(?://[^\n]*)*$''',
4357 r'\g<callback_data>', code)
4358
4359
4360def js_to_json(code, vars={}):
4361 # vars is a dict of var, val pairs to substitute
4362 COMMENT_RE = r'/\*(?:(?!\*/).)*?\*/|//[^\n]*'
4363 SKIP_RE = r'\s*(?:{comment})?\s*'.format(comment=COMMENT_RE)
4364 INTEGER_TABLE = (
4365 (r'(?s)^(0[xX][0-9a-fA-F]+){skip}:?$'.format(skip=SKIP_RE), 16),
4366 (r'(?s)^(0+[0-7]+){skip}:?$'.format(skip=SKIP_RE), 8),
4367 )
4368
4369 def fix_kv(m):
4370 v = m.group(0)
4371 if v in ('true', 'false', 'null'):
4372 return v
4373 elif v.startswith('/*') or v.startswith('//') or v.startswith('!') or v == ',':
4374 return ""
4375
4376 if v[0] in ("'", '"'):
4377 v = re.sub(r'(?s)\\.|"', lambda m: {
4378 '"': '\\"',
4379 "\\'": "'",
4380 '\\\n': '',
4381 '\\x': '\\u00',
4382 }.get(m.group(0), m.group(0)), v[1:-1])
4383 else:
4384 for regex, base in INTEGER_TABLE:
4385 im = re.match(regex, v)
4386 if im:
4387 i = int(im.group(1), base)
4388 return '"%d":' % i if v.endswith(':') else '%d' % i
4389
4390 if v in vars:
4391 return vars[v]
4392
4393 return '"%s"' % v
4394
4395 return re.sub(r'''(?sx)
4396 "(?:[^"\\]*(?:\\\\|\\['"nurtbfx/\n]))*[^"\\]*"|
4397 '(?:[^'\\]*(?:\\\\|\\['"nurtbfx/\n]))*[^'\\]*'|
4398 {comment}|,(?={skip}[\]}}])|
4399 (?:(?<![0-9])[eE]|[a-df-zA-DF-Z_])[.a-zA-Z_0-9]*|
4400 \b(?:0[xX][0-9a-fA-F]+|0+[0-7]+)(?:{skip}:)?|
4401 [0-9]+(?={skip}:)|
4402 !+
4403 '''.format(comment=COMMENT_RE, skip=SKIP_RE), fix_kv, code)
4404
4405
4406def qualities(quality_ids):
4407 """ Get a numeric quality value out of a list of possible values """
4408 def q(qid):
4409 try:
4410 return quality_ids.index(qid)
4411 except ValueError:
4412 return -1
4413 return q
4414
4415
4416DEFAULT_OUTTMPL = {
4417 'default': '%(title)s [%(id)s].%(ext)s',
4418 'chapter': '%(title)s - %(section_number)03d %(section_title)s [%(id)s].%(ext)s',
4419}
4420OUTTMPL_TYPES = {
4421 'chapter': None,
4422 'subtitle': None,
4423 'thumbnail': None,
4424 'description': 'description',
4425 'annotation': 'annotations.xml',
4426 'infojson': 'info.json',
4427 'pl_thumbnail': None,
4428 'pl_description': 'description',
4429 'pl_infojson': 'info.json',
4430}
4431
4432# As of [1] format syntax is:
4433# %[mapping_key][conversion_flags][minimum_width][.precision][length_modifier]type
4434# 1. https://docs.python.org/2/library/stdtypes.html#string-formatting
4435STR_FORMAT_RE = r'''(?x)
4436 (?<!%)
4437 %
4438 (?P<has_key>\((?P<key>{0})\))? # mapping key
4439 (?P<format>
4440 (?:[#0\-+ ]+)? # conversion flags (optional)
4441 (?:\d+)? # minimum field width (optional)
4442 (?:\.\d+)? # precision (optional)
4443 [hlL]? # length modifier (optional)
4444 [diouxXeEfFgGcrs] # conversion type
4445 )
4446'''
4447
4448
4449def limit_length(s, length):
4450 """ Add ellipses to overly long strings """
4451 if s is None:
4452 return None
4453 ELLIPSES = '...'
4454 if len(s) > length:
4455 return s[:length - len(ELLIPSES)] + ELLIPSES
4456 return s
4457
4458
4459def version_tuple(v):
4460 return tuple(int(e) for e in re.split(r'[-.]', v))
4461
4462
4463def is_outdated_version(version, limit, assume_new=True):
4464 if not version:
4465 return not assume_new
4466 try:
4467 return version_tuple(version) < version_tuple(limit)
4468 except ValueError:
4469 return not assume_new
4470
4471
4472def ytdl_is_updateable():
4473 """ Returns if yt-dlp can be updated with -U """
4474 return False
4475
4476 from zipimport import zipimporter
4477
4478 return isinstance(globals().get('__loader__'), zipimporter) or hasattr(sys, 'frozen')
4479
4480
4481def args_to_str(args):
4482 # Get a short string representation for a subprocess command
4483 return ' '.join(compat_shlex_quote(a) for a in args)
4484
4485
4486def error_to_compat_str(err):
4487 err_str = str(err)
4488 # On python 2 error byte string must be decoded with proper
4489 # encoding rather than ascii
4490 if sys.version_info[0] < 3:
4491 err_str = err_str.decode(preferredencoding())
4492 return err_str
4493
4494
4495def mimetype2ext(mt):
4496 if mt is None:
4497 return None
4498
4499 ext = {
4500 'audio/mp4': 'm4a',
4501 # Per RFC 3003, audio/mpeg can be .mp1, .mp2 or .mp3. Here use .mp3 as
4502 # it's the most popular one
4503 'audio/mpeg': 'mp3',
4504 'audio/x-wav': 'wav',
4505 }.get(mt)
4506 if ext is not None:
4507 return ext
4508
4509 _, _, res = mt.rpartition('/')
4510 res = res.split(';')[0].strip().lower()
4511
4512 return {
4513 '3gpp': '3gp',
4514 'smptett+xml': 'tt',
4515 'ttaf+xml': 'dfxp',
4516 'ttml+xml': 'ttml',
4517 'x-flv': 'flv',
4518 'x-mp4-fragmented': 'mp4',
4519 'x-ms-sami': 'sami',
4520 'x-ms-wmv': 'wmv',
4521 'mpegurl': 'm3u8',
4522 'x-mpegurl': 'm3u8',
4523 'vnd.apple.mpegurl': 'm3u8',
4524 'dash+xml': 'mpd',
4525 'f4m+xml': 'f4m',
4526 'hds+xml': 'f4m',
4527 'vnd.ms-sstr+xml': 'ism',
4528 'quicktime': 'mov',
4529 'mp2t': 'ts',
4530 'x-wav': 'wav',
4531 }.get(res, res)
4532
4533
4534def parse_codecs(codecs_str):
4535 # http://tools.ietf.org/html/rfc6381
4536 if not codecs_str:
4537 return {}
4538 split_codecs = list(filter(None, map(
4539 lambda str: str.strip(), codecs_str.strip().strip(',').split(','))))
4540 vcodec, acodec = None, None
4541 for full_codec in split_codecs:
4542 codec = full_codec.split('.')[0]
4543 if codec in ('avc1', 'avc2', 'avc3', 'avc4', 'vp9', 'vp8', 'hev1', 'hev2', 'h263', 'h264', 'mp4v', 'hvc1', 'av01', 'theora'):
4544 if not vcodec:
4545 vcodec = full_codec
4546 elif codec in ('mp4a', 'opus', 'vorbis', 'mp3', 'aac', 'ac-3', 'ec-3', 'eac3', 'dtsc', 'dtse', 'dtsh', 'dtsl'):
4547 if not acodec:
4548 acodec = full_codec
4549 else:
4550 write_string('WARNING: Unknown codec %s\n' % full_codec, sys.stderr)
4551 if not vcodec and not acodec:
4552 if len(split_codecs) == 2:
4553 return {
4554 'vcodec': split_codecs[0],
4555 'acodec': split_codecs[1],
4556 }
4557 else:
4558 return {
4559 'vcodec': vcodec or 'none',
4560 'acodec': acodec or 'none',
4561 }
4562 return {}
4563
4564
4565def urlhandle_detect_ext(url_handle):
4566 getheader = url_handle.headers.get
4567
4568 cd = getheader('Content-Disposition')
4569 if cd:
4570 m = re.match(r'attachment;\s*filename="(?P<filename>[^"]+)"', cd)
4571 if m:
4572 e = determine_ext(m.group('filename'), default_ext=None)
4573 if e:
4574 return e
4575
4576 return mimetype2ext(getheader('Content-Type'))
4577
4578
4579def encode_data_uri(data, mime_type):
4580 return 'data:%s;base64,%s' % (mime_type, base64.b64encode(data).decode('ascii'))
4581
4582
4583def age_restricted(content_limit, age_limit):
4584 """ Returns True iff the content should be blocked """
4585
4586 if age_limit is None: # No limit set
4587 return False
4588 if content_limit is None:
4589 return False # Content available for everyone
4590 return age_limit < content_limit
4591
4592
4593def is_html(first_bytes):
4594 """ Detect whether a file contains HTML by examining its first bytes. """
4595
4596 BOMS = [
4597 (b'\xef\xbb\xbf', 'utf-8'),
4598 (b'\x00\x00\xfe\xff', 'utf-32-be'),
4599 (b'\xff\xfe\x00\x00', 'utf-32-le'),
4600 (b'\xff\xfe', 'utf-16-le'),
4601 (b'\xfe\xff', 'utf-16-be'),
4602 ]
4603 for bom, enc in BOMS:
4604 if first_bytes.startswith(bom):
4605 s = first_bytes[len(bom):].decode(enc, 'replace')
4606 break
4607 else:
4608 s = first_bytes.decode('utf-8', 'replace')
4609
4610 return re.match(r'^\s*<', s)
4611
4612
4613def determine_protocol(info_dict):
4614 protocol = info_dict.get('protocol')
4615 if protocol is not None:
4616 return protocol
4617
4618 url = info_dict['url']
4619 if url.startswith('rtmp'):
4620 return 'rtmp'
4621 elif url.startswith('mms'):
4622 return 'mms'
4623 elif url.startswith('rtsp'):
4624 return 'rtsp'
4625
4626 ext = determine_ext(url)
4627 if ext == 'm3u8':
4628 return 'm3u8'
4629 elif ext == 'f4m':
4630 return 'f4m'
4631
4632 return compat_urllib_parse_urlparse(url).scheme
4633
4634
4635def render_table(header_row, data, delim=False, extraGap=0, hideEmpty=False):
4636 """ Render a list of rows, each as a list of values """
4637
4638 def get_max_lens(table):
4639 return [max(len(compat_str(v)) for v in col) for col in zip(*table)]
4640
4641 def filter_using_list(row, filterArray):
4642 return [col for (take, col) in zip(filterArray, row) if take]
4643
4644 if hideEmpty:
4645 max_lens = get_max_lens(data)
4646 header_row = filter_using_list(header_row, max_lens)
4647 data = [filter_using_list(row, max_lens) for row in data]
4648
4649 table = [header_row] + data
4650 max_lens = get_max_lens(table)
4651 if delim:
4652 table = [header_row] + [['-' * ml for ml in max_lens]] + data
4653 format_str = ' '.join('%-' + compat_str(ml + extraGap) + 's' for ml in max_lens[:-1]) + ' %s'
4654 return '\n'.join(format_str % tuple(row) for row in table)
4655
4656
4657def _match_one(filter_part, dct):
4658 COMPARISON_OPERATORS = {
4659 '<': operator.lt,
4660 '<=': operator.le,
4661 '>': operator.gt,
4662 '>=': operator.ge,
4663 '=': operator.eq,
4664 '!=': operator.ne,
4665 }
4666 operator_rex = re.compile(r'''(?x)\s*
4667 (?P<key>[a-z_]+)
4668 \s*(?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
4669 (?:
4670 (?P<intval>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)|
4671 (?P<quote>["\'])(?P<quotedstrval>(?:\\.|(?!(?P=quote)|\\).)+?)(?P=quote)|
4672 (?P<strval>(?![0-9.])[a-z0-9A-Z]*)
4673 )
4674 \s*$
4675 ''' % '|'.join(map(re.escape, COMPARISON_OPERATORS.keys())))
4676 m = operator_rex.search(filter_part)
4677 if m:
4678 op = COMPARISON_OPERATORS[m.group('op')]
4679 actual_value = dct.get(m.group('key'))
4680 if (m.group('quotedstrval') is not None
4681 or m.group('strval') is not None
4682 # If the original field is a string and matching comparisonvalue is
4683 # a number we should respect the origin of the original field
4684 # and process comparison value as a string (see
4685 # https://github.com/ytdl-org/youtube-dl/issues/11082).
4686 or actual_value is not None and m.group('intval') is not None
4687 and isinstance(actual_value, compat_str)):
4688 if m.group('op') not in ('=', '!='):
4689 raise ValueError(
4690 'Operator %s does not support string values!' % m.group('op'))
4691 comparison_value = m.group('quotedstrval') or m.group('strval') or m.group('intval')
4692 quote = m.group('quote')
4693 if quote is not None:
4694 comparison_value = comparison_value.replace(r'\%s' % quote, quote)
4695 else:
4696 try:
4697 comparison_value = int(m.group('intval'))
4698 except ValueError:
4699 comparison_value = parse_filesize(m.group('intval'))
4700 if comparison_value is None:
4701 comparison_value = parse_filesize(m.group('intval') + 'B')
4702 if comparison_value is None:
4703 raise ValueError(
4704 'Invalid integer value %r in filter part %r' % (
4705 m.group('intval'), filter_part))
4706 if actual_value is None:
4707 return m.group('none_inclusive')
4708 return op(actual_value, comparison_value)
4709
4710 UNARY_OPERATORS = {
4711 '': lambda v: (v is True) if isinstance(v, bool) else (v is not None),
4712 '!': lambda v: (v is False) if isinstance(v, bool) else (v is None),
4713 }
4714 operator_rex = re.compile(r'''(?x)\s*
4715 (?P<op>%s)\s*(?P<key>[a-z_]+)
4716 \s*$
4717 ''' % '|'.join(map(re.escape, UNARY_OPERATORS.keys())))
4718 m = operator_rex.search(filter_part)
4719 if m:
4720 op = UNARY_OPERATORS[m.group('op')]
4721 actual_value = dct.get(m.group('key'))
4722 return op(actual_value)
4723
4724 raise ValueError('Invalid filter part %r' % filter_part)
4725
4726
4727def match_str(filter_str, dct):
4728 """ Filter a dictionary with a simple string syntax. Returns True (=passes filter) or false """
4729
4730 return all(
4731 _match_one(filter_part, dct) for filter_part in filter_str.split('&'))
4732
4733
4734def match_filter_func(filter_str):
4735 def _match_func(info_dict):
4736 if match_str(filter_str, info_dict):
4737 return None
4738 else:
4739 video_title = info_dict.get('title', info_dict.get('id', 'video'))
4740 return '%s does not pass filter %s, skipping ..' % (video_title, filter_str)
4741 return _match_func
4742
4743
4744def parse_dfxp_time_expr(time_expr):
4745 if not time_expr:
4746 return
4747
4748 mobj = re.match(r'^(?P<time_offset>\d+(?:\.\d+)?)s?$', time_expr)
4749 if mobj:
4750 return float(mobj.group('time_offset'))
4751
4752 mobj = re.match(r'^(\d+):(\d\d):(\d\d(?:(?:\.|:)\d+)?)$', time_expr)
4753 if mobj:
4754 return 3600 * int(mobj.group(1)) + 60 * int(mobj.group(2)) + float(mobj.group(3).replace(':', '.'))
4755
4756
4757def srt_subtitles_timecode(seconds):
4758 return '%02d:%02d:%02d,%03d' % (seconds / 3600, (seconds % 3600) / 60, seconds % 60, (seconds % 1) * 1000)
4759
4760
4761def dfxp2srt(dfxp_data):
4762 '''
4763 @param dfxp_data A bytes-like object containing DFXP data
4764 @returns A unicode object containing converted SRT data
4765 '''
4766 LEGACY_NAMESPACES = (
4767 (b'http://www.w3.org/ns/ttml', [
4768 b'http://www.w3.org/2004/11/ttaf1',
4769 b'http://www.w3.org/2006/04/ttaf1',
4770 b'http://www.w3.org/2006/10/ttaf1',
4771 ]),
4772 (b'http://www.w3.org/ns/ttml#styling', [
4773 b'http://www.w3.org/ns/ttml#style',
4774 ]),
4775 )
4776
4777 SUPPORTED_STYLING = [
4778 'color',
4779 'fontFamily',
4780 'fontSize',
4781 'fontStyle',
4782 'fontWeight',
4783 'textDecoration'
4784 ]
4785
4786 _x = functools.partial(xpath_with_ns, ns_map={
4787 'xml': 'http://www.w3.org/XML/1998/namespace',
4788 'ttml': 'http://www.w3.org/ns/ttml',
4789 'tts': 'http://www.w3.org/ns/ttml#styling',
4790 })
4791
4792 styles = {}
4793 default_style = {}
4794
4795 class TTMLPElementParser(object):
4796 _out = ''
4797 _unclosed_elements = []
4798 _applied_styles = []
4799
4800 def start(self, tag, attrib):
4801 if tag in (_x('ttml:br'), 'br'):
4802 self._out += '\n'
4803 else:
4804 unclosed_elements = []
4805 style = {}
4806 element_style_id = attrib.get('style')
4807 if default_style:
4808 style.update(default_style)
4809 if element_style_id:
4810 style.update(styles.get(element_style_id, {}))
4811 for prop in SUPPORTED_STYLING:
4812 prop_val = attrib.get(_x('tts:' + prop))
4813 if prop_val:
4814 style[prop] = prop_val
4815 if style:
4816 font = ''
4817 for k, v in sorted(style.items()):
4818 if self._applied_styles and self._applied_styles[-1].get(k) == v:
4819 continue
4820 if k == 'color':
4821 font += ' color="%s"' % v
4822 elif k == 'fontSize':
4823 font += ' size="%s"' % v
4824 elif k == 'fontFamily':
4825 font += ' face="%s"' % v
4826 elif k == 'fontWeight' and v == 'bold':
4827 self._out += '<b>'
4828 unclosed_elements.append('b')
4829 elif k == 'fontStyle' and v == 'italic':
4830 self._out += '<i>'
4831 unclosed_elements.append('i')
4832 elif k == 'textDecoration' and v == 'underline':
4833 self._out += '<u>'
4834 unclosed_elements.append('u')
4835 if font:
4836 self._out += '<font' + font + '>'
4837 unclosed_elements.append('font')
4838 applied_style = {}
4839 if self._applied_styles:
4840 applied_style.update(self._applied_styles[-1])
4841 applied_style.update(style)
4842 self._applied_styles.append(applied_style)
4843 self._unclosed_elements.append(unclosed_elements)
4844
4845 def end(self, tag):
4846 if tag not in (_x('ttml:br'), 'br'):
4847 unclosed_elements = self._unclosed_elements.pop()
4848 for element in reversed(unclosed_elements):
4849 self._out += '</%s>' % element
4850 if unclosed_elements and self._applied_styles:
4851 self._applied_styles.pop()
4852
4853 def data(self, data):
4854 self._out += data
4855
4856 def close(self):
4857 return self._out.strip()
4858
4859 def parse_node(node):
4860 target = TTMLPElementParser()
4861 parser = xml.etree.ElementTree.XMLParser(target=target)
4862 parser.feed(xml.etree.ElementTree.tostring(node))
4863 return parser.close()
4864
4865 for k, v in LEGACY_NAMESPACES:
4866 for ns in v:
4867 dfxp_data = dfxp_data.replace(ns, k)
4868
4869 dfxp = compat_etree_fromstring(dfxp_data)
4870 out = []
4871 paras = dfxp.findall(_x('.//ttml:p')) or dfxp.findall('.//p')
4872
4873 if not paras:
4874 raise ValueError('Invalid dfxp/TTML subtitle')
4875
4876 repeat = False
4877 while True:
4878 for style in dfxp.findall(_x('.//ttml:style')):
4879 style_id = style.get('id') or style.get(_x('xml:id'))
4880 if not style_id:
4881 continue
4882 parent_style_id = style.get('style')
4883 if parent_style_id:
4884 if parent_style_id not in styles:
4885 repeat = True
4886 continue
4887 styles[style_id] = styles[parent_style_id].copy()
4888 for prop in SUPPORTED_STYLING:
4889 prop_val = style.get(_x('tts:' + prop))
4890 if prop_val:
4891 styles.setdefault(style_id, {})[prop] = prop_val
4892 if repeat:
4893 repeat = False
4894 else:
4895 break
4896
4897 for p in ('body', 'div'):
4898 ele = xpath_element(dfxp, [_x('.//ttml:' + p), './/' + p])
4899 if ele is None:
4900 continue
4901 style = styles.get(ele.get('style'))
4902 if not style:
4903 continue
4904 default_style.update(style)
4905
4906 for para, index in zip(paras, itertools.count(1)):
4907 begin_time = parse_dfxp_time_expr(para.attrib.get('begin'))
4908 end_time = parse_dfxp_time_expr(para.attrib.get('end'))
4909 dur = parse_dfxp_time_expr(para.attrib.get('dur'))
4910 if begin_time is None:
4911 continue
4912 if not end_time:
4913 if not dur:
4914 continue
4915 end_time = begin_time + dur
4916 out.append('%d\n%s --> %s\n%s\n\n' % (
4917 index,
4918 srt_subtitles_timecode(begin_time),
4919 srt_subtitles_timecode(end_time),
4920 parse_node(para)))
4921
4922 return ''.join(out)
4923
4924
4925def cli_option(params, command_option, param):
4926 param = params.get(param)
4927 if param:
4928 param = compat_str(param)
4929 return [command_option, param] if param is not None else []
4930
4931
4932def cli_bool_option(params, command_option, param, true_value='true', false_value='false', separator=None):
4933 param = params.get(param)
4934 if param is None:
4935 return []
4936 assert isinstance(param, bool)
4937 if separator:
4938 return [command_option + separator + (true_value if param else false_value)]
4939 return [command_option, true_value if param else false_value]
4940
4941
4942def cli_valueless_option(params, command_option, param, expected_value=True):
4943 param = params.get(param)
4944 return [command_option] if param == expected_value else []
4945
4946
4947def cli_configuration_args(argdict, keys, default=[], use_compat=True):
4948 if isinstance(argdict, (list, tuple)): # for backward compatibility
4949 if use_compat:
4950 return argdict
4951 else:
4952 argdict = None
4953 if argdict is None:
4954 return default
4955 assert isinstance(argdict, dict)
4956
4957 assert isinstance(keys, (list, tuple))
4958 for key_list in keys:
4959 if isinstance(key_list, compat_str):
4960 key_list = (key_list,)
4961 arg_list = list(filter(
4962 lambda x: x is not None,
4963 [argdict.get(key.lower()) for key in key_list]))
4964 if arg_list:
4965 return [arg for args in arg_list for arg in args]
4966 return default
4967
4968
4969class ISO639Utils(object):
4970 # See http://www.loc.gov/standards/iso639-2/ISO-639-2_utf-8.txt
4971 _lang_map = {
4972 'aa': 'aar',
4973 'ab': 'abk',
4974 'ae': 'ave',
4975 'af': 'afr',
4976 'ak': 'aka',
4977 'am': 'amh',
4978 'an': 'arg',
4979 'ar': 'ara',
4980 'as': 'asm',
4981 'av': 'ava',
4982 'ay': 'aym',
4983 'az': 'aze',
4984 'ba': 'bak',
4985 'be': 'bel',
4986 'bg': 'bul',
4987 'bh': 'bih',
4988 'bi': 'bis',
4989 'bm': 'bam',
4990 'bn': 'ben',
4991 'bo': 'bod',
4992 'br': 'bre',
4993 'bs': 'bos',
4994 'ca': 'cat',
4995 'ce': 'che',
4996 'ch': 'cha',
4997 'co': 'cos',
4998 'cr': 'cre',
4999 'cs': 'ces',
5000 'cu': 'chu',
5001 'cv': 'chv',
5002 'cy': 'cym',
5003 'da': 'dan',
5004 'de': 'deu',
5005 'dv': 'div',
5006 'dz': 'dzo',
5007 'ee': 'ewe',
5008 'el': 'ell',
5009 'en': 'eng',
5010 'eo': 'epo',
5011 'es': 'spa',
5012 'et': 'est',
5013 'eu': 'eus',
5014 'fa': 'fas',
5015 'ff': 'ful',
5016 'fi': 'fin',
5017 'fj': 'fij',
5018 'fo': 'fao',
5019 'fr': 'fra',
5020 'fy': 'fry',
5021 'ga': 'gle',
5022 'gd': 'gla',
5023 'gl': 'glg',
5024 'gn': 'grn',
5025 'gu': 'guj',
5026 'gv': 'glv',
5027 'ha': 'hau',
5028 'he': 'heb',
5029 'iw': 'heb', # Replaced by he in 1989 revision
5030 'hi': 'hin',
5031 'ho': 'hmo',
5032 'hr': 'hrv',
5033 'ht': 'hat',
5034 'hu': 'hun',
5035 'hy': 'hye',
5036 'hz': 'her',
5037 'ia': 'ina',
5038 'id': 'ind',
5039 'in': 'ind', # Replaced by id in 1989 revision
5040 'ie': 'ile',
5041 'ig': 'ibo',
5042 'ii': 'iii',
5043 'ik': 'ipk',
5044 'io': 'ido',
5045 'is': 'isl',
5046 'it': 'ita',
5047 'iu': 'iku',
5048 'ja': 'jpn',
5049 'jv': 'jav',
5050 'ka': 'kat',
5051 'kg': 'kon',
5052 'ki': 'kik',
5053 'kj': 'kua',
5054 'kk': 'kaz',
5055 'kl': 'kal',
5056 'km': 'khm',
5057 'kn': 'kan',
5058 'ko': 'kor',
5059 'kr': 'kau',
5060 'ks': 'kas',
5061 'ku': 'kur',
5062 'kv': 'kom',
5063 'kw': 'cor',
5064 'ky': 'kir',
5065 'la': 'lat',
5066 'lb': 'ltz',
5067 'lg': 'lug',
5068 'li': 'lim',
5069 'ln': 'lin',
5070 'lo': 'lao',
5071 'lt': 'lit',
5072 'lu': 'lub',
5073 'lv': 'lav',
5074 'mg': 'mlg',
5075 'mh': 'mah',
5076 'mi': 'mri',
5077 'mk': 'mkd',
5078 'ml': 'mal',
5079 'mn': 'mon',
5080 'mr': 'mar',
5081 'ms': 'msa',
5082 'mt': 'mlt',
5083 'my': 'mya',
5084 'na': 'nau',
5085 'nb': 'nob',
5086 'nd': 'nde',
5087 'ne': 'nep',
5088 'ng': 'ndo',
5089 'nl': 'nld',
5090 'nn': 'nno',
5091 'no': 'nor',
5092 'nr': 'nbl',
5093 'nv': 'nav',
5094 'ny': 'nya',
5095 'oc': 'oci',
5096 'oj': 'oji',
5097 'om': 'orm',
5098 'or': 'ori',
5099 'os': 'oss',
5100 'pa': 'pan',
5101 'pi': 'pli',
5102 'pl': 'pol',
5103 'ps': 'pus',
5104 'pt': 'por',
5105 'qu': 'que',
5106 'rm': 'roh',
5107 'rn': 'run',
5108 'ro': 'ron',
5109 'ru': 'rus',
5110 'rw': 'kin',
5111 'sa': 'san',
5112 'sc': 'srd',
5113 'sd': 'snd',
5114 'se': 'sme',
5115 'sg': 'sag',
5116 'si': 'sin',
5117 'sk': 'slk',
5118 'sl': 'slv',
5119 'sm': 'smo',
5120 'sn': 'sna',
5121 'so': 'som',
5122 'sq': 'sqi',
5123 'sr': 'srp',
5124 'ss': 'ssw',
5125 'st': 'sot',
5126 'su': 'sun',
5127 'sv': 'swe',
5128 'sw': 'swa',
5129 'ta': 'tam',
5130 'te': 'tel',
5131 'tg': 'tgk',
5132 'th': 'tha',
5133 'ti': 'tir',
5134 'tk': 'tuk',
5135 'tl': 'tgl',
5136 'tn': 'tsn',
5137 'to': 'ton',
5138 'tr': 'tur',
5139 'ts': 'tso',
5140 'tt': 'tat',
5141 'tw': 'twi',
5142 'ty': 'tah',
5143 'ug': 'uig',
5144 'uk': 'ukr',
5145 'ur': 'urd',
5146 'uz': 'uzb',
5147 've': 'ven',
5148 'vi': 'vie',
5149 'vo': 'vol',
5150 'wa': 'wln',
5151 'wo': 'wol',
5152 'xh': 'xho',
5153 'yi': 'yid',
5154 'ji': 'yid', # Replaced by yi in 1989 revision
5155 'yo': 'yor',
5156 'za': 'zha',
5157 'zh': 'zho',
5158 'zu': 'zul',
5159 }
5160
5161 @classmethod
5162 def short2long(cls, code):
5163 """Convert language code from ISO 639-1 to ISO 639-2/T"""
5164 return cls._lang_map.get(code[:2])
5165
5166 @classmethod
5167 def long2short(cls, code):
5168 """Convert language code from ISO 639-2/T to ISO 639-1"""
5169 for short_name, long_name in cls._lang_map.items():
5170 if long_name == code:
5171 return short_name
5172
5173
5174class ISO3166Utils(object):
5175 # From http://data.okfn.org/data/core/country-list
5176 _country_map = {
5177 'AF': 'Afghanistan',
5178 'AX': 'Åland Islands',
5179 'AL': 'Albania',
5180 'DZ': 'Algeria',
5181 'AS': 'American Samoa',
5182 'AD': 'Andorra',
5183 'AO': 'Angola',
5184 'AI': 'Anguilla',
5185 'AQ': 'Antarctica',
5186 'AG': 'Antigua and Barbuda',
5187 'AR': 'Argentina',
5188 'AM': 'Armenia',
5189 'AW': 'Aruba',
5190 'AU': 'Australia',
5191 'AT': 'Austria',
5192 'AZ': 'Azerbaijan',
5193 'BS': 'Bahamas',
5194 'BH': 'Bahrain',
5195 'BD': 'Bangladesh',
5196 'BB': 'Barbados',
5197 'BY': 'Belarus',
5198 'BE': 'Belgium',
5199 'BZ': 'Belize',
5200 'BJ': 'Benin',
5201 'BM': 'Bermuda',
5202 'BT': 'Bhutan',
5203 'BO': 'Bolivia, Plurinational State of',
5204 'BQ': 'Bonaire, Sint Eustatius and Saba',
5205 'BA': 'Bosnia and Herzegovina',
5206 'BW': 'Botswana',
5207 'BV': 'Bouvet Island',
5208 'BR': 'Brazil',
5209 'IO': 'British Indian Ocean Territory',
5210 'BN': 'Brunei Darussalam',
5211 'BG': 'Bulgaria',
5212 'BF': 'Burkina Faso',
5213 'BI': 'Burundi',
5214 'KH': 'Cambodia',
5215 'CM': 'Cameroon',
5216 'CA': 'Canada',
5217 'CV': 'Cape Verde',
5218 'KY': 'Cayman Islands',
5219 'CF': 'Central African Republic',
5220 'TD': 'Chad',
5221 'CL': 'Chile',
5222 'CN': 'China',
5223 'CX': 'Christmas Island',
5224 'CC': 'Cocos (Keeling) Islands',
5225 'CO': 'Colombia',
5226 'KM': 'Comoros',
5227 'CG': 'Congo',
5228 'CD': 'Congo, the Democratic Republic of the',
5229 'CK': 'Cook Islands',
5230 'CR': 'Costa Rica',
5231 'CI': 'Côte d\'Ivoire',
5232 'HR': 'Croatia',
5233 'CU': 'Cuba',
5234 'CW': 'Curaçao',
5235 'CY': 'Cyprus',
5236 'CZ': 'Czech Republic',
5237 'DK': 'Denmark',
5238 'DJ': 'Djibouti',
5239 'DM': 'Dominica',
5240 'DO': 'Dominican Republic',
5241 'EC': 'Ecuador',
5242 'EG': 'Egypt',
5243 'SV': 'El Salvador',
5244 'GQ': 'Equatorial Guinea',
5245 'ER': 'Eritrea',
5246 'EE': 'Estonia',
5247 'ET': 'Ethiopia',
5248 'FK': 'Falkland Islands (Malvinas)',
5249 'FO': 'Faroe Islands',
5250 'FJ': 'Fiji',
5251 'FI': 'Finland',
5252 'FR': 'France',
5253 'GF': 'French Guiana',
5254 'PF': 'French Polynesia',
5255 'TF': 'French Southern Territories',
5256 'GA': 'Gabon',
5257 'GM': 'Gambia',
5258 'GE': 'Georgia',
5259 'DE': 'Germany',
5260 'GH': 'Ghana',
5261 'GI': 'Gibraltar',
5262 'GR': 'Greece',
5263 'GL': 'Greenland',
5264 'GD': 'Grenada',
5265 'GP': 'Guadeloupe',
5266 'GU': 'Guam',
5267 'GT': 'Guatemala',
5268 'GG': 'Guernsey',
5269 'GN': 'Guinea',
5270 'GW': 'Guinea-Bissau',
5271 'GY': 'Guyana',
5272 'HT': 'Haiti',
5273 'HM': 'Heard Island and McDonald Islands',
5274 'VA': 'Holy See (Vatican City State)',
5275 'HN': 'Honduras',
5276 'HK': 'Hong Kong',
5277 'HU': 'Hungary',
5278 'IS': 'Iceland',
5279 'IN': 'India',
5280 'ID': 'Indonesia',
5281 'IR': 'Iran, Islamic Republic of',
5282 'IQ': 'Iraq',
5283 'IE': 'Ireland',
5284 'IM': 'Isle of Man',
5285 'IL': 'Israel',
5286 'IT': 'Italy',
5287 'JM': 'Jamaica',
5288 'JP': 'Japan',
5289 'JE': 'Jersey',
5290 'JO': 'Jordan',
5291 'KZ': 'Kazakhstan',
5292 'KE': 'Kenya',
5293 'KI': 'Kiribati',
5294 'KP': 'Korea, Democratic People\'s Republic of',
5295 'KR': 'Korea, Republic of',
5296 'KW': 'Kuwait',
5297 'KG': 'Kyrgyzstan',
5298 'LA': 'Lao People\'s Democratic Republic',
5299 'LV': 'Latvia',
5300 'LB': 'Lebanon',
5301 'LS': 'Lesotho',
5302 'LR': 'Liberia',
5303 'LY': 'Libya',
5304 'LI': 'Liechtenstein',
5305 'LT': 'Lithuania',
5306 'LU': 'Luxembourg',
5307 'MO': 'Macao',
5308 'MK': 'Macedonia, the Former Yugoslav Republic of',
5309 'MG': 'Madagascar',
5310 'MW': 'Malawi',
5311 'MY': 'Malaysia',
5312 'MV': 'Maldives',
5313 'ML': 'Mali',
5314 'MT': 'Malta',
5315 'MH': 'Marshall Islands',
5316 'MQ': 'Martinique',
5317 'MR': 'Mauritania',
5318 'MU': 'Mauritius',
5319 'YT': 'Mayotte',
5320 'MX': 'Mexico',
5321 'FM': 'Micronesia, Federated States of',
5322 'MD': 'Moldova, Republic of',
5323 'MC': 'Monaco',
5324 'MN': 'Mongolia',
5325 'ME': 'Montenegro',
5326 'MS': 'Montserrat',
5327 'MA': 'Morocco',
5328 'MZ': 'Mozambique',
5329 'MM': 'Myanmar',
5330 'NA': 'Namibia',
5331 'NR': 'Nauru',
5332 'NP': 'Nepal',
5333 'NL': 'Netherlands',
5334 'NC': 'New Caledonia',
5335 'NZ': 'New Zealand',
5336 'NI': 'Nicaragua',
5337 'NE': 'Niger',
5338 'NG': 'Nigeria',
5339 'NU': 'Niue',
5340 'NF': 'Norfolk Island',
5341 'MP': 'Northern Mariana Islands',
5342 'NO': 'Norway',
5343 'OM': 'Oman',
5344 'PK': 'Pakistan',
5345 'PW': 'Palau',
5346 'PS': 'Palestine, State of',
5347 'PA': 'Panama',
5348 'PG': 'Papua New Guinea',
5349 'PY': 'Paraguay',
5350 'PE': 'Peru',
5351 'PH': 'Philippines',
5352 'PN': 'Pitcairn',
5353 'PL': 'Poland',
5354 'PT': 'Portugal',
5355 'PR': 'Puerto Rico',
5356 'QA': 'Qatar',
5357 'RE': 'Réunion',
5358 'RO': 'Romania',
5359 'RU': 'Russian Federation',
5360 'RW': 'Rwanda',
5361 'BL': 'Saint Barthélemy',
5362 'SH': 'Saint Helena, Ascension and Tristan da Cunha',
5363 'KN': 'Saint Kitts and Nevis',
5364 'LC': 'Saint Lucia',
5365 'MF': 'Saint Martin (French part)',
5366 'PM': 'Saint Pierre and Miquelon',
5367 'VC': 'Saint Vincent and the Grenadines',
5368 'WS': 'Samoa',
5369 'SM': 'San Marino',
5370 'ST': 'Sao Tome and Principe',
5371 'SA': 'Saudi Arabia',
5372 'SN': 'Senegal',
5373 'RS': 'Serbia',
5374 'SC': 'Seychelles',
5375 'SL': 'Sierra Leone',
5376 'SG': 'Singapore',
5377 'SX': 'Sint Maarten (Dutch part)',
5378 'SK': 'Slovakia',
5379 'SI': 'Slovenia',
5380 'SB': 'Solomon Islands',
5381 'SO': 'Somalia',
5382 'ZA': 'South Africa',
5383 'GS': 'South Georgia and the South Sandwich Islands',
5384 'SS': 'South Sudan',
5385 'ES': 'Spain',
5386 'LK': 'Sri Lanka',
5387 'SD': 'Sudan',
5388 'SR': 'Suriname',
5389 'SJ': 'Svalbard and Jan Mayen',
5390 'SZ': 'Swaziland',
5391 'SE': 'Sweden',
5392 'CH': 'Switzerland',
5393 'SY': 'Syrian Arab Republic',
5394 'TW': 'Taiwan, Province of China',
5395 'TJ': 'Tajikistan',
5396 'TZ': 'Tanzania, United Republic of',
5397 'TH': 'Thailand',
5398 'TL': 'Timor-Leste',
5399 'TG': 'Togo',
5400 'TK': 'Tokelau',
5401 'TO': 'Tonga',
5402 'TT': 'Trinidad and Tobago',
5403 'TN': 'Tunisia',
5404 'TR': 'Turkey',
5405 'TM': 'Turkmenistan',
5406 'TC': 'Turks and Caicos Islands',
5407 'TV': 'Tuvalu',
5408 'UG': 'Uganda',
5409 'UA': 'Ukraine',
5410 'AE': 'United Arab Emirates',
5411 'GB': 'United Kingdom',
5412 'US': 'United States',
5413 'UM': 'United States Minor Outlying Islands',
5414 'UY': 'Uruguay',
5415 'UZ': 'Uzbekistan',
5416 'VU': 'Vanuatu',
5417 'VE': 'Venezuela, Bolivarian Republic of',
5418 'VN': 'Viet Nam',
5419 'VG': 'Virgin Islands, British',
5420 'VI': 'Virgin Islands, U.S.',
5421 'WF': 'Wallis and Futuna',
5422 'EH': 'Western Sahara',
5423 'YE': 'Yemen',
5424 'ZM': 'Zambia',
5425 'ZW': 'Zimbabwe',
5426 }
5427
5428 @classmethod
5429 def short2full(cls, code):
5430 """Convert an ISO 3166-2 country code to the corresponding full name"""
5431 return cls._country_map.get(code.upper())
5432
5433
5434class GeoUtils(object):
5435 # Major IPv4 address blocks per country
5436 _country_ip_map = {
5437 'AD': '46.172.224.0/19',
5438 'AE': '94.200.0.0/13',
5439 'AF': '149.54.0.0/17',
5440 'AG': '209.59.64.0/18',
5441 'AI': '204.14.248.0/21',
5442 'AL': '46.99.0.0/16',
5443 'AM': '46.70.0.0/15',
5444 'AO': '105.168.0.0/13',
5445 'AP': '182.50.184.0/21',
5446 'AQ': '23.154.160.0/24',
5447 'AR': '181.0.0.0/12',
5448 'AS': '202.70.112.0/20',
5449 'AT': '77.116.0.0/14',
5450 'AU': '1.128.0.0/11',
5451 'AW': '181.41.0.0/18',
5452 'AX': '185.217.4.0/22',
5453 'AZ': '5.197.0.0/16',
5454 'BA': '31.176.128.0/17',
5455 'BB': '65.48.128.0/17',
5456 'BD': '114.130.0.0/16',
5457 'BE': '57.0.0.0/8',
5458 'BF': '102.178.0.0/15',
5459 'BG': '95.42.0.0/15',
5460 'BH': '37.131.0.0/17',
5461 'BI': '154.117.192.0/18',
5462 'BJ': '137.255.0.0/16',
5463 'BL': '185.212.72.0/23',
5464 'BM': '196.12.64.0/18',
5465 'BN': '156.31.0.0/16',
5466 'BO': '161.56.0.0/16',
5467 'BQ': '161.0.80.0/20',
5468 'BR': '191.128.0.0/12',
5469 'BS': '24.51.64.0/18',
5470 'BT': '119.2.96.0/19',
5471 'BW': '168.167.0.0/16',
5472 'BY': '178.120.0.0/13',
5473 'BZ': '179.42.192.0/18',
5474 'CA': '99.224.0.0/11',
5475 'CD': '41.243.0.0/16',
5476 'CF': '197.242.176.0/21',
5477 'CG': '160.113.0.0/16',
5478 'CH': '85.0.0.0/13',
5479 'CI': '102.136.0.0/14',
5480 'CK': '202.65.32.0/19',
5481 'CL': '152.172.0.0/14',
5482 'CM': '102.244.0.0/14',
5483 'CN': '36.128.0.0/10',
5484 'CO': '181.240.0.0/12',
5485 'CR': '201.192.0.0/12',
5486 'CU': '152.206.0.0/15',
5487 'CV': '165.90.96.0/19',
5488 'CW': '190.88.128.0/17',
5489 'CY': '31.153.0.0/16',
5490 'CZ': '88.100.0.0/14',
5491 'DE': '53.0.0.0/8',
5492 'DJ': '197.241.0.0/17',
5493 'DK': '87.48.0.0/12',
5494 'DM': '192.243.48.0/20',
5495 'DO': '152.166.0.0/15',
5496 'DZ': '41.96.0.0/12',
5497 'EC': '186.68.0.0/15',
5498 'EE': '90.190.0.0/15',
5499 'EG': '156.160.0.0/11',
5500 'ER': '196.200.96.0/20',
5501 'ES': '88.0.0.0/11',
5502 'ET': '196.188.0.0/14',
5503 'EU': '2.16.0.0/13',
5504 'FI': '91.152.0.0/13',
5505 'FJ': '144.120.0.0/16',
5506 'FK': '80.73.208.0/21',
5507 'FM': '119.252.112.0/20',
5508 'FO': '88.85.32.0/19',
5509 'FR': '90.0.0.0/9',
5510 'GA': '41.158.0.0/15',
5511 'GB': '25.0.0.0/8',
5512 'GD': '74.122.88.0/21',
5513 'GE': '31.146.0.0/16',
5514 'GF': '161.22.64.0/18',
5515 'GG': '62.68.160.0/19',
5516 'GH': '154.160.0.0/12',
5517 'GI': '95.164.0.0/16',
5518 'GL': '88.83.0.0/19',
5519 'GM': '160.182.0.0/15',
5520 'GN': '197.149.192.0/18',
5521 'GP': '104.250.0.0/19',
5522 'GQ': '105.235.224.0/20',
5523 'GR': '94.64.0.0/13',
5524 'GT': '168.234.0.0/16',
5525 'GU': '168.123.0.0/16',
5526 'GW': '197.214.80.0/20',
5527 'GY': '181.41.64.0/18',
5528 'HK': '113.252.0.0/14',
5529 'HN': '181.210.0.0/16',
5530 'HR': '93.136.0.0/13',
5531 'HT': '148.102.128.0/17',
5532 'HU': '84.0.0.0/14',
5533 'ID': '39.192.0.0/10',
5534 'IE': '87.32.0.0/12',
5535 'IL': '79.176.0.0/13',
5536 'IM': '5.62.80.0/20',
5537 'IN': '117.192.0.0/10',
5538 'IO': '203.83.48.0/21',
5539 'IQ': '37.236.0.0/14',
5540 'IR': '2.176.0.0/12',
5541 'IS': '82.221.0.0/16',
5542 'IT': '79.0.0.0/10',
5543 'JE': '87.244.64.0/18',
5544 'JM': '72.27.0.0/17',
5545 'JO': '176.29.0.0/16',
5546 'JP': '133.0.0.0/8',
5547 'KE': '105.48.0.0/12',
5548 'KG': '158.181.128.0/17',
5549 'KH': '36.37.128.0/17',
5550 'KI': '103.25.140.0/22',
5551 'KM': '197.255.224.0/20',
5552 'KN': '198.167.192.0/19',
5553 'KP': '175.45.176.0/22',
5554 'KR': '175.192.0.0/10',
5555 'KW': '37.36.0.0/14',
5556 'KY': '64.96.0.0/15',
5557 'KZ': '2.72.0.0/13',
5558 'LA': '115.84.64.0/18',
5559 'LB': '178.135.0.0/16',
5560 'LC': '24.92.144.0/20',
5561 'LI': '82.117.0.0/19',
5562 'LK': '112.134.0.0/15',
5563 'LR': '102.183.0.0/16',
5564 'LS': '129.232.0.0/17',
5565 'LT': '78.56.0.0/13',
5566 'LU': '188.42.0.0/16',
5567 'LV': '46.109.0.0/16',
5568 'LY': '41.252.0.0/14',
5569 'MA': '105.128.0.0/11',
5570 'MC': '88.209.64.0/18',
5571 'MD': '37.246.0.0/16',
5572 'ME': '178.175.0.0/17',
5573 'MF': '74.112.232.0/21',
5574 'MG': '154.126.0.0/17',
5575 'MH': '117.103.88.0/21',
5576 'MK': '77.28.0.0/15',
5577 'ML': '154.118.128.0/18',
5578 'MM': '37.111.0.0/17',
5579 'MN': '49.0.128.0/17',
5580 'MO': '60.246.0.0/16',
5581 'MP': '202.88.64.0/20',
5582 'MQ': '109.203.224.0/19',
5583 'MR': '41.188.64.0/18',
5584 'MS': '208.90.112.0/22',
5585 'MT': '46.11.0.0/16',
5586 'MU': '105.16.0.0/12',
5587 'MV': '27.114.128.0/18',
5588 'MW': '102.70.0.0/15',
5589 'MX': '187.192.0.0/11',
5590 'MY': '175.136.0.0/13',
5591 'MZ': '197.218.0.0/15',
5592 'NA': '41.182.0.0/16',
5593 'NC': '101.101.0.0/18',
5594 'NE': '197.214.0.0/18',
5595 'NF': '203.17.240.0/22',
5596 'NG': '105.112.0.0/12',
5597 'NI': '186.76.0.0/15',
5598 'NL': '145.96.0.0/11',
5599 'NO': '84.208.0.0/13',
5600 'NP': '36.252.0.0/15',
5601 'NR': '203.98.224.0/19',
5602 'NU': '49.156.48.0/22',
5603 'NZ': '49.224.0.0/14',
5604 'OM': '5.36.0.0/15',
5605 'PA': '186.72.0.0/15',
5606 'PE': '186.160.0.0/14',
5607 'PF': '123.50.64.0/18',
5608 'PG': '124.240.192.0/19',
5609 'PH': '49.144.0.0/13',
5610 'PK': '39.32.0.0/11',
5611 'PL': '83.0.0.0/11',
5612 'PM': '70.36.0.0/20',
5613 'PR': '66.50.0.0/16',
5614 'PS': '188.161.0.0/16',
5615 'PT': '85.240.0.0/13',
5616 'PW': '202.124.224.0/20',
5617 'PY': '181.120.0.0/14',
5618 'QA': '37.210.0.0/15',
5619 'RE': '102.35.0.0/16',
5620 'RO': '79.112.0.0/13',
5621 'RS': '93.86.0.0/15',
5622 'RU': '5.136.0.0/13',
5623 'RW': '41.186.0.0/16',
5624 'SA': '188.48.0.0/13',
5625 'SB': '202.1.160.0/19',
5626 'SC': '154.192.0.0/11',
5627 'SD': '102.120.0.0/13',
5628 'SE': '78.64.0.0/12',
5629 'SG': '8.128.0.0/10',
5630 'SI': '188.196.0.0/14',
5631 'SK': '78.98.0.0/15',
5632 'SL': '102.143.0.0/17',
5633 'SM': '89.186.32.0/19',
5634 'SN': '41.82.0.0/15',
5635 'SO': '154.115.192.0/18',
5636 'SR': '186.179.128.0/17',
5637 'SS': '105.235.208.0/21',
5638 'ST': '197.159.160.0/19',
5639 'SV': '168.243.0.0/16',
5640 'SX': '190.102.0.0/20',
5641 'SY': '5.0.0.0/16',
5642 'SZ': '41.84.224.0/19',
5643 'TC': '65.255.48.0/20',
5644 'TD': '154.68.128.0/19',
5645 'TG': '196.168.0.0/14',
5646 'TH': '171.96.0.0/13',
5647 'TJ': '85.9.128.0/18',
5648 'TK': '27.96.24.0/21',
5649 'TL': '180.189.160.0/20',
5650 'TM': '95.85.96.0/19',
5651 'TN': '197.0.0.0/11',
5652 'TO': '175.176.144.0/21',
5653 'TR': '78.160.0.0/11',
5654 'TT': '186.44.0.0/15',
5655 'TV': '202.2.96.0/19',
5656 'TW': '120.96.0.0/11',
5657 'TZ': '156.156.0.0/14',
5658 'UA': '37.52.0.0/14',
5659 'UG': '102.80.0.0/13',
5660 'US': '6.0.0.0/8',
5661 'UY': '167.56.0.0/13',
5662 'UZ': '84.54.64.0/18',
5663 'VA': '212.77.0.0/19',
5664 'VC': '207.191.240.0/21',
5665 'VE': '186.88.0.0/13',
5666 'VG': '66.81.192.0/20',
5667 'VI': '146.226.0.0/16',
5668 'VN': '14.160.0.0/11',
5669 'VU': '202.80.32.0/20',
5670 'WF': '117.20.32.0/21',
5671 'WS': '202.4.32.0/19',
5672 'YE': '134.35.0.0/16',
5673 'YT': '41.242.116.0/22',
5674 'ZA': '41.0.0.0/11',
5675 'ZM': '102.144.0.0/13',
5676 'ZW': '102.177.192.0/18',
5677 }
5678
5679 @classmethod
5680 def random_ipv4(cls, code_or_block):
5681 if len(code_or_block) == 2:
5682 block = cls._country_ip_map.get(code_or_block.upper())
5683 if not block:
5684 return None
5685 else:
5686 block = code_or_block
5687 addr, preflen = block.split('/')
5688 addr_min = compat_struct_unpack('!L', socket.inet_aton(addr))[0]
5689 addr_max = addr_min | (0xffffffff >> int(preflen))
5690 return compat_str(socket.inet_ntoa(
5691 compat_struct_pack('!L', random.randint(addr_min, addr_max))))
5692
5693
5694class PerRequestProxyHandler(compat_urllib_request.ProxyHandler):
5695 def __init__(self, proxies=None):
5696 # Set default handlers
5697 for type in ('http', 'https'):
5698 setattr(self, '%s_open' % type,
5699 lambda r, proxy='__noproxy__', type=type, meth=self.proxy_open:
5700 meth(r, proxy, type))
5701 compat_urllib_request.ProxyHandler.__init__(self, proxies)
5702
5703 def proxy_open(self, req, proxy, type):
5704 req_proxy = req.headers.get('Ytdl-request-proxy')
5705 if req_proxy is not None:
5706 proxy = req_proxy
5707 del req.headers['Ytdl-request-proxy']
5708
5709 if proxy == '__noproxy__':
5710 return None # No Proxy
5711 if compat_urlparse.urlparse(proxy).scheme.lower() in ('socks', 'socks4', 'socks4a', 'socks5'):
5712 req.add_header('Ytdl-socks-proxy', proxy)
5713 # yt-dlp's http/https handlers do wrapping the socket with socks
5714 return None
5715 return compat_urllib_request.ProxyHandler.proxy_open(
5716 self, req, proxy, type)
5717
5718
5719# Both long_to_bytes and bytes_to_long are adapted from PyCrypto, which is
5720# released into Public Domain
5721# https://github.com/dlitz/pycrypto/blob/master/lib/Crypto/Util/number.py#L387
5722
5723def long_to_bytes(n, blocksize=0):
5724 """long_to_bytes(n:long, blocksize:int) : string
5725 Convert a long integer to a byte string.
5726
5727 If optional blocksize is given and greater than zero, pad the front of the
5728 byte string with binary zeros so that the length is a multiple of
5729 blocksize.
5730 """
5731 # after much testing, this algorithm was deemed to be the fastest
5732 s = b''
5733 n = int(n)
5734 while n > 0:
5735 s = compat_struct_pack('>I', n & 0xffffffff) + s
5736 n = n >> 32
5737 # strip off leading zeros
5738 for i in range(len(s)):
5739 if s[i] != b'\000'[0]:
5740 break
5741 else:
5742 # only happens when n == 0
5743 s = b'\000'
5744 i = 0
5745 s = s[i:]
5746 # add back some pad bytes. this could be done more efficiently w.r.t. the
5747 # de-padding being done above, but sigh...
5748 if blocksize > 0 and len(s) % blocksize:
5749 s = (blocksize - len(s) % blocksize) * b'\000' + s
5750 return s
5751
5752
5753def bytes_to_long(s):
5754 """bytes_to_long(string) : long
5755 Convert a byte string to a long integer.
5756
5757 This is (essentially) the inverse of long_to_bytes().
5758 """
5759 acc = 0
5760 length = len(s)
5761 if length % 4:
5762 extra = (4 - length % 4)
5763 s = b'\000' * extra + s
5764 length = length + extra
5765 for i in range(0, length, 4):
5766 acc = (acc << 32) + compat_struct_unpack('>I', s[i:i + 4])[0]
5767 return acc
5768
5769
5770def ohdave_rsa_encrypt(data, exponent, modulus):
5771 '''
5772 Implement OHDave's RSA algorithm. See http://www.ohdave.com/rsa/
5773
5774 Input:
5775 data: data to encrypt, bytes-like object
5776 exponent, modulus: parameter e and N of RSA algorithm, both integer
5777 Output: hex string of encrypted data
5778
5779 Limitation: supports one block encryption only
5780 '''
5781
5782 payload = int(binascii.hexlify(data[::-1]), 16)
5783 encrypted = pow(payload, exponent, modulus)
5784 return '%x' % encrypted
5785
5786
5787def pkcs1pad(data, length):
5788 """
5789 Padding input data with PKCS#1 scheme
5790
5791 @param {int[]} data input data
5792 @param {int} length target length
5793 @returns {int[]} padded data
5794 """
5795 if len(data) > length - 11:
5796 raise ValueError('Input data too long for PKCS#1 padding')
5797
5798 pseudo_random = [random.randint(0, 254) for _ in range(length - len(data) - 3)]
5799 return [0, 2] + pseudo_random + [0] + data
5800
5801
5802def encode_base_n(num, n, table=None):
5803 FULL_TABLE = '0123456789abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ'
5804 if not table:
5805 table = FULL_TABLE[:n]
5806
5807 if n > len(table):
5808 raise ValueError('base %d exceeds table length %d' % (n, len(table)))
5809
5810 if num == 0:
5811 return table[0]
5812
5813 ret = ''
5814 while num:
5815 ret = table[num % n] + ret
5816 num = num // n
5817 return ret
5818
5819
5820def decode_packed_codes(code):
5821 mobj = re.search(PACKED_CODES_RE, code)
5822 obfuscated_code, base, count, symbols = mobj.groups()
5823 base = int(base)
5824 count = int(count)
5825 symbols = symbols.split('|')
5826 symbol_table = {}
5827
5828 while count:
5829 count -= 1
5830 base_n_count = encode_base_n(count, base)
5831 symbol_table[base_n_count] = symbols[count] or base_n_count
5832
5833 return re.sub(
5834 r'\b(\w+)\b', lambda mobj: symbol_table[mobj.group(0)],
5835 obfuscated_code)
5836
5837
5838def caesar(s, alphabet, shift):
5839 if shift == 0:
5840 return s
5841 l = len(alphabet)
5842 return ''.join(
5843 alphabet[(alphabet.index(c) + shift) % l] if c in alphabet else c
5844 for c in s)
5845
5846
5847def rot47(s):
5848 return caesar(s, r'''!"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~''', 47)
5849
5850
5851def parse_m3u8_attributes(attrib):
5852 info = {}
5853 for (key, val) in re.findall(r'(?P<key>[A-Z0-9-]+)=(?P<val>"[^"]+"|[^",]+)(?:,|$)', attrib):
5854 if val.startswith('"'):
5855 val = val[1:-1]
5856 info[key] = val
5857 return info
5858
5859
5860def urshift(val, n):
5861 return val >> n if val >= 0 else (val + 0x100000000) >> n
5862
5863
5864# Based on png2str() written by @gdkchan and improved by @yokrysty
5865# Originally posted at https://github.com/ytdl-org/youtube-dl/issues/9706
5866def decode_png(png_data):
5867 # Reference: https://www.w3.org/TR/PNG/
5868 header = png_data[8:]
5869
5870 if png_data[:8] != b'\x89PNG\x0d\x0a\x1a\x0a' or header[4:8] != b'IHDR':
5871 raise IOError('Not a valid PNG file.')
5872
5873 int_map = {1: '>B', 2: '>H', 4: '>I'}
5874 unpack_integer = lambda x: compat_struct_unpack(int_map[len(x)], x)[0]
5875
5876 chunks = []
5877
5878 while header:
5879 length = unpack_integer(header[:4])
5880 header = header[4:]
5881
5882 chunk_type = header[:4]
5883 header = header[4:]
5884
5885 chunk_data = header[:length]
5886 header = header[length:]
5887
5888 header = header[4:] # Skip CRC
5889
5890 chunks.append({
5891 'type': chunk_type,
5892 'length': length,
5893 'data': chunk_data
5894 })
5895
5896 ihdr = chunks[0]['data']
5897
5898 width = unpack_integer(ihdr[:4])
5899 height = unpack_integer(ihdr[4:8])
5900
5901 idat = b''
5902
5903 for chunk in chunks:
5904 if chunk['type'] == b'IDAT':
5905 idat += chunk['data']
5906
5907 if not idat:
5908 raise IOError('Unable to read PNG data.')
5909
5910 decompressed_data = bytearray(zlib.decompress(idat))
5911
5912 stride = width * 3
5913 pixels = []
5914
5915 def _get_pixel(idx):
5916 x = idx % stride
5917 y = idx // stride
5918 return pixels[y][x]
5919
5920 for y in range(height):
5921 basePos = y * (1 + stride)
5922 filter_type = decompressed_data[basePos]
5923
5924 current_row = []
5925
5926 pixels.append(current_row)
5927
5928 for x in range(stride):
5929 color = decompressed_data[1 + basePos + x]
5930 basex = y * stride + x
5931 left = 0
5932 up = 0
5933
5934 if x > 2:
5935 left = _get_pixel(basex - 3)
5936 if y > 0:
5937 up = _get_pixel(basex - stride)
5938
5939 if filter_type == 1: # Sub
5940 color = (color + left) & 0xff
5941 elif filter_type == 2: # Up
5942 color = (color + up) & 0xff
5943 elif filter_type == 3: # Average
5944 color = (color + ((left + up) >> 1)) & 0xff
5945 elif filter_type == 4: # Paeth
5946 a = left
5947 b = up
5948 c = 0
5949
5950 if x > 2 and y > 0:
5951 c = _get_pixel(basex - stride - 3)
5952
5953 p = a + b - c
5954
5955 pa = abs(p - a)
5956 pb = abs(p - b)
5957 pc = abs(p - c)
5958
5959 if pa <= pb and pa <= pc:
5960 color = (color + a) & 0xff
5961 elif pb <= pc:
5962 color = (color + b) & 0xff
5963 else:
5964 color = (color + c) & 0xff
5965
5966 current_row.append(color)
5967
5968 return width, height, pixels
5969
5970
5971def write_xattr(path, key, value):
5972 # This mess below finds the best xattr tool for the job
5973 try:
5974 # try the pyxattr module...
5975 import xattr
5976
5977 if hasattr(xattr, 'set'): # pyxattr
5978 # Unicode arguments are not supported in python-pyxattr until
5979 # version 0.5.0
5980 # See https://github.com/ytdl-org/youtube-dl/issues/5498
5981 pyxattr_required_version = '0.5.0'
5982 if version_tuple(xattr.__version__) < version_tuple(pyxattr_required_version):
5983 # TODO: fallback to CLI tools
5984 raise XAttrUnavailableError(
5985 'python-pyxattr is detected but is too old. '
5986 'yt-dlp requires %s or above while your version is %s. '
5987 'Falling back to other xattr implementations' % (
5988 pyxattr_required_version, xattr.__version__))
5989
5990 setxattr = xattr.set
5991 else: # xattr
5992 setxattr = xattr.setxattr
5993
5994 try:
5995 setxattr(path, key, value)
5996 except EnvironmentError as e:
5997 raise XAttrMetadataError(e.errno, e.strerror)
5998
5999 except ImportError:
6000 if compat_os_name == 'nt':
6001 # Write xattrs to NTFS Alternate Data Streams:
6002 # http://en.wikipedia.org/wiki/NTFS#Alternate_data_streams_.28ADS.29
6003 assert ':' not in key
6004 assert os.path.exists(path)
6005
6006 ads_fn = path + ':' + key
6007 try:
6008 with open(ads_fn, 'wb') as f:
6009 f.write(value)
6010 except EnvironmentError as e:
6011 raise XAttrMetadataError(e.errno, e.strerror)
6012 else:
6013 user_has_setfattr = check_executable('setfattr', ['--version'])
6014 user_has_xattr = check_executable('xattr', ['-h'])
6015
6016 if user_has_setfattr or user_has_xattr:
6017
6018 value = value.decode('utf-8')
6019 if user_has_setfattr:
6020 executable = 'setfattr'
6021 opts = ['-n', key, '-v', value]
6022 elif user_has_xattr:
6023 executable = 'xattr'
6024 opts = ['-w', key, value]
6025
6026 cmd = ([encodeFilename(executable, True)]
6027 + [encodeArgument(o) for o in opts]
6028 + [encodeFilename(path, True)])
6029
6030 try:
6031 p = subprocess.Popen(
6032 cmd, stdout=subprocess.PIPE, stderr=subprocess.PIPE, stdin=subprocess.PIPE)
6033 except EnvironmentError as e:
6034 raise XAttrMetadataError(e.errno, e.strerror)
6035 stdout, stderr = process_communicate_or_kill(p)
6036 stderr = stderr.decode('utf-8', 'replace')
6037 if p.returncode != 0:
6038 raise XAttrMetadataError(p.returncode, stderr)
6039
6040 else:
6041 # On Unix, and can't find pyxattr, setfattr, or xattr.
6042 if sys.platform.startswith('linux'):
6043 raise XAttrUnavailableError(
6044 "Couldn't find a tool to set the xattrs. "
6045 "Install either the python 'pyxattr' or 'xattr' "
6046 "modules, or the GNU 'attr' package "
6047 "(which contains the 'setfattr' tool).")
6048 else:
6049 raise XAttrUnavailableError(
6050 "Couldn't find a tool to set the xattrs. "
6051 "Install either the python 'xattr' module, "
6052 "or the 'xattr' binary.")
6053
6054
6055def random_birthday(year_field, month_field, day_field):
6056 start_date = datetime.date(1950, 1, 1)
6057 end_date = datetime.date(1995, 12, 31)
6058 offset = random.randint(0, (end_date - start_date).days)
6059 random_date = start_date + datetime.timedelta(offset)
6060 return {
6061 year_field: str(random_date.year),
6062 month_field: str(random_date.month),
6063 day_field: str(random_date.day),
6064 }
6065
6066
6067# Templates for internet shortcut files, which are plain text files.
6068DOT_URL_LINK_TEMPLATE = '''
6069[InternetShortcut]
6070URL=%(url)s
6071'''.lstrip()
6072
6073DOT_WEBLOC_LINK_TEMPLATE = '''
6074<?xml version="1.0" encoding="UTF-8"?>
6075<!DOCTYPE plist PUBLIC "-//Apple//DTD PLIST 1.0//EN" "http://www.apple.com/DTDs/PropertyList-1.0.dtd">
6076<plist version="1.0">
6077<dict>
6078\t<key>URL</key>
6079\t<string>%(url)s</string>
6080</dict>
6081</plist>
6082'''.lstrip()
6083
6084DOT_DESKTOP_LINK_TEMPLATE = '''
6085[Desktop Entry]
6086Encoding=UTF-8
6087Name=%(filename)s
6088Type=Link
6089URL=%(url)s
6090Icon=text-html
6091'''.lstrip()
6092
6093
6094def iri_to_uri(iri):
6095 """
6096 Converts an IRI (Internationalized Resource Identifier, allowing Unicode characters) to a URI (Uniform Resource Identifier, ASCII-only).
6097
6098 The function doesn't add an additional layer of escaping; e.g., it doesn't escape `%3C` as `%253C`. Instead, it percent-escapes characters with an underlying UTF-8 encoding *besides* those already escaped, leaving the URI intact.
6099 """
6100
6101 iri_parts = compat_urllib_parse_urlparse(iri)
6102
6103 if '[' in iri_parts.netloc:
6104 raise ValueError('IPv6 URIs are not, yet, supported.')
6105 # Querying `.netloc`, when there's only one bracket, also raises a ValueError.
6106
6107 # The `safe` argument values, that the following code uses, contain the characters that should not be percent-encoded. Everything else but letters, digits and '_.-' will be percent-encoded with an underlying UTF-8 encoding. Everything already percent-encoded will be left as is.
6108
6109 net_location = ''
6110 if iri_parts.username:
6111 net_location += compat_urllib_parse_quote(iri_parts.username, safe=r"!$%&'()*+,~")
6112 if iri_parts.password is not None:
6113 net_location += ':' + compat_urllib_parse_quote(iri_parts.password, safe=r"!$%&'()*+,~")
6114 net_location += '@'
6115
6116 net_location += iri_parts.hostname.encode('idna').decode('utf-8') # Punycode for Unicode hostnames.
6117 # The 'idna' encoding produces ASCII text.
6118 if iri_parts.port is not None and iri_parts.port != 80:
6119 net_location += ':' + str(iri_parts.port)
6120
6121 return compat_urllib_parse_urlunparse(
6122 (iri_parts.scheme,
6123 net_location,
6124
6125 compat_urllib_parse_quote_plus(iri_parts.path, safe=r"!$%&'()*+,/:;=@|~"),
6126
6127 # Unsure about the `safe` argument, since this is a legacy way of handling parameters.
6128 compat_urllib_parse_quote_plus(iri_parts.params, safe=r"!$%&'()*+,/:;=@|~"),
6129
6130 # Not totally sure about the `safe` argument, since the source does not explicitly mention the query URI component.
6131 compat_urllib_parse_quote_plus(iri_parts.query, safe=r"!$%&'()*+,/:;=?@{|}~"),
6132
6133 compat_urllib_parse_quote_plus(iri_parts.fragment, safe=r"!#$%&'()*+,/:;=?@{|}~")))
6134
6135 # Source for `safe` arguments: https://url.spec.whatwg.org/#percent-encoded-bytes.
6136
6137
6138def to_high_limit_path(path):
6139 if sys.platform in ['win32', 'cygwin']:
6140 # Work around MAX_PATH limitation on Windows. The maximum allowed length for the individual path segments may still be quite limited.
6141 return r'\\?\ '.rstrip() + os.path.abspath(path)
6142
6143 return path
6144
6145
6146def format_field(obj, field, template='%s', ignore=(None, ''), default='', func=None):
6147 val = obj.get(field, default)
6148 if func and val not in ignore:
6149 val = func(val)
6150 return template % val if val not in ignore else default
6151
6152
6153def clean_podcast_url(url):
6154 return re.sub(r'''(?x)
6155 (?:
6156 (?:
6157 chtbl\.com/track|
6158 media\.blubrry\.com| # https://create.blubrry.com/resources/podcast-media-download-statistics/getting-started/
6159 play\.podtrac\.com
6160 )/[^/]+|
6161 (?:dts|www)\.podtrac\.com/(?:pts/)?redirect\.[0-9a-z]{3,4}| # http://analytics.podtrac.com/how-to-measure
6162 flex\.acast\.com|
6163 pd(?:
6164 cn\.co| # https://podcorn.com/analytics-prefix/
6165 st\.fm # https://podsights.com/docs/
6166 )/e
6167 )/''', '', url)
6168
6169
6170_HEX_TABLE = '0123456789abcdef'
6171
6172
6173def random_uuidv4():
6174 return re.sub(r'[xy]', lambda x: _HEX_TABLE[random.randint(0, 15)], 'xxxxxxxx-xxxx-4xxx-yxxx-xxxxxxxxxxxx')
6175
6176
6177def make_dir(path, to_screen=None):
6178 try:
6179 dn = os.path.dirname(path)
6180 if dn and not os.path.exists(dn):
6181 os.makedirs(dn)
6182 return True
6183 except (OSError, IOError) as err:
6184 if callable(to_screen) is not None:
6185 to_screen('unable to create directory ' + error_to_compat_str(err))
6186 return False
6187
6188
6189def get_executable_path():
6190 from zipimport import zipimporter
6191 if hasattr(sys, 'frozen'): # Running from PyInstaller
6192 path = os.path.dirname(sys.executable)
6193 elif isinstance(globals().get('__loader__'), zipimporter): # Running from ZIP
6194 path = os.path.join(os.path.dirname(__file__), '../..')
6195 else:
6196 path = os.path.join(os.path.dirname(__file__), '..')
6197 return os.path.abspath(path)
6198
6199
6200def load_plugins(name, suffix, namespace):
6201 plugin_info = [None]
6202 classes = []
6203 try:
6204 plugin_info = imp.find_module(
6205 name, [os.path.join(get_executable_path(), 'ytdlp_plugins')])
6206 plugins = imp.load_module(name, *plugin_info)
6207 for name in dir(plugins):
6208 if name in namespace:
6209 continue
6210 if not name.endswith(suffix):
6211 continue
6212 klass = getattr(plugins, name)
6213 classes.append(klass)
6214 namespace[name] = klass
6215 except ImportError:
6216 pass
6217 finally:
6218 if plugin_info[0] is not None:
6219 plugin_info[0].close()
6220 return classes
6221
6222
6223def traverse_obj(obj, keys, *, casesense=True, is_user_input=False, traverse_string=False):
6224 ''' Traverse nested list/dict/tuple
6225 @param casesense Whether to consider dictionary keys as case sensitive
6226 @param is_user_input Whether the keys are generated from user input. If True,
6227 strings are converted to int/slice if necessary
6228 @param traverse_string Whether to traverse inside strings. If True, any
6229 non-compatible object will also be converted into a string
6230 '''
6231 keys = list(keys)[::-1]
6232 while keys:
6233 key = keys.pop()
6234 if isinstance(obj, dict):
6235 assert isinstance(key, compat_str)
6236 if not casesense:
6237 obj = {k.lower(): v for k, v in obj.items()}
6238 key = key.lower()
6239 obj = obj.get(key)
6240 else:
6241 if is_user_input:
6242 key = (int_or_none(key) if ':' not in key
6243 else slice(*map(int_or_none, key.split(':'))))
6244 if not isinstance(obj, (list, tuple)):
6245 if traverse_string:
6246 obj = compat_str(obj)
6247 else:
6248 return None
6249 assert isinstance(key, (int, slice))
6250 obj = try_get(obj, lambda x: x[key])
6251 return obj
6252
6253
6254def traverse_dict(dictn, keys, casesense=True):
6255 ''' For backward compatibility. Do not use '''
6256 return traverse_obj(dictn, keys, casesense=casesense,
6257 is_user_input=True, traverse_string=True)