aboutsummaryrefslogtreecommitdiffstats
path: root/youtube_dl/extractor/common.py
blob: 5b6a09c0b7f732ef5e9361775e42719af0deb8b7 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
1317
1318
1319
1320
1321
1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
1345
1346
1347
1348
1349
1350
1351
1352
1353
1354
1355
1356
1357
1358
1359
1360
1361
1362
1363
1364
1365
1366
1367
1368
1369
1370
1371
1372
1373
1374
1375
1376
1377
1378
1379
1380
1381
1382
1383
1384
1385
1386
1387
1388
1389
1390
1391
1392
1393
1394
1395
1396
1397
1398
1399
1400
1401
1402
1403
1404
1405
1406
1407
1408
1409
1410
1411
1412
1413
1414
1415
1416
1417
1418
1419
1420
1421
1422
1423
1424
1425
1426
1427
1428
1429
1430
1431
1432
1433
1434
1435
1436
1437
1438
1439
1440
1441
1442
1443
1444
1445
1446
1447
1448
1449
1450
1451
1452
1453
1454
1455
1456
1457
1458
1459
1460
1461
1462
1463
1464
1465
1466
1467
1468
1469
1470
1471
1472
1473
1474
1475
1476
1477
1478
1479
1480
1481
1482
1483
1484
1485
1486
1487
1488
1489
1490
1491
1492
1493
1494
1495
1496
1497
1498
1499
1500
1501
1502
1503
1504
1505
1506
1507
1508
1509
1510
1511
1512
1513
1514
1515
1516
1517
1518
1519
1520
1521
1522
1523
1524
1525
1526
1527
1528
1529
1530
1531
1532
1533
1534
1535
1536
1537
1538
1539
1540
1541
1542
1543
1544
1545
1546
1547
1548
1549
1550
1551
1552
1553
1554
1555
1556
1557
1558
1559
1560
1561
1562
1563
1564
1565
1566
1567
1568
1569
1570
1571
1572
1573
1574
1575
1576
1577
1578
1579
1580
1581
1582
1583
1584
1585
1586
1587
1588
1589
1590
1591
1592
1593
1594
1595
1596
1597
1598
1599
1600
1601
1602
1603
1604
1605
1606
1607
1608
1609
1610
1611
1612
1613
1614
1615
1616
1617
1618
1619
1620
1621
1622
1623
1624
1625
1626
1627
1628
1629
1630
1631
1632
1633
1634
1635
1636
1637
1638
1639
1640
1641
1642
1643
1644
1645
1646
1647
1648
1649
1650
1651
1652
1653
1654
1655
1656
1657
1658
1659
1660
1661
1662
1663
1664
1665
1666
1667
1668
1669
1670
1671
1672
1673
1674
1675
1676
1677
1678
1679
1680
1681
1682
1683
1684
1685
1686
1687
1688
1689
1690
1691
1692
1693
1694
1695
1696
1697
1698
1699
1700
1701
1702
1703
1704
1705
1706
1707
1708
1709
1710
1711
1712
1713
1714
1715
1716
1717
1718
1719
1720
1721
1722
1723
1724
1725
1726
1727
1728
1729
1730
1731
1732
1733
1734
1735
1736
1737
1738
1739
1740
1741
1742
1743
1744
1745
1746
1747
1748
1749
1750
1751
1752
1753
1754
1755
1756
1757
1758
1759
1760
1761
1762
1763
1764
1765
1766
1767
1768
1769
1770
1771
1772
1773
1774
1775
1776
1777
1778
1779
1780
1781
1782
1783
1784
1785
1786
1787
1788
1789
1790
1791
1792
1793
1794
1795
1796
1797
1798
1799
1800
1801
1802
1803
1804
1805
1806
1807
1808
1809
1810
1811
1812
1813
1814
1815
1816
1817
1818
1819
1820
1821
1822
1823
1824
1825
1826
1827
1828
1829
1830
1831
1832
1833
1834
1835
1836
1837
1838
1839
1840
1841
1842
1843
1844
1845
1846
1847
1848
1849
1850
1851
1852
1853
1854
1855
1856
1857
1858
1859
1860
1861
1862
1863
1864
1865
1866
1867
1868
1869
1870
1871
1872
1873
1874
1875
1876
1877
1878
1879
1880
1881
1882
1883
1884
1885
1886
1887
1888
1889
1890
1891
1892
1893
1894
1895
1896
1897
1898
1899
1900
1901
1902
1903
1904
1905
1906
1907
1908
1909
1910
1911
1912
1913
1914
1915
1916
1917
1918
1919
1920
1921
1922
1923
1924
1925
1926
1927
1928
1929
1930
1931
1932
1933
1934
1935
1936
1937
1938
1939
1940
1941
1942
1943
1944
1945
1946
1947
1948
1949
1950
1951
1952
1953
1954
1955
1956
1957
1958
1959
1960
1961
1962
1963
1964
1965
1966
1967
1968
1969
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
2031
2032
2033
2034
2035
2036
2037
2038
2039
2040
2041
2042
2043
2044
2045
2046
2047
2048
2049
2050
2051
2052
2053
2054
2055
2056
2057
2058
2059
2060
2061
2062
2063
2064
2065
2066
2067
2068
2069
2070
2071
2072
2073
2074
2075
2076
2077
2078
2079
2080
2081
2082
2083
2084
2085
2086
2087
2088
2089
2090
2091
2092
2093
2094
2095
2096
2097
2098
2099
2100
2101
2102
2103
2104
2105
2106
2107
2108
2109
2110
2111
2112
2113
2114
2115
2116
2117
2118
2119
2120
2121
2122
2123
2124
2125
2126
2127
2128
2129
2130
2131
2132
2133
2134
2135
2136
2137
2138
2139
2140
2141
2142
2143
2144
2145
2146
2147
2148
2149
2150
2151
2152
2153
2154
2155
2156
2157
2158
2159
2160
2161
2162
2163
2164
2165
2166
2167
2168
2169
2170
2171
2172
2173
2174
2175
2176
2177
2178
2179
2180
2181
2182
2183
2184
2185
2186
2187
2188
2189
2190
2191
2192
2193
2194
2195
2196
2197
2198
2199
2200
2201
2202
2203
2204
2205
2206
2207
2208
2209
2210
2211
2212
2213
2214
2215
2216
2217
2218
2219
2220
2221
2222
2223
2224
2225
2226
2227
2228
2229
2230
2231
2232
2233
2234
2235
2236
2237
2238
2239
2240
2241
2242
2243
2244
2245
2246
2247
2248
2249
2250
2251
2252
2253
2254
2255
2256
2257
2258
2259
2260
2261
2262
2263
2264
2265
2266
2267
2268
2269
2270
2271
2272
2273
2274
2275
2276
2277
2278
2279
2280
2281
2282
2283
2284
2285
2286
2287
2288
2289
2290
2291
2292
2293
2294
2295
2296
2297
2298
2299
2300
2301
2302
2303
2304
2305
2306
2307
2308
2309
2310
2311
2312
2313
2314
2315
2316
2317
2318
2319
2320
2321
2322
2323
2324
2325
2326
2327
2328
2329
2330
2331
2332
2333
2334
2335
2336
2337
2338
2339
2340
2341
2342
2343
2344
2345
2346
2347
2348
2349
2350
2351
2352
2353
2354
2355
2356
2357
2358
2359
2360
2361
2362
2363
2364
2365
2366
2367
2368
2369
2370
2371
2372
2373
2374
2375
2376
2377
2378
2379
2380
2381
2382
2383
2384
2385
2386
2387
2388
2389
2390
2391
2392
2393
2394
2395
2396
2397
2398
2399
2400
2401
2402
2403
2404
2405
2406
2407
2408
2409
2410
2411
2412
2413
2414
2415
2416
2417
2418
2419
2420
2421
2422
2423
2424
2425
2426
2427
2428
2429
2430
2431
2432
2433
2434
2435
2436
2437
2438
2439
2440
2441
2442
2443
2444
2445
2446
2447
2448
2449
2450
2451
2452
2453
2454
2455
2456
2457
2458
2459
2460
2461
2462
2463
2464
2465
2466
2467
2468
2469
2470
2471
2472
2473
2474
2475
2476
2477
2478
2479
2480
2481
2482
2483
2484
2485
2486
2487
2488
2489
2490
2491
2492
2493
2494
2495
2496
2497
2498
2499
2500
2501
2502
2503
2504
2505
2506
2507
2508
2509
2510
2511
2512
2513
2514
2515
2516
2517
2518
2519
2520
2521
2522
2523
2524
2525
2526
2527
2528
2529
2530
2531
2532
2533
2534
2535
2536
2537
2538
2539
2540
2541
2542
2543
2544
2545
2546
2547
2548
2549
2550
2551
2552
2553
2554
2555
2556
2557
2558
2559
2560
2561
2562
2563
2564
2565
2566
2567
2568
2569
2570
2571
2572
2573
2574
2575
2576
2577
2578
2579
2580
2581
2582
2583
2584
2585
2586
2587
2588
2589
2590
2591
2592
2593
2594
2595
2596
2597
2598
2599
2600
2601
2602
2603
2604
2605
2606
2607
2608
2609
2610
2611
2612
2613
2614
2615
2616
2617
2618
2619
2620
2621
2622
2623
# coding: utf-8
from __future__ import unicode_literals

import base64
import datetime
import hashlib
import json
import netrc
import os
import random
import re
import socket
import sys
import time
import math

from ..compat import (
    compat_cookiejar,
    compat_cookies,
    compat_etree_fromstring,
    compat_getpass,
    compat_http_client,
    compat_os_name,
    compat_str,
    compat_urllib_error,
    compat_urllib_parse_unquote,
    compat_urllib_parse_urlencode,
    compat_urllib_request,
    compat_urlparse,
    compat_xml_parse_error,
)
from ..downloader.f4m import (
    get_base_url,
    remove_encrypted_media,
)
from ..utils import (
    NO_DEFAULT,
    age_restricted,
    base_url,
    bug_reports_message,
    clean_html,
    compiled_regex_type,
    determine_ext,
    determine_protocol,
    error_to_compat_str,
    ExtractorError,
    extract_attributes,
    fix_xml_ampersands,
    float_or_none,
    GeoRestrictedError,
    GeoUtils,
    int_or_none,
    js_to_json,
    mimetype2ext,
    orderedSet,
    parse_codecs,
    parse_duration,
    parse_iso8601,
    parse_m3u8_attributes,
    RegexNotFoundError,
    sanitized_Request,
    sanitize_filename,
    unescapeHTML,
    unified_strdate,
    unified_timestamp,
    update_Request,
    update_url_query,
    urljoin,
    url_basename,
    xpath_element,
    xpath_text,
    xpath_with_ns,
)


class InfoExtractor(object):
    """Information Extractor class.

    Information extractors are the classes that, given a URL, extract
    information about the video (or videos) the URL refers to. This
    information includes the real video URL, the video title, author and
    others. The information is stored in a dictionary which is then
    passed to the YoutubeDL. The YoutubeDL processes this
    information possibly downloading the video to the file system, among
    other possible outcomes.

    The type field determines the type of the result.
    By far the most common value (and the default if _type is missing) is
    "video", which indicates a single video.

    For a video, the dictionaries must include the following fields:

    id:             Video identifier.
    title:          Video title, unescaped.

    Additionally, it must contain either a formats entry or a url one:

    formats:        A list of dictionaries for each format available, ordered
                    from worst to best quality.

                    Potential fields:
                    * url        Mandatory. The URL of the video file
                    * manifest_url
                                 The URL of the manifest file in case of
                                 fragmented media (DASH, hls, hds)
                    * ext        Will be calculated from URL if missing
                    * format     A human-readable description of the format
                                 ("mp4 container with h264/opus").
                                 Calculated from the format_id, width, height.
                                 and format_note fields if missing.
                    * format_id  A short description of the format
                                 ("mp4_h264_opus" or "19").
                                Technically optional, but strongly recommended.
                    * format_note Additional info about the format
                                 ("3D" or "DASH video")
                    * width      Width of the video, if known
                    * height     Height of the video, if known
                    * resolution Textual description of width and height
                    * tbr        Average bitrate of audio and video in KBit/s
                    * abr        Average audio bitrate in KBit/s
                    * acodec     Name of the audio codec in use
                    * asr        Audio sampling rate in Hertz
                    * vbr        Average video bitrate in KBit/s
                    * fps        Frame rate
                    * vcodec     Name of the video codec in use
                    * container  Name of the container format
                    * filesize   The number of bytes, if known in advance
                    * filesize_approx  An estimate for the number of bytes
                    * player_url SWF Player URL (used for rtmpdump).
                    * protocol   The protocol that will be used for the actual
                                 download, lower-case.
                                 "http", "https", "rtsp", "rtmp", "rtmpe",
                                 "m3u8", "m3u8_native" or "http_dash_segments".
                    * fragment_base_url
                                 Base URL for fragments. Each fragment's path
                                 value (if present) will be relative to
                                 this URL.
                    * fragments  A list of fragments of a fragmented media.
                                 Each fragment entry must contain either an url
                                 or a path. If an url is present it should be
                                 considered by a client. Otherwise both path and
                                 fragment_base_url must be present. Here is
                                 the list of all potential fields:
                                 * "url" - fragment's URL
                                 * "path" - fragment's path relative to
                                            fragment_base_url
                                 * "duration" (optional, int or float)
                                 * "filesize" (optional, int)
                    * preference Order number of this format. If this field is
                                 present and not None, the formats get sorted
                                 by this field, regardless of all other values.
                                 -1 for default (order by other properties),
                                 -2 or smaller for less than default.
                                 < -1000 to hide the format (if there is
                                    another one which is strictly better)
                    * language   Language code, e.g. "de" or "en-US".
                    * language_preference  Is this in the language mentioned in
                                 the URL?
                                 10 if it's what the URL is about,
                                 -1 for default (don't know),
                                 -10 otherwise, other values reserved for now.
                    * quality    Order number of the video quality of this
                                 format, irrespective of the file format.
                                 -1 for default (order by other properties),
                                 -2 or smaller for less than default.
                    * source_preference  Order number for this video source
                                  (quality takes higher priority)
                                 -1 for default (order by other properties),
                                 -2 or smaller for less than default.
                    * http_headers  A dictionary of additional HTTP headers
                                 to add to the request.
                    * stretched_ratio  If given and not 1, indicates that the
                                 video's pixels are not square.
                                 width : height ratio as float.
                    * no_resume  The server does not support resuming the
                                 (HTTP or RTMP) download. Boolean.

    url:            Final video URL.
    ext:            Video filename extension.
    format:         The video format, defaults to ext (used for --get-format)
    player_url:     SWF Player URL (used for rtmpdump).

    The following fields are optional:

    alt_title:      A secondary title of the video.
    display_id      An alternative identifier for the video, not necessarily
                    unique, but available before title. Typically, id is
                    something like "4234987", title "Dancing naked mole rats",
                    and display_id "dancing-naked-mole-rats"
    thumbnails:     A list of dictionaries, with the following entries:
                        * "id" (optional, string) - Thumbnail format ID
                        * "url"
                        * "preference" (optional, int) - quality of the image
                        * "width" (optional, int)
                        * "height" (optional, int)
                        * "resolution" (optional, string "{width}x{height"},
                                        deprecated)
                        * "filesize" (optional, int)
    thumbnail:      Full URL to a video thumbnail image.
    description:    Full video description.
    uploader:       Full name of the video uploader.
    license:        License name the video is licensed under.
    creator:        The creator of the video.
    release_date:   The date (YYYYMMDD) when the video was released.
    timestamp:      UNIX timestamp of the moment the video became available.
    upload_date:    Video upload date (YYYYMMDD).
                    If not explicitly set, calculated from timestamp.
    uploader_id:    Nickname or id of the video uploader.
    uploader_url:   Full URL to a personal webpage of the video uploader.
    location:       Physical location where the video was filmed.
    subtitles:      The available subtitles as a dictionary in the format
                    {tag: subformats}. "tag" is usually a language code, and
                    "subformats" is a list sorted from lower to higher
                    preference, each element is a dictionary with the "ext"
                    entry and one of:
                        * "data": The subtitles file contents
                        * "url": A URL pointing to the subtitles file
                    "ext" will be calculated from URL if missing
    automatic_captions: Like 'subtitles', used by the YoutubeIE for
                    automatically generated captions
    duration:       Length of the video in seconds, as an integer or float.
    view_count:     How many users have watched the video on the platform.
    like_count:     Number of positive ratings of the video
    dislike_count:  Number of negative ratings of the video
    repost_count:   Number of reposts of the video
    average_rating: Average rating give by users, the scale used depends on the webpage
    comment_count:  Number of comments on the video
    comments:       A list of comments, each with one or more of the following
                    properties (all but one of text or html optional):
                        * "author" - human-readable name of the comment author
                        * "author_id" - user ID of the comment author
                        * "id" - Comment ID
                        * "html" - Comment as HTML
                        * "text" - Plain text of the comment
                        * "timestamp" - UNIX timestamp of comment
                        * "parent" - ID of the comment this one is replying to.
                                     Set to "root" to indicate that this is a
                                     comment to the original video.
    age_limit:      Age restriction for the video, as an integer (years)
    webpage_url:    The URL to the video webpage, if given to youtube-dl it
                    should allow to get the same result again. (It will be set
                    by YoutubeDL if it's missing)
    categories:     A list of categories that the video falls in, for example
                    ["Sports", "Berlin"]
    tags:           A list of tags assigned to the video, e.g. ["sweden", "pop music"]
    is_live:        True, False, or None (=unknown). Whether this video is a
                    live stream that goes on instead of a fixed-length video.
    start_time:     Time in seconds where the reproduction should start, as
                    specified in the URL.
    end_time:       Time in seconds where the reproduction should end, as
                    specified in the URL.
    chapters:       A list of dictionaries, with the following entries:
                        * "start_time" - The start time of the chapter in seconds
                        * "end_time" - The end time of the chapter in seconds
                        * "title" (optional, string)

    The following fields should only be used when the video belongs to some logical
    chapter or section:

    chapter:        Name or title of the chapter the video belongs to.
    chapter_number: Number of the chapter the video belongs to, as an integer.
    chapter_id:     Id of the chapter the video belongs to, as a unicode string.

    The following fields should only be used when the video is an episode of some
    series, programme or podcast:

    series:         Title of the series or programme the video episode belongs to.
    season:         Title of the season the video episode belongs to.
    season_number:  Number of the season the video episode belongs to, as an integer.
    season_id:      Id of the season the video episode belongs to, as a unicode string.
    episode:        Title of the video episode. Unlike mandatory video title field,
                    this field should denote the exact title of the video episode
                    without any kind of decoration.
    episode_number: Number of the video episode within a season, as an integer.
    episode_id:     Id of the video episode, as a unicode string.

    The following fields should only be used when the media is a track or a part of
    a music album:

    track:          Title of the track.
    track_number:   Number of the track within an album or a disc, as an integer.
    track_id:       Id of the track (useful in case of custom indexing, e.g. 6.iii),
                    as a unicode string.
    artist:         Artist(s) of the track.
    genre:          Genre(s) of the track.
    album:          Title of the album the track belongs to.
    album_type:     Type of the album (e.g. "Demo", "Full-length", "Split", "Compilation", etc).
    album_artist:   List of all artists appeared on the album (e.g.
                    "Ash Borer / Fell Voices" or "Various Artists", useful for splits
                    and compilations).
    disc_number:    Number of the disc or other physical medium the track belongs to,
                    as an integer.
    release_year:   Year (YYYY) when the album was released.

    Unless mentioned otherwise, the fields should be Unicode strings.

    Unless mentioned otherwise, None is equivalent to absence of information.


    _type "playlist" indicates multiple videos.
    There must be a key "entries", which is a list, an iterable, or a PagedList
    object, each element of which is a valid dictionary by this specification.

    Additionally, playlists can have "id", "title", "description", "uploader",
    "uploader_id", "uploader_url" attributes with the same semantics as videos
    (see above).


    _type "multi_video" indicates that there are multiple videos that
    form a single show, for examples multiple acts of an opera or TV episode.
    It must have an entries key like a playlist and contain all the keys
    required for a video at the same time.


    _type "url" indicates that the video must be extracted from another
    location, possibly by a different extractor. Its only required key is:
    "url" - the next URL to extract.
    The key "ie_key" can be set to the class name (minus the trailing "IE",
    e.g. "Youtube") if the extractor class is known in advance.
    Additionally, the dictionary may have any properties of the resolved entity
    known in advance, for example "title" if the title of the referred video is
    known ahead of time.


    _type "url_transparent" entities have the same specification as "url", but
    indicate that the given additional information is more precise than the one
    associated with the resolved URL.
    This is useful when a site employs a video service that hosts the video and
    its technical metadata, but that video service does not embed a useful
    title, description etc.


    Subclasses of this one should re-define the _real_initialize() and
    _real_extract() methods and define a _VALID_URL regexp.
    Probably, they should also be added to the list of extractors.

    _GEO_BYPASS attribute may be set to False in order to disable
    geo restriction bypass mechanisms for a particular extractor.
    Though it won't disable explicit geo restriction bypass based on
    country code provided with geo_bypass_country. (experimental)

    _GEO_COUNTRIES attribute may contain a list of presumably geo unrestricted
    countries for this extractor. One of these countries will be used by
    geo restriction bypass mechanism right away in order to bypass
    geo restriction, of course, if the mechanism is not disabled. (experimental)

    NB: both these geo attributes are experimental and may change in future
    or be completely removed.

    Finally, the _WORKING attribute should be set to False for broken IEs
    in order to warn the users and skip the tests.
    """

    _ready = False
    _downloader = None
    _x_forwarded_for_ip = None
    _GEO_BYPASS = True
    _GEO_COUNTRIES = None
    _WORKING = True

    def __init__(self, downloader=None):
        """Constructor. Receives an optional downloader."""
        self._ready = False
        self._x_forwarded_for_ip = None
        self.set_downloader(downloader)

    @classmethod
    def suitable(cls, url):
        """Receives a URL and returns True if suitable for this IE."""

        # This does not use has/getattr intentionally - we want to know whether
        # we have cached the regexp for *this* class, whereas getattr would also
        # match the superclass
        if '_VALID_URL_RE' not in cls.__dict__:
            cls._VALID_URL_RE = re.compile(cls._VALID_URL)
        return cls._VALID_URL_RE.match(url) is not None

    @classmethod
    def _match_id(cls, url):
        if '_VALID_URL_RE' not in cls.__dict__:
            cls._VALID_URL_RE = re.compile(cls._VALID_URL)
        m = cls._VALID_URL_RE.match(url)
        assert m
        return compat_str(m.group('id'))

    @classmethod
    def working(cls):
        """Getter method for _WORKING."""
        return cls._WORKING

    def initialize(self):
        """Initializes an instance (authentication, etc)."""
        self._initialize_geo_bypass(self._GEO_COUNTRIES)
        if not self._ready:
            self._real_initialize()
            self._ready = True

    def _initialize_geo_bypass(self, countries):
        """
        Initialize geo restriction bypass mechanism.

        This method is used to initialize geo bypass mechanism based on faking
        X-Forwarded-For HTTP header. A random country from provided country list
        is selected and a random IP belonging to this country is generated. This
        IP will be passed as X-Forwarded-For HTTP header in all subsequent
        HTTP requests.

        This method will be used for initial geo bypass mechanism initialization
        during the instance initialization with _GEO_COUNTRIES.

        You may also manually call it from extractor's code if geo countries
        information is not available beforehand (e.g. obtained during
        extraction) or due to some another reason.
        """
        if not self._x_forwarded_for_ip:
            country_code = self._downloader.params.get('geo_bypass_country', None)
            # If there is no explicit country for geo bypass specified and
            # the extractor is known to be geo restricted let's fake IP
            # as X-Forwarded-For right away.
            if (not country_code and
                    self._GEO_BYPASS and
                    self._downloader.params.get('geo_bypass', True) and
                    countries):
                country_code = random.choice(countries)
            if country_code:
                self._x_forwarded_for_ip = GeoUtils.random_ipv4(country_code)
                if self._downloader.params.get('verbose', False):
                    self._downloader.to_screen(
                        '[debug] Using fake IP %s (%s) as X-Forwarded-For.'
                        % (self._x_forwarded_for_ip, country_code.upper()))

    def extract(self, url):
        """Extracts URL information and returns it in list of dicts."""
        try:
            for _ in range(2):
                try:
                    self.initialize()
                    ie_result = self._real_extract(url)
                    if self._x_forwarded_for_ip:
                        ie_result['__x_forwarded_for_ip'] = self._x_forwarded_for_ip
                    return ie_result
                except GeoRestrictedError as e:
                    if self.__maybe_fake_ip_and_retry(e.countries):
                        continue
                    raise
        except ExtractorError:
            raise
        except compat_http_client.IncompleteRead as e:
            raise ExtractorError('A network error has occurred.', cause=e, expected=True)
        except (KeyError, StopIteration) as e:
            raise ExtractorError('An extractor error has occurred.', cause=e)

    def __maybe_fake_ip_and_retry(self, countries):
        if (not self._downloader.params.get('geo_bypass_country', None) and
                self._GEO_BYPASS and
                self._downloader.params.get('geo_bypass', True) and
                not self._x_forwarded_for_ip and
                countries):
            country_code = random.choice(countries)
            self._x_forwarded_for_ip = GeoUtils.random_ipv4(country_code)
            if self._x_forwarded_for_ip:
                self.report_warning(
                    'Video is geo restricted. Retrying extraction with fake IP %s (%s) as X-Forwarded-For.'
                    % (self._x_forwarded_for_ip, country_code.upper()))
                return True
        return False

    def set_downloader(self, downloader):
        """Sets the downloader for this IE."""
        self._downloader = downloader

    def _real_initialize(self):
        """Real initialization process. Redefine in subclasses."""
        pass

    def _real_extract(self, url):
        """Real extraction process. Redefine in subclasses."""
        pass

    @classmethod
    def ie_key(cls):
        """A string for getting the InfoExtractor with get_info_extractor"""
        return compat_str(cls.__name__[:-2])

    @property
    def IE_NAME(self):
        return compat_str(type(self).__name__[:-2])

    def _request_webpage(self, url_or_request, video_id, note=None, errnote=None, fatal=True, data=None, headers={}, query={}):
        """ Returns the response handle """
        if note is None:
            self.report_download_webpage(video_id)
        elif note is not False:
            if video_id is None:
                self.to_screen('%s' % (note,))
            else:
                self.to_screen('%s: %s' % (video_id, note))

        # Some sites check X-Forwarded-For HTTP header in order to figure out
        # the origin of the client behind proxy. This allows bypassing geo
        # restriction by faking this header's value to IP that belongs to some
        # geo unrestricted country. We will do so once we encounter any
        # geo restriction error.
        if self._x_forwarded_for_ip:
            if 'X-Forwarded-For' not in headers:
                headers['X-Forwarded-For'] = self._x_forwarded_for_ip

        if isinstance(url_or_request, compat_urllib_request.Request):
            url_or_request = update_Request(
                url_or_request, data=data, headers=headers, query=query)
        else:
            if query:
                url_or_request = update_url_query(url_or_request, query)
            if data is not None or headers:
                url_or_request = sanitized_Request(url_or_request, data, headers)
        try:
            return self._downloader.urlopen(url_or_request)
        except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
            if errnote is False:
                return False
            if errnote is None:
                errnote = 'Unable to download webpage'

            errmsg = '%s: %s' % (errnote, error_to_compat_str(err))
            if fatal:
                raise ExtractorError(errmsg, sys.exc_info()[2], cause=err)
            else:
                self._downloader.report_warning(errmsg)
                return False

    def _download_webpage_handle(self, url_or_request, video_id, note=None, errnote=None, fatal=True, encoding=None, data=None, headers={}, query={}):
        """ Returns a tuple (page content as string, URL handle) """
        # Strip hashes from the URL (#1038)
        if isinstance(url_or_request, (compat_str, str)):
            url_or_request = url_or_request.partition('#')[0]

        urlh = self._request_webpage(url_or_request, video_id, note, errnote, fatal, data=data, headers=headers, query=query)
        if urlh is False:
            assert not fatal
            return False
        content = self._webpage_read_content(urlh, url_or_request, video_id, note, errnote, fatal, encoding=encoding)
        return (content, urlh)

    @staticmethod
    def _guess_encoding_from_content(content_type, webpage_bytes):
        m = re.match(r'[a-zA-Z0-9_.-]+/[a-zA-Z0-9_.-]+\s*;\s*charset=(.+)', content_type)
        if m:
            encoding = m.group(1)
        else:
            m = re.search(br'<meta[^>]+charset=[\'"]?([^\'")]+)[ /\'">]',
                          webpage_bytes[:1024])
            if m:
                encoding = m.group(1).decode('ascii')
            elif webpage_bytes.startswith(b'\xff\xfe'):
                encoding = 'utf-16'
            else:
                encoding = 'utf-8'

        return encoding

    def __check_blocked(self, content):
        first_block = content[:512]
        if ('<title>Access to this site is blocked</title>' in content and
                'Websense' in first_block):
            msg = 'Access to this webpage has been blocked by Websense filtering software in your network.'
            blocked_iframe = self._html_search_regex(
                r'<iframe src="([^"]+)"', content,
                'Websense information URL', default=None)
            if blocked_iframe:
                msg += ' Visit %s for more details' % blocked_iframe
            raise ExtractorError(msg, expected=True)
        if '<title>The URL you requested has been blocked</title>' in first_block:
            msg = (
                'Access to this webpage has been blocked by Indian censorship. '
                'Use a VPN or proxy server (with --proxy) to route around it.')
            block_msg = self._html_search_regex(
                r'</h1><p>(.*?)</p>',
                content, 'block message', default=None)
            if block_msg:
                msg += ' (Message: "%s")' % block_msg.replace('\n', ' ')
            raise ExtractorError(msg, expected=True)
        if ('<title>TTK :: Доступ к ресурсу ограничен</title>' in content and
                'blocklist.rkn.gov.ru' in content):
            raise ExtractorError(
                'Access to this webpage has been blocked by decision of the Russian government. '
                'Visit http://blocklist.rkn.gov.ru/ for a block reason.',
                expected=True)

    def _webpage_read_content(self, urlh, url_or_request, video_id, note=None, errnote=None, fatal=True, prefix=None, encoding=None):
        content_type = urlh.headers.get('Content-Type', '')
        webpage_bytes = urlh.read()
        if prefix is not None:
            webpage_bytes = prefix + webpage_bytes
        if not encoding:
            encoding = self._guess_encoding_from_content(content_type, webpage_bytes)
        if self._downloader.params.get('dump_intermediate_pages', False):
            self.to_screen('Dumping request to ' + urlh.geturl())
            dump = base64.b64encode(webpage_bytes).decode('ascii')
            self._downloader.to_screen(dump)
        if self._downloader.params.get('write_pages', False):
            basen = '%s_%s' % (video_id, urlh.geturl())
            if len(basen) > 240:
                h = '___' + hashlib.md5(basen.encode('utf-8')).hexdigest()
                basen = basen[:240 - len(h)] + h
            raw_filename = basen + '.dump'
            filename = sanitize_filename(raw_filename, restricted=True)
            self.to_screen('Saving request to ' + filename)
            # Working around MAX_PATH limitation on Windows (see
            # http://msdn.microsoft.com/en-us/library/windows/desktop/aa365247(v=vs.85).aspx)
            if compat_os_name == 'nt':
                absfilepath = os.path.abspath(filename)
                if len(absfilepath) > 259:
                    filename = '\\\\?\\' + absfilepath
            with open(filename, 'wb') as outf:
                outf.write(webpage_bytes)

        try:
            content = webpage_bytes.decode(encoding, 'replace')
        except LookupError:
            content = webpage_bytes.decode('utf-8', 'replace')

        self.__check_blocked(content)

        return content

    def _download_webpage(self, url_or_request, video_id, note=None, errnote=None, fatal=True, tries=1, timeout=5, encoding=None, data=None, headers={}, query={}):
        """ Returns the data of the page as a string """
        success = False
        try_count = 0
        while success is False:
            try:
                res = self._download_webpage_handle(url_or_request, video_id, note, errnote, fatal, encoding=encoding, data=data, headers=headers, query=query)
                success = True
            except compat_http_client.IncompleteRead as e:
                try_count += 1
                if try_count >= tries:
                    raise e
                self._sleep(timeout, video_id)
        if res is False:
            return res
        else:
            content, _ = res
            return content

    def _download_xml(self, url_or_request, video_id,
                      note='Downloading XML', errnote='Unable to download XML',
                      transform_source=None, fatal=True, encoding=None,
                      data=None, headers={}, query={}):
        """Return the xml as an xml.etree.ElementTree.Element"""
        xml_string = self._download_webpage(
            url_or_request, video_id, note, errnote, fatal=fatal,
            encoding=encoding, data=data, headers=headers, query=query)
        if xml_string is False:
            return xml_string
        return self._parse_xml(
            xml_string, video_id, transform_source=transform_source,
            fatal=fatal)

    def _parse_xml(self, xml_string, video_id, transform_source=None, fatal=True):
        if transform_source:
            xml_string = transform_source(xml_string)
        try:
            return compat_etree_fromstring(xml_string.encode('utf-8'))
        except compat_xml_parse_error as ve:
            errmsg = '%s: Failed to parse XML ' % video_id
            if fatal:
                raise ExtractorError(errmsg, cause=ve)
            else:
                self.report_warning(errmsg + str(ve))

    def _download_json(self, url_or_request, video_id,
                       note='Downloading JSON metadata',
                       errnote='Unable to download JSON metadata',
                       transform_source=None,
                       fatal=True, encoding=None, data=None, headers={}, query={}):
        json_string = self._download_webpage(
            url_or_request, video_id, note, errnote, fatal=fatal,
            encoding=encoding, data=data, headers=headers, query=query)
        if (not fatal) and json_string is False:
            return None
        return self._parse_json(
            json_string, video_id, transform_source=transform_source, fatal=fatal)

    def _parse_json(self, json_string, video_id, transform_source=None, fatal=True):
        if transform_source:
            json_string = transform_source(json_string)
        try:
            return json.loads(json_string)
        except ValueError as ve:
            errmsg = '%s: Failed to parse JSON ' % video_id
            if fatal:
                raise ExtractorError(errmsg, cause=ve)
            else:
                self.report_warning(errmsg + str(ve))

    def report_warning(self, msg, video_id=None):
        idstr = '' if video_id is None else '%s: ' % video_id
        self._downloader.report_warning(
            '[%s] %s%s' % (self.IE_NAME, idstr, msg))

    def to_screen(self, msg):
        """Print msg to screen, prefixing it with '[ie_name]'"""
        self._downloader.to_screen('[%s] %s' % (self.IE_NAME, msg))

    def report_extraction(self, id_or_name):
        """Report information extraction."""
        self.to_screen('%s: Extracting information' % id_or_name)

    def report_download_webpage(self, video_id):
        """Report webpage download."""
        self.to_screen('%s: Downloading webpage' % video_id)

    def report_age_confirmation(self):
        """Report attempt to confirm age."""
        self.to_screen('Confirming age')

    def report_login(self):
        """Report attempt to log in."""
        self.to_screen('Logging in')

    @staticmethod
    def raise_login_required(msg='This video is only available for registered users'):
        raise ExtractorError(
            '%s. Use --username and --password or --netrc to provide account credentials.' % msg,
            expected=True)

    @staticmethod
    def raise_geo_restricted(msg='This video is not available from your location due to geo restriction', countries=None):
        raise GeoRestrictedError(msg, countries=countries)

    # Methods for following #608
    @staticmethod
    def url_result(url, ie=None, video_id=None, video_title=None):
        """Returns a URL that points to a page that should be processed"""
        # TODO: ie should be the class used for getting the info
        video_info = {'_type': 'url',
                      'url': url,
                      'ie_key': ie}
        if video_id is not None:
            video_info['id'] = video_id
        if video_title is not None:
            video_info['title'] = video_title
        return video_info

    def playlist_from_matches(self, matches, playlist_id=None, playlist_title=None, getter=None, ie=None):
        urls = orderedSet(
            self.url_result(self._proto_relative_url(getter(m) if getter else m), ie)
            for m in matches)
        return self.playlist_result(
            urls, playlist_id=playlist_id, playlist_title=playlist_title)

    @staticmethod
    def playlist_result(entries, playlist_id=None, playlist_title=None, playlist_description=None):
        """Returns a playlist"""
        video_info = {'_type': 'playlist',
                      'entries': entries}
        if playlist_id:
            video_info['id'] = playlist_id
        if playlist_title:
            video_info['title'] = playlist_title
        if playlist_description:
            video_info['description'] = playlist_description
        return video_info

    def _search_regex(self, pattern, string, name, default=NO_DEFAULT, fatal=True, flags=0, group=None):
        """
        Perform a regex search on the given string, using a single or a list of
        patterns returning the first matching group.
        In case of failure return a default value or raise a WARNING or a
        RegexNotFoundError, depending on fatal, specifying the field name.
        """
        if isinstance(pattern, (str, compat_str, compiled_regex_type)):
            mobj = re.search(pattern, string, flags)
        else:
            for p in pattern:
                mobj = re.search(p, string, flags)
                if mobj:
                    break

        if not self._downloader.params.get('no_color') and compat_os_name != 'nt' and sys.stderr.isatty():
            _name = '\033[0;34m%s\033[0m' % name
        else:
            _name = name

        if mobj:
            if group is None:
                # return the first matching group
                return next(g for g in mobj.groups() if g is not None)
            else:
                return mobj.group(group)
        elif default is not NO_DEFAULT:
            return default
        elif fatal:
            raise RegexNotFoundError('Unable to extract %s' % _name)
        else:
            self._downloader.report_warning('unable to extract %s' % _name + bug_reports_message())
            return None

    def _html_search_regex(self, pattern, string, name, default=NO_DEFAULT, fatal=True, flags=0, group=None):
        """
        Like _search_regex, but strips HTML tags and unescapes entities.
        """
        res = self._search_regex(pattern, string, name, default, fatal, flags, group)
        if res:
            return clean_html(res).strip()
        else:
            return res

    def _get_netrc_login_info(self, netrc_machine=None):
        username = None
        password = None
        netrc_machine = netrc_machine or self._NETRC_MACHINE

        if self._downloader.params.get('usenetrc', False):
            try:
                info = netrc.netrc().authenticators(netrc_machine)
                if info is not None:
                    username = info[0]
                    password = info[2]
                else:
                    raise netrc.NetrcParseError(
                        'No authenticators for %s' % netrc_machine)
            except (IOError, netrc.NetrcParseError) as err:
                self._downloader.report_warning(
                    'parsing .netrc: %s' % error_to_compat_str(err))

        return username, password

    def _get_login_info(self, username_option='username', password_option='password', netrc_machine=None):
        """
        Get the login info as (username, password)
        First look for the manually specified credentials using username_option
        and password_option as keys in params dictionary. If no such credentials
        available look in the netrc file using the netrc_machine or _NETRC_MACHINE
        value.
        If there's no info available, return (None, None)
        """
        if self._downloader is None:
            return (None, None)

        downloader_params = self._downloader.params

        # Attempt to use provided username and password or .netrc data
        if downloader_params.get(username_option) is not None:
            username = downloader_params[username_option]
            password = downloader_params[password_option]
        else:
            username, password = self._get_netrc_login_info(netrc_machine)

        return username, password

    def _get_tfa_info(self, note='two-factor verification code'):
        """
        Get the two-factor authentication info
        TODO - asking the user will be required for sms/phone verify
        currently just uses the command line option
        If there's no info available, return None
        """
        if self._downloader is None:
            return None
        downloader_params = self._downloader.params

        if downloader_params.get('twofactor') is not None:
            return downloader_params['twofactor']

        return compat_getpass('Type %s and press [Return]: ' % note)

    # Helper functions for extracting OpenGraph info
    @staticmethod
    def _og_regexes(prop):
        content_re = r'content=(?:"([^"]+?)"|\'([^\']+?)\'|\s*([^\s"\'=<>`]+?))'
        property_re = (r'(?:name|property)=(?:\'og:%(prop)s\'|"og:%(prop)s"|\s*og:%(prop)s\b)'
                       % {'prop': re.escape(prop)})
        template = r'<meta[^>]+?%s[^>]+?%s'
        return [
            template % (property_re, content_re),
            template % (content_re, property_re),
        ]

    @staticmethod
    def _meta_regex(prop):
        return r'''(?isx)<meta
                    (?=[^>]+(?:itemprop|name|property|id|http-equiv)=(["\']?)%s\1)
                    [^>]+?content=(["\'])(?P<content>.*?)\2''' % re.escape(prop)

    def _og_search_property(self, prop, html, name=None, **kargs):
        if not isinstance(prop, (list, tuple)):
            prop = [prop]
        if name is None:
            name = 'OpenGraph %s' % prop[0]
        og_regexes = []
        for p in prop:
            og_regexes.extend(self._og_regexes(p))
        escaped = self._search_regex(og_regexes, html, name, flags=re.DOTALL, **kargs)
        if escaped is None:
            return None
        return unescapeHTML(escaped)

    def _og_search_thumbnail(self, html, **kargs):
        return self._og_search_property('image', html, 'thumbnail URL', fatal=False, **kargs)

    def _og_search_description(self, html, **kargs):
        return self._og_search_property('description', html, fatal=False, **kargs)

    def _og_search_title(self, html, **kargs):
        return self._og_search_property('title', html, **kargs)

    def _og_search_video_url(self, html, name='video url', secure=True, **kargs):
        regexes = self._og_regexes('video') + self._og_regexes('video:url')
        if secure:
            regexes = self._og_regexes('video:secure_url') + regexes
        return self._html_search_regex(regexes, html, name, **kargs)

    def _og_search_url(self, html, **kargs):
        return self._og_search_property('url', html, **kargs)

    def _html_search_meta(self, name, html, display_name=None, fatal=False, **kwargs):
        if not isinstance(name, (list, tuple)):
            name = [name]
        if display_name is None:
            display_name = name[0]
        return self._html_search_regex(
            [self._meta_regex(n) for n in name],
            html, display_name, fatal=fatal, group='content', **kwargs)

    def _dc_search_uploader(self, html):
        return self._html_search_meta('dc.creator', html, 'uploader')

    def _rta_search(self, html):
        # See http://www.rtalabel.org/index.php?content=howtofaq#single
        if re.search(r'(?ix)<meta\s+name="rating"\s+'
                     r'     content="RTA-5042-1996-1400-1577-RTA"',
                     html):
            return 18
        return 0

    def _media_rating_search(self, html):
        # See http://www.tjg-designs.com/WP/metadata-code-examples-adding-metadata-to-your-web-pages/
        rating = self._html_search_meta('rating', html)

        if not rating:
            return None

        RATING_TABLE = {
            'safe for kids': 0,
            'general': 8,
            '14 years': 14,
            'mature': 17,
            'restricted': 19,
        }
        return RATING_TABLE.get(rating.lower())

    def _family_friendly_search(self, html):
        # See http://schema.org/VideoObject
        family_friendly = self._html_search_meta(
            'isFamilyFriendly', html, default=None)

        if not family_friendly:
            return None

        RATING_TABLE = {
            '1': 0,
            'true': 0,
            '0': 18,
            'false': 18,
        }
        return RATING_TABLE.get(family_friendly.lower())

    def _twitter_search_player(self, html):
        return self._html_search_meta('twitter:player', html,
                                      'twitter card player')

    def _search_json_ld(self, html, video_id, expected_type=None, **kwargs):
        json_ld = self._search_regex(
            r'(?s)<script[^>]+type=(["\'])application/ld\+json\1[^>]*>(?P<json_ld>.+?)</script>',
            html, 'JSON-LD', group='json_ld', **kwargs)
        default = kwargs.get('default', NO_DEFAULT)
        if not json_ld:
            return default if default is not NO_DEFAULT else {}
        # JSON-LD may be malformed and thus `fatal` should be respected.
        # At the same time `default` may be passed that assumes `fatal=False`
        # for _search_regex. Let's simulate the same behavior here as well.
        fatal = kwargs.get('fatal', True) if default == NO_DEFAULT else False
        return self._json_ld(json_ld, video_id, fatal=fatal, expected_type=expected_type)

    def _json_ld(self, json_ld, video_id, fatal=True, expected_type=None):
        if isinstance(json_ld, compat_str):
            json_ld = self._parse_json(json_ld, video_id, fatal=fatal)
        if not json_ld:
            return {}
        info = {}
        if not isinstance(json_ld, (list, tuple, dict)):
            return info
        if isinstance(json_ld, dict):
            json_ld = [json_ld]

        def extract_video_object(e):
            assert e['@type'] == 'VideoObject'
            info.update({
                'url': e.get('contentUrl'),
                'title': unescapeHTML(e.get('name')),
                'description': unescapeHTML(e.get('description')),
                'thumbnail': e.get('thumbnailUrl') or e.get('thumbnailURL'),
                'duration': parse_duration(e.get('duration')),
                'timestamp': unified_timestamp(e.get('uploadDate')),
                'filesize': float_or_none(e.get('contentSize')),
                'tbr': int_or_none(e.get('bitrate')),
                'width': int_or_none(e.get('width')),
                'height': int_or_none(e.get('height')),
                'view_count': int_or_none(e.get('interactionCount')),
            })

        for e in json_ld:
            if e.get('@context') == 'http://schema.org':
                item_type = e.get('@type')
                if expected_type is not None and expected_type != item_type:
                    return info
                if item_type in ('TVEpisode', 'Episode'):
                    info.update({
                        'episode': unescapeHTML(e.get('name')),
                        'episode_number': int_or_none(e.get('episodeNumber')),
                        'description': unescapeHTML(e.get('description')),
                    })
                    part_of_season = e.get('partOfSeason')
                    if isinstance(part_of_season, dict) and part_of_season.get('@type') in ('TVSeason', 'Season', 'CreativeWorkSeason'):
                        info['season_number'] = int_or_none(part_of_season.get('seasonNumber'))
                    part_of_series = e.get('partOfSeries') or e.get('partOfTVSeries')
                    if isinstance(part_of_series, dict) and part_of_series.get('@type') in ('TVSeries', 'Series', 'CreativeWorkSeries'):
                        info['series'] = unescapeHTML(part_of_series.get('name'))
                elif item_type == 'Article':
                    info.update({
                        'timestamp': parse_iso8601(e.get('datePublished')),
                        'title': unescapeHTML(e.get('headline')),
                        'description': unescapeHTML(e.get('articleBody')),
                    })
                elif item_type == 'VideoObject':
                    extract_video_object(e)
                    continue
                video = e.get('video')
                if isinstance(video, dict) and video.get('@type') == 'VideoObject':
                    extract_video_object(video)
                break
        return dict((k, v) for k, v in info.items() if v is not None)

    @staticmethod
    def _hidden_inputs(html):
        html = re.sub(r'<!--(?:(?!<!--).)*-->', '', html)
        hidden_inputs = {}
        for input in re.findall(r'(?i)(<input[^>]+>)', html):
            attrs = extract_attributes(input)
            if not input:
                continue
            if attrs.get('type') not in ('hidden', 'submit'):
                continue
            name = attrs.get('name') or attrs.get('id')
            value = attrs.get('value')
            if name and value is not None:
                hidden_inputs[name] = value
        return hidden_inputs

    def _form_hidden_inputs(self, form_id, html):
        form = self._search_regex(
            r'(?is)<form[^>]+?id=(["\'])%s\1[^>]*>(?P<form>.+?)</form>' % form_id,
            html, '%s form' % form_id, group='form')
        return self._hidden_inputs(form)

    def _sort_formats(self, formats, field_preference=None):
        if not formats:
            raise ExtractorError('No video formats found')

        for f in formats:
            # Automatically determine tbr when missing based on abr and vbr (improves
            # formats sorting in some cases)
            if 'tbr' not in f and f.get('abr') is not None and f.get('vbr') is not None:
                f['tbr'] = f['abr'] + f['vbr']

        def _formats_key(f):
            # TODO remove the following workaround
            from ..utils import determine_ext
            if not f.get('ext') and 'url' in f:
                f['ext'] = determine_ext(f['url'])

            if isinstance(field_preference, (list, tuple)):
                return tuple(
                    f.get(field)
                    if f.get(field) is not None
                    else ('' if field == 'format_id' else -1)
                    for field in field_preference)

            preference = f.get('preference')
            if preference is None:
                preference = 0
                if f.get('ext') in ['f4f', 'f4m']:  # Not yet supported
                    preference -= 0.5

            protocol = f.get('protocol') or determine_protocol(f)
            proto_preference = 0 if protocol in ['http', 'https'] else (-0.5 if protocol == 'rtsp' else -0.1)

            if f.get('vcodec') == 'none':  # audio only
                preference -= 50
                if self._downloader.params.get('prefer_free_formats'):
                    ORDER = ['aac', 'mp3', 'm4a', 'webm', 'ogg', 'opus']
                else:
                    ORDER = ['webm', 'opus', 'ogg', 'mp3', 'aac', 'm4a']
                ext_preference = 0
                try:
                    audio_ext_preference = ORDER.index(f['ext'])
                except ValueError:
                    audio_ext_preference = -1
            else:
                if f.get('acodec') == 'none':  # video only
                    preference -= 40
                if self._downloader.params.get('prefer_free_formats'):
                    ORDER = ['flv', 'mp4', 'webm']
                else:
                    ORDER = ['webm', 'flv', 'mp4']
                try:
                    ext_preference = ORDER.index(f['ext'])
                except ValueError:
                    ext_preference = -1
                audio_ext_preference = 0

            return (
                preference,
                f.get('language_preference') if f.get('language_preference') is not None else -1,
                f.get('quality') if f.get('quality') is not None else -1,
                f.get('tbr') if f.get('tbr') is not None else -1,
                f.get('filesize') if f.get('filesize') is not None else -1,
                f.get('vbr') if f.get('vbr') is not None else -1,
                f.get('height') if f.get('height') is not None else -1,
                f.get('width') if f.get('width') is not None else -1,
                proto_preference,
                ext_preference,
                f.get('abr') if f.get('abr') is not None else -1,
                audio_ext_preference,
                f.get('fps') if f.get('fps') is not None else -1,
                f.get('filesize_approx') if f.get('filesize_approx') is not None else -1,
                f.get('source_preference') if f.get('source_preference') is not None else -1,
                f.get('format_id') if f.get('format_id') is not None else '',
            )
        formats.sort(key=_formats_key)

    def _check_formats(self, formats, video_id):
        if formats:
            formats[:] = filter(
                lambda f: self._is_valid_url(
                    f['url'], video_id,
                    item='%s video format' % f.get('format_id') if f.get('format_id') else 'video'),
                formats)

    @staticmethod
    def _remove_duplicate_formats(formats):
        format_urls = set()
        unique_formats = []
        for f in formats:
            if f['url'] not in format_urls:
                format_urls.add(f['url'])
                unique_formats.append(f)
        formats[:] = unique_formats

    def _is_valid_url(self, url, video_id, item='video', headers={}):
        url = self._proto_relative_url(url, scheme='http:')
        # For now assume non HTTP(S) URLs always valid
        if not (url.startswith('http://') or url.startswith('https://')):
            return True
        try:
            self._request_webpage(url, video_id, 'Checking %s URL' % item, headers=headers)
            return True
        except ExtractorError as e:
            if isinstance(e.cause, compat_urllib_error.URLError):
                self.to_screen(
                    '%s: %s URL is invalid, skipping' % (video_id, item))
                return False
            raise

    def http_scheme(self):
        """ Either "http:" or "https:", depending on the user's preferences """
        return (
            'http:'
            if self._downloader.params.get('prefer_insecure', False)
            else 'https:')

    def _proto_relative_url(self, url, scheme=None):
        if url is None:
            return url
        if url.startswith('//'):
            if scheme is None:
                scheme = self.http_scheme()
            return scheme + url
        else:
            return url

    def _sleep(self, timeout, video_id, msg_template=None):
        if msg_template is None:
            msg_template = '%(video_id)s: Waiting for %(timeout)s seconds'
        msg = msg_template % {'video_id': video_id, 'timeout': timeout}
        self.to_screen(msg)
        time.sleep(timeout)

    def _extract_f4m_formats(self, manifest_url, video_id, preference=None, f4m_id=None,
                             transform_source=lambda s: fix_xml_ampersands(s).strip(),
                             fatal=True, m3u8_id=None):
        manifest = self._download_xml(
            manifest_url, video_id, 'Downloading f4m manifest',
            'Unable to download f4m manifest',
            # Some manifests may be malformed, e.g. prosiebensat1 generated manifests
            # (see https://github.com/rg3/youtube-dl/issues/6215#issuecomment-121704244)
            transform_source=transform_source,
            fatal=fatal)

        if manifest is False:
            return []

        return self._parse_f4m_formats(
            manifest, manifest_url, video_id, preference=preference, f4m_id=f4m_id,
            transform_source=transform_source, fatal=fatal, m3u8_id=m3u8_id)

    def _parse_f4m_formats(self, manifest, manifest_url, video_id, preference=None, f4m_id=None,
                           transform_source=lambda s: fix_xml_ampersands(s).strip(),
                           fatal=True, m3u8_id=None):
        # currently youtube-dl cannot decode the playerVerificationChallenge as Akamai uses Adobe Alchemy
        akamai_pv = manifest.find('{http://ns.adobe.com/f4m/1.0}pv-2.0')
        if akamai_pv is not None and ';' in akamai_pv.text:
            playerVerificationChallenge = akamai_pv.text.split(';')[0]
            if playerVerificationChallenge.strip() != '':
                return []

        formats = []
        manifest_version = '1.0'
        media_nodes = manifest.findall('{http://ns.adobe.com/f4m/1.0}media')
        if not media_nodes:
            manifest_version = '2.0'
            media_nodes = manifest.findall('{http://ns.adobe.com/f4m/2.0}media')
        # Remove unsupported DRM protected media from final formats
        # rendition (see https://github.com/rg3/youtube-dl/issues/8573).
        media_nodes = remove_encrypted_media(media_nodes)
        if not media_nodes:
            return formats

        manifest_base_url = get_base_url(manifest)

        bootstrap_info = xpath_element(
            manifest, ['{http://ns.adobe.com/f4m/1.0}bootstrapInfo', '{http://ns.adobe.com/f4m/2.0}bootstrapInfo'],
            'bootstrap info', default=None)

        vcodec = None
        mime_type = xpath_text(
            manifest, ['{http://ns.adobe.com/f4m/1.0}mimeType', '{http://ns.adobe.com/f4m/2.0}mimeType'],
            'base URL', default=None)
        if mime_type and mime_type.startswith('audio/'):
            vcodec = 'none'

        for i, media_el in enumerate(media_nodes):
            tbr = int_or_none(media_el.attrib.get('bitrate'))
            width = int_or_none(media_el.attrib.get('width'))
            height = int_or_none(media_el.attrib.get('height'))
            format_id = '-'.join(filter(None, [f4m_id, compat_str(i if tbr is None else tbr)]))
            # If <bootstrapInfo> is present, the specified f4m is a
            # stream-level manifest, and only set-level manifests may refer to
            # external resources.  See section 11.4 and section 4 of F4M spec
            if bootstrap_info is None:
                media_url = None
                # @href is introduced in 2.0, see section 11.6 of F4M spec
                if manifest_version == '2.0':
                    media_url = media_el.attrib.get('href')
                if media_url is None:
                    media_url = media_el.attrib.get('url')
                if not media_url:
                    continue
                manifest_url = (
                    media_url if media_url.startswith('http://') or media_url.startswith('https://')
                    else ((manifest_base_url or '/'.join(manifest_url.split('/')[:-1])) + '/' + media_url))
                # If media_url is itself a f4m manifest do the recursive extraction
                # since bitrates in parent manifest (this one) and media_url manifest
                # may differ leading to inability to resolve the format by requested
                # bitrate in f4m downloader
                ext = determine_ext(manifest_url)
                if ext == 'f4m':
                    f4m_formats = self._extract_f4m_formats(
                        manifest_url, video_id, preference=preference, f4m_id=f4m_id,
                        transform_source=transform_source, fatal=fatal)
                    # Sometimes stream-level manifest contains single media entry that
                    # does not contain any quality metadata (e.g. http://matchtv.ru/#live-player).
                    # At the same time parent's media entry in set-level manifest may
                    # contain it. We will copy it from parent in such cases.
                    if len(f4m_formats) == 1:
                        f = f4m_formats[0]
                        f.update({
                            'tbr': f.get('tbr') or tbr,
                            'width': f.get('width') or width,
                            'height': f.get('height') or height,
                            'format_id': f.get('format_id') if not tbr else format_id,
                            'vcodec': vcodec,
                        })
                    formats.extend(f4m_formats)
                    continue
                elif ext == 'm3u8':
                    formats.extend(self._extract_m3u8_formats(
                        manifest_url, video_id, 'mp4', preference=preference,
                        m3u8_id=m3u8_id, fatal=fatal))
                    continue
            formats.append({
                'format_id': format_id,
                'url': manifest_url,
                'manifest_url': manifest_url,
                'ext': 'flv' if bootstrap_info is not None else None,
                'protocol': 'f4m',
                'tbr': tbr,
                'width': width,
                'height': height,
                'vcodec': vcodec,
                'preference': preference,
            })
        return formats

    def _m3u8_meta_format(self, m3u8_url, ext=None, preference=None, m3u8_id=None):
        return {
            'format_id': '-'.join(filter(None, [m3u8_id, 'meta'])),
            'url': m3u8_url,
            'ext': ext,
            'protocol': 'm3u8',
            'preference': preference - 100 if preference else -100,
            'resolution': 'multiple',
            'format_note': 'Quality selection URL',
        }

    def _extract_m3u8_formats(self, m3u8_url, video_id, ext=None,
                              entry_protocol='m3u8', preference=None,
                              m3u8_id=None, note=None, errnote=None,
                              fatal=True, live=False):
        res = self._download_webpage_handle(
            m3u8_url, video_id,
            note=note or 'Downloading m3u8 information',
            errnote=errnote or 'Failed to download m3u8 information',
            fatal=fatal)

        if res is False:
            return []

        m3u8_doc, urlh = res
        m3u8_url = urlh.geturl()

        return self._parse_m3u8_formats(
            m3u8_doc, m3u8_url, ext=ext, entry_protocol=entry_protocol,
            preference=preference, m3u8_id=m3u8_id, live=live)

    def _parse_m3u8_formats(self, m3u8_doc, m3u8_url, ext=None,
                            entry_protocol='m3u8', preference=None,
                            m3u8_id=None, live=False):
        if '#EXT-X-FAXS-CM:' in m3u8_doc:  # Adobe Flash Access
            return []

        if re.search(r'#EXT-X-SESSION-KEY:.*?URI="skd://', m3u8_doc):  # Apple FairPlay
            return []

        formats = []

        format_url = lambda u: (
            u
            if re.match(r'^https?://', u)
            else compat_urlparse.urljoin(m3u8_url, u))

        # References:
        # 1. https://tools.ietf.org/html/draft-pantos-http-live-streaming-21
        # 2. https://github.com/rg3/youtube-dl/issues/12211

        # We should try extracting formats only from master playlists [1, 4.3.4],
        # i.e. playlists that describe available qualities. On the other hand
        # media playlists [1, 4.3.3] should be returned as is since they contain
        # just the media without qualities renditions.
        # Fortunately, master playlist can be easily distinguished from media
        # playlist based on particular tags availability. As of [1, 4.3.3, 4.3.4]
        # master playlist tags MUST NOT appear in a media playist and vice versa.
        # As of [1, 4.3.3.1] #EXT-X-TARGETDURATION tag is REQUIRED for every
        # media playlist and MUST NOT appear in master playlist thus we can
        # clearly detect media playlist with this criterion.

        if '#EXT-X-TARGETDURATION' in m3u8_doc:  # media playlist, return as is
            return [{
                'url': m3u8_url,
                'format_id': m3u8_id,
                'ext': ext,
                'protocol': entry_protocol,
                'preference': preference,
            }]

        groups = {}
        last_stream_inf = {}

        def extract_media(x_media_line):
            media = parse_m3u8_attributes(x_media_line)
            # As per [1, 4.3.4.1] TYPE, GROUP-ID and NAME are REQUIRED
            media_type, group_id, name = media.get('TYPE'), media.get('GROUP-ID'), media.get('NAME')
            if not (media_type and group_id and name):
                return
            groups.setdefault(group_id, []).append(media)
            if media_type not in ('VIDEO', 'AUDIO'):
                return
            media_url = media.get('URI')
            if media_url:
                format_id = []
                for v in (m3u8_id, group_id, name):
                    if v:
                        format_id.append(v)
                f = {
                    'format_id': '-'.join(format_id),
                    'url': format_url(media_url),
                    'manifest_url': m3u8_url,
                    'language': media.get('LANGUAGE'),
                    'ext': ext,
                    'protocol': entry_protocol,
                    'preference': preference,
                }
                if media_type == 'AUDIO':
                    f['vcodec'] = 'none'
                formats.append(f)

        def build_stream_name():
            # Despite specification does not mention NAME attribute for
            # EXT-X-STREAM-INF tag it still sometimes may be present (see [1]
            # or vidio test in TestInfoExtractor.test_parse_m3u8_formats)
            # 1. http://www.vidio.com/watch/165683-dj_ambred-booyah-live-2015
            stream_name = last_stream_inf.get('NAME')
            if stream_name:
                return stream_name
            # If there is no NAME in EXT-X-STREAM-INF it will be obtained
            # from corresponding rendition group
            stream_group_id = last_stream_inf.get('VIDEO')
            if not stream_group_id:
                return
            stream_group = groups.get(stream_group_id)
            if not stream_group:
                return stream_group_id
            rendition = stream_group[0]
            return rendition.get('NAME') or stream_group_id

        for line in m3u8_doc.splitlines():
            if line.startswith('#EXT-X-STREAM-INF:'):
                last_stream_inf = parse_m3u8_attributes(line)
            elif line.startswith('#EXT-X-MEDIA:'):
                extract_media(line)
            elif line.startswith('#') or not line.strip():
                continue
            else:
                tbr = float_or_none(
                    last_stream_inf.get('AVERAGE-BANDWIDTH') or
                    last_stream_inf.get('BANDWIDTH'), scale=1000)
                format_id = []
                if m3u8_id:
                    format_id.append(m3u8_id)
                stream_name = build_stream_name()
                # Bandwidth of live streams may differ over time thus making
                # format_id unpredictable. So it's better to keep provided
                # format_id intact.
                if not live:
                    format_id.append(stream_name if stream_name else '%d' % (tbr if tbr else len(formats)))
                manifest_url = format_url(line.strip())
                f = {
                    'format_id': '-'.join(format_id),
                    'url': manifest_url,
                    'manifest_url': m3u8_url,
                    'tbr': tbr,
                    'ext': ext,
                    'fps': float_or_none(last_stream_inf.get('FRAME-RATE')),
                    'protocol': entry_protocol,
                    'preference': preference,
                }
                resolution = last_stream_inf.get('RESOLUTION')
                if resolution:
                    mobj = re.search(r'(?P<width>\d+)[xX](?P<height>\d+)', resolution)
                    if mobj:
                        f['width'] = int(mobj.group('width'))
                        f['height'] = int(mobj.group('height'))
                # Unified Streaming Platform
                mobj = re.search(
                    r'audio.*?(?:%3D|=)(\d+)(?:-video.*?(?:%3D|=)(\d+))?', f['url'])
                if mobj:
                    abr, vbr = mobj.groups()
                    abr, vbr = float_or_none(abr, 1000), float_or_none(vbr, 1000)
                    f.update({
                        'vbr': vbr,
                        'abr': abr,
                    })
                codecs = parse_codecs(last_stream_inf.get('CODECS'))
                f.update(codecs)
                audio_group_id = last_stream_inf.get('AUDIO')
                # As per [1, 4.3.4.1.1] any EXT-X-STREAM-INF tag which
                # references a rendition group MUST have a CODECS attribute.
                # However, this is not always respected, for example, [2]
                # contains EXT-X-STREAM-INF tag which references AUDIO
                # rendition group but does not have CODECS and despite
                # referencing audio group an audio group, it represents
                # a complete (with audio and video) format. So, for such cases
                # we will ignore references to rendition groups and treat them
                # as complete formats.
                if audio_group_id and codecs and f.get('vcodec') != 'none':
                    audio_group = groups.get(audio_group_id)
                    if audio_group and audio_group[0].get('URI'):
                        # TODO: update acodec for audio only formats with
                        # the same GROUP-ID
                        f['acodec'] = 'none'
                formats.append(f)
                last_stream_inf = {}
        return formats

    @staticmethod
    def _xpath_ns(path, namespace=None):
        if not namespace:
            return path
        out = []
        for c in path.split('/'):
            if not c or c == '.':
                out.append(c)
            else:
                out.append('{%s}%s' % (namespace, c))
        return '/'.join(out)

    def _extract_smil_formats(self, smil_url, video_id, fatal=True, f4m_params=None, transform_source=None):
        smil = self._download_smil(smil_url, video_id, fatal=fatal, transform_source=transform_source)

        if smil is False:
            assert not fatal
            return []

        namespace = self._parse_smil_namespace(smil)

        return self._parse_smil_formats(
            smil, smil_url, video_id, namespace=namespace, f4m_params=f4m_params)

    def _extract_smil_info(self, smil_url, video_id, fatal=True, f4m_params=None):
        smil = self._download_smil(smil_url, video_id, fatal=fatal)
        if smil is False:
            return {}
        return self._parse_smil(smil, smil_url, video_id, f4m_params=f4m_params)

    def _download_smil(self, smil_url, video_id, fatal=True, transform_source=None):
        return self._download_xml(
            smil_url, video_id, 'Downloading SMIL file',
            'Unable to download SMIL file', fatal=fatal, transform_source=transform_source)

    def _parse_smil(self, smil, smil_url, video_id, f4m_params=None):
        namespace = self._parse_smil_namespace(smil)

        formats = self._parse_smil_formats(
            smil, smil_url, video_id, namespace=namespace, f4m_params=f4m_params)
        subtitles = self._parse_smil_subtitles(smil, namespace=namespace)

        video_id = os.path.splitext(url_basename(smil_url))[0]
        title = None
        description = None
        upload_date = None
        for meta in smil.findall(self._xpath_ns('./head/meta', namespace)):
            name = meta.attrib.get('name')
            content = meta.attrib.get('content')
            if not name or not content:
                continue
            if not title and name == 'title':
                title = content
            elif not description and name in ('description', 'abstract'):
                description = content
            elif not upload_date and name == 'date':
                upload_date = unified_strdate(content)

        thumbnails = [{
            'id': image.get('type'),
            'url': image.get('src'),
            'width': int_or_none(image.get('width')),
            'height': int_or_none(image.get('height')),
        } for image in smil.findall(self._xpath_ns('.//image', namespace)) if image.get('src')]

        return {
            'id': video_id,
            'title': title or video_id,
            'description': description,
            'upload_date': upload_date,
            'thumbnails': thumbnails,
            'formats': formats,
            'subtitles': subtitles,
        }

    def _parse_smil_namespace(self, smil):
        return self._search_regex(
            r'(?i)^{([^}]+)?}smil$', smil.tag, 'namespace', default=None)

    def _parse_smil_formats(self, smil, smil_url, video_id, namespace=None, f4m_params=None, transform_rtmp_url=None):
        base = smil_url
        for meta in smil.findall(self._xpath_ns('./head/meta', namespace)):
            b = meta.get('base') or meta.get('httpBase')
            if b:
                base = b
                break

        formats = []
        rtmp_count = 0
        http_count = 0
        m3u8_count = 0

        srcs = []
        media = smil.findall(self._xpath_ns('.//video', namespace)) + smil.findall(self._xpath_ns('.//audio', namespace))
        for medium in media:
            src = medium.get('src')
            if not src or src in srcs:
                continue
            srcs.append(src)

            bitrate = float_or_none(medium.get('system-bitrate') or medium.get('systemBitrate'), 1000)
            filesize = int_or_none(medium.get('size') or medium.get('fileSize'))
            width = int_or_none(medium.get('width'))
            height = int_or_none(medium.get('height'))
            proto = medium.get('proto')
            ext = medium.get('ext')
            src_ext = determine_ext(src)
            streamer = medium.get('streamer') or base

            if proto == 'rtmp' or streamer.startswith('rtmp'):
                rtmp_count += 1
                formats.append({
                    'url': streamer,
                    'play_path': src,
                    'ext': 'flv',
                    'format_id': 'rtmp-%d' % (rtmp_count if bitrate is None else bitrate),
                    'tbr': bitrate,
                    'filesize': filesize,
                    'width': width,
                    'height': height,
                })
                if transform_rtmp_url:
                    streamer, src = transform_rtmp_url(streamer, src)
                    formats[-1].update({
                        'url': streamer,
                        'play_path': src,
                    })
                continue

            src_url = src if src.startswith('http') else compat_urlparse.urljoin(base, src)
            src_url = src_url.strip()

            if proto == 'm3u8' or src_ext == 'm3u8':
                m3u8_formats = self._extract_m3u8_formats(
                    src_url, video_id, ext or 'mp4', m3u8_id='hls', fatal=False)
                if len(m3u8_formats) == 1:
                    m3u8_count += 1
                    m3u8_formats[0].update({
                        'format_id': 'hls-%d' % (m3u8_count if bitrate is None else bitrate),
                        'tbr': bitrate,
                        'width': width,
                        'height': height,
                    })
                formats.extend(m3u8_formats)
                continue

            if src_ext == 'f4m':
                f4m_url = src_url
                if not f4m_params:
                    f4m_params = {
                        'hdcore': '3.2.0',
                        'plugin': 'flowplayer-3.2.0.1',
                    }
                f4m_url += '&' if '?' in f4m_url else '?'
                f4m_url += compat_urllib_parse_urlencode(f4m_params)
                formats.extend(self._extract_f4m_formats(f4m_url, video_id, f4m_id='hds', fatal=False))
                continue

            if src_url.startswith('http') and self._is_valid_url(src, video_id):
                http_count += 1
                formats.append({
                    'url': src_url,
                    'ext': ext or src_ext or 'flv',
                    'format_id': 'http-%d' % (bitrate or http_count),
                    'tbr': bitrate,
                    'filesize': filesize,
                    'width': width,
                    'height': height,
                })
                continue

        return formats

    def _parse_smil_subtitles(self, smil, namespace=None, subtitles_lang='en'):
        urls = []
        subtitles = {}
        for num, textstream in enumerate(smil.findall(self._xpath_ns('.//textstream', namespace))):
            src = textstream.get('src')
            if not src or src in urls:
                continue
            urls.append(src)
            ext = textstream.get('ext') or mimetype2ext(textstream.get('type')) or determine_ext(src)
            lang = textstream.get('systemLanguage') or textstream.get('systemLanguageName') or textstream.get('lang') or subtitles_lang
            subtitles.setdefault(lang, []).append({
                'url': src,
                'ext': ext,
            })
        return subtitles

    def _extract_xspf_playlist(self, playlist_url, playlist_id, fatal=True):
        xspf = self._download_xml(
            playlist_url, playlist_id, 'Downloading xpsf playlist',
            'Unable to download xspf manifest', fatal=fatal)
        if xspf is False:
            return []
        return self._parse_xspf(xspf, playlist_id)

    def _parse_xspf(self, playlist, playlist_id):
        NS_MAP = {
            'xspf': 'http://xspf.org/ns/0/',
            's1': 'http://static.streamone.nl/player/ns/0',
        }

        entries = []
        for track in playlist.findall(xpath_with_ns('./xspf:trackList/xspf:track', NS_MAP)):
            title = xpath_text(
                track, xpath_with_ns('./xspf:title', NS_MAP), 'title', default=playlist_id)
            description = xpath_text(
                track, xpath_with_ns('./xspf:annotation', NS_MAP), 'description')
            thumbnail = xpath_text(
                track, xpath_with_ns('./xspf:image', NS_MAP), 'thumbnail')
            duration = float_or_none(
                xpath_text(track, xpath_with_ns('./xspf:duration', NS_MAP), 'duration'), 1000)

            formats = [{
                'url': location.text,
                'format_id': location.get(xpath_with_ns('s1:label', NS_MAP)),
                'width': int_or_none(location.get(xpath_with_ns('s1:width', NS_MAP))),
                'height': int_or_none(location.get(xpath_with_ns('s1:height', NS_MAP))),
            } for location in track.findall(xpath_with_ns('./xspf:location', NS_MAP))]
            self._sort_formats(formats)

            entries.append({
                'id': playlist_id,
                'title': title,
                'description': description,
                'thumbnail': thumbnail,
                'duration': duration,
                'formats': formats,
            })
        return entries

    def _extract_mpd_formats(self, mpd_url, video_id, mpd_id=None, note=None, errnote=None, fatal=True, formats_dict={}):
        res = self._download_webpage_handle(
            mpd_url, video_id,
            note=note or 'Downloading MPD manifest',
            errnote=errnote or 'Failed to download MPD manifest',
            fatal=fatal)
        if res is False:
            return []
        mpd, urlh = res
        mpd_base_url = base_url(urlh.geturl())

        return self._parse_mpd_formats(
            compat_etree_fromstring(mpd.encode('utf-8')), mpd_id, mpd_base_url,
            formats_dict=formats_dict, mpd_url=mpd_url)

    def _parse_mpd_formats(self, mpd_doc, mpd_id=None, mpd_base_url='', formats_dict={}, mpd_url=None):
        """
        Parse formats from MPD manifest.
        References:
         1. MPEG-DASH Standard, ISO/IEC 23009-1:2014(E),
            http://standards.iso.org/ittf/PubliclyAvailableStandards/c065274_ISO_IEC_23009-1_2014.zip
         2. https://en.wikipedia.org/wiki/Dynamic_Adaptive_Streaming_over_HTTP
        """
        if mpd_doc.get('type') == 'dynamic':
            return []

        namespace = self._search_regex(r'(?i)^{([^}]+)?}MPD$', mpd_doc.tag, 'namespace', default=None)

        def _add_ns(path):
            return self._xpath_ns(path, namespace)

        def is_drm_protected(element):
            return element.find(_add_ns('ContentProtection')) is not None

        def extract_multisegment_info(element, ms_parent_info):
            ms_info = ms_parent_info.copy()

            # As per [1, 5.3.9.2.2] SegmentList and SegmentTemplate share some
            # common attributes and elements.  We will only extract relevant
            # for us.
            def extract_common(source):
                segment_timeline = source.find(_add_ns('SegmentTimeline'))
                if segment_timeline is not None:
                    s_e = segment_timeline.findall(_add_ns('S'))
                    if s_e:
                        ms_info['total_number'] = 0
                        ms_info['s'] = []
                        for s in s_e:
                            r = int(s.get('r', 0))
                            ms_info['total_number'] += 1 + r
                            ms_info['s'].append({
                                't': int(s.get('t', 0)),
                                # @d is mandatory (see [1, 5.3.9.6.2, Table 17, page 60])
                                'd': int(s.attrib['d']),
                                'r': r,
                            })
                start_number = source.get('startNumber')
                if start_number:
                    ms_info['start_number'] = int(start_number)
                timescale = source.get('timescale')
                if timescale:
                    ms_info['timescale'] = int(timescale)
                segment_duration = source.get('duration')
                if segment_duration:
                    ms_info['segment_duration'] = float(segment_duration)

            def extract_Initialization(source):
                initialization = source.find(_add_ns('Initialization'))
                if initialization is not None:
                    ms_info['initialization_url'] = initialization.attrib['sourceURL']

            segment_list = element.find(_add_ns('SegmentList'))
            if segment_list is not None:
                extract_common(segment_list)
                extract_Initialization(segment_list)
                segment_urls_e = segment_list.findall(_add_ns('SegmentURL'))
                if segment_urls_e:
                    ms_info['segment_urls'] = [segment.attrib['media'] for segment in segment_urls_e]
            else:
                segment_template = element.find(_add_ns('SegmentTemplate'))
                if segment_template is not None:
                    extract_common(segment_template)
                    media = segment_template.get('media')
                    if media:
                        ms_info['media'] = media
                    initialization = segment_template.get('initialization')
                    if initialization:
                        ms_info['initialization'] = initialization
                    else:
                        extract_Initialization(segment_template)
            return ms_info

        mpd_duration = parse_duration(mpd_doc.get('mediaPresentationDuration'))
        formats = []
        for period in mpd_doc.findall(_add_ns('Period')):
            period_duration = parse_duration(period.get('duration')) or mpd_duration
            period_ms_info = extract_multisegment_info(period, {
                'start_number': 1,
                'timescale': 1,
            })
            for adaptation_set in period.findall(_add_ns('AdaptationSet')):
                if is_drm_protected(adaptation_set):
                    continue
                adaption_set_ms_info = extract_multisegment_info(adaptation_set, period_ms_info)
                for representation in adaptation_set.findall(_add_ns('Representation')):
                    if is_drm_protected(representation):
                        continue
                    representation_attrib = adaptation_set.attrib.copy()
                    representation_attrib.update(representation.attrib)
                    # According to [1, 5.3.7.2, Table 9, page 41], @mimeType is mandatory
                    mime_type = representation_attrib['mimeType']
                    content_type = mime_type.split('/')[0]
                    if content_type == 'text':
                        # TODO implement WebVTT downloading
                        pass
                    elif content_type in ('video', 'audio'):
                        base_url = ''
                        for element in (representation, adaptation_set, period, mpd_doc):
                            base_url_e = element.find(_add_ns('BaseURL'))
                            if base_url_e is not None:
                                base_url = base_url_e.text + base_url
                                if re.match(r'^https?://', base_url):
                                    break
                        if mpd_base_url and not re.match(r'^https?://', base_url):
                            if not mpd_base_url.endswith('/') and not base_url.startswith('/'):
                                mpd_base_url += '/'
                            base_url = mpd_base_url + base_url
                        representation_id = representation_attrib.get('id')
                        lang = representation_attrib.get('lang')
                        url_el = representation.find(_add_ns('BaseURL'))
                        filesize = int_or_none(url_el.attrib.get('{http://youtube.com/yt/2012/10/10}contentLength') if url_el is not None else None)
                        bandwidth = int_or_none(representation_attrib.get('bandwidth'))
                        f = {
                            'format_id': '%s-%s' % (mpd_id, representation_id) if mpd_id else representation_id,
                            'url': base_url,
                            'manifest_url': mpd_url,
                            'ext': mimetype2ext(mime_type),
                            'width': int_or_none(representation_attrib.get('width')),
                            'height': int_or_none(representation_attrib.get('height')),
                            'tbr': float_or_none(bandwidth, 1000),
                            'asr': int_or_none(representation_attrib.get('audioSamplingRate')),
                            'fps': int_or_none(representation_attrib.get('frameRate')),
                            'language': lang if lang not in ('mul', 'und', 'zxx', 'mis') else None,
                            'format_note': 'DASH %s' % content_type,
                            'filesize': filesize,
                            'container': mimetype2ext(mime_type) + '_dash',
                        }
                        f.update(parse_codecs(representation_attrib.get('codecs')))
                        representation_ms_info = extract_multisegment_info(representation, adaption_set_ms_info)

                        def prepare_template(template_name, identifiers):
                            t = representation_ms_info[template_name]
                            t = t.replace('$RepresentationID$', representation_id)
                            t = re.sub(r'\$(%s)\$' % '|'.join(identifiers), r'%(\1)d', t)
                            t = re.sub(r'\$(%s)%%([^$]+)\$' % '|'.join(identifiers), r'%(\1)\2', t)
                            t.replace('$$', '$')
                            return t

                        # @initialization is a regular template like @media one
                        # so it should be handled just the same way (see
                        # https://github.com/rg3/youtube-dl/issues/11605)
                        if 'initialization' in representation_ms_info:
                            initialization_template = prepare_template(
                                'initialization',
                                # As per [1, 5.3.9.4.2, Table 15, page 54] $Number$ and
                                # $Time$ shall not be included for @initialization thus
                                # only $Bandwidth$ remains
                                ('Bandwidth', ))
                            representation_ms_info['initialization_url'] = initialization_template % {
                                'Bandwidth': bandwidth,
                            }

                        def location_key(location):
                            return 'url' if re.match(r'^https?://', location) else 'path'

                        if 'segment_urls' not in representation_ms_info and 'media' in representation_ms_info:

                            media_template = prepare_template('media', ('Number', 'Bandwidth', 'Time'))
                            media_location_key = location_key(media_template)

                            # As per [1, 5.3.9.4.4, Table 16, page 55] $Number$ and $Time$
                            # can't be used at the same time
                            if '%(Number' in media_template and 's' not in representation_ms_info:
                                segment_duration = None
                                if 'total_number' not in representation_ms_info and 'segment_duration' in representation_ms_info:
                                    segment_duration = float_or_none(representation_ms_info['segment_duration'], representation_ms_info['timescale'])
                                    representation_ms_info['total_number'] = int(math.ceil(float(period_duration) / segment_duration))
                                representation_ms_info['fragments'] = [{
                                    media_location_key: media_template % {
                                        'Number': segment_number,
                                        'Bandwidth': bandwidth,
                                    },
                                    'duration': segment_duration,
                                } for segment_number in range(
                                    representation_ms_info['start_number'],
                                    representation_ms_info['total_number'] + representation_ms_info['start_number'])]
                            else:
                                # $Number*$ or $Time$ in media template with S list available
                                # Example $Number*$: http://www.svtplay.se/klipp/9023742/stopptid-om-bjorn-borg
                                # Example $Time$: https://play.arkena.com/embed/avp/v2/player/media/b41dda37-d8e7-4d3f-b1b5-9a9db578bdfe/1/129411
                                representation_ms_info['fragments'] = []
                                segment_time = 0
                                segment_d = None
                                segment_number = representation_ms_info['start_number']

                                def add_segment_url():
                                    segment_url = media_template % {
                                        'Time': segment_time,
                                        'Bandwidth': bandwidth,
                                        'Number': segment_number,
                                    }
                                    representation_ms_info['fragments'].append({
                                        media_location_key: segment_url,
                                        'duration': float_or_none(segment_d, representation_ms_info['timescale']),
                                    })

                                for num, s in enumerate(representation_ms_info['s']):
                                    segment_time = s.get('t') or segment_time
                                    segment_d = s['d']
                                    add_segment_url()
                                    segment_number += 1
                                    for r in range(s.get('r', 0)):
                                        segment_time += segment_d
                                        add_segment_url()
                                        segment_number += 1
                                    segment_time += segment_d
                        elif 'segment_urls' in representation_ms_info and 's' in representation_ms_info:
                            # No media template
                            # Example: https://www.youtube.com/watch?v=iXZV5uAYMJI
                            # or any YouTube dashsegments video
                            fragments = []
                            segment_index = 0
                            timescale = representation_ms_info['timescale']
                            for s in representation_ms_info['s']:
                                duration = float_or_none(s['d'], timescale)
                                for r in range(s.get('r', 0) + 1):
                                    segment_uri = representation_ms_info['segment_urls'][segment_index]
                                    fragments.append({
                                        location_key(segment_uri): segment_uri,
                                        'duration': duration,
                                    })
                                    segment_index += 1
                            representation_ms_info['fragments'] = fragments
                        elif 'segment_urls' in representation_ms_info:
                            # Segment URLs with no SegmentTimeline
                            # Example: https://www.seznam.cz/zpravy/clanek/cesko-zasahne-vitr-o-sile-vichrice-muze-byt-i-zivotu-nebezpecny-39091
                            # https://github.com/rg3/youtube-dl/pull/14844
                            fragments = []
                            segment_duration = float_or_none(
                                representation_ms_info['segment_duration'],
                                representation_ms_info['timescale']) if 'segment_duration' in representation_ms_info else None
                            for segment_url in representation_ms_info['segment_urls']:
                                fragment = {
                                    location_key(segment_url): segment_url,
                                }
                                if segment_duration:
                                    fragment['duration'] = segment_duration
                                fragments.append(fragment)
                            representation_ms_info['fragments'] = fragments
                        # NB: MPD manifest may contain direct URLs to unfragmented media.
                        # No fragments key is present in this case.
                        if 'fragments' in representation_ms_info:
                            f.update({
                                'fragment_base_url': base_url,
                                'fragments': [],
                                'protocol': 'http_dash_segments',
                            })
                            if 'initialization_url' in representation_ms_info:
                                initialization_url = representation_ms_info['initialization_url']
                                if not f.get('url'):
                                    f['url'] = initialization_url
                                f['fragments'].append({location_key(initialization_url): initialization_url})
                            f['fragments'].extend(representation_ms_info['fragments'])
                        # According to [1, 5.3.5.2, Table 7, page 35] @id of Representation
                        # is not necessarily unique within a Period thus formats with
                        # the same `format_id` are quite possible. There are numerous examples
                        # of such manifests (see https://github.com/rg3/youtube-dl/issues/15111,
                        # https://github.com/rg3/youtube-dl/issues/13919)
                        full_info = formats_dict.get(representation_id, {}).copy()
                        full_info.update(f)
                        formats.append(full_info)
                    else:
                        self.report_warning('Unknown MIME type %s in DASH manifest' % mime_type)
        return formats

    def _extract_ism_formats(self, ism_url, video_id, ism_id=None, note=None, errnote=None, fatal=True):
        res = self._download_webpage_handle(
            ism_url, video_id,
            note=note or 'Downloading ISM manifest',
            errnote=errnote or 'Failed to download ISM manifest',
            fatal=fatal)
        if res is False:
            return []
        ism, urlh = res

        return self._parse_ism_formats(
            compat_etree_fromstring(ism.encode('utf-8')), urlh.geturl(), ism_id)

    def _parse_ism_formats(self, ism_doc, ism_url, ism_id=None):
        """
        Parse formats from ISM manifest.
        References:
         1. [MS-SSTR]: Smooth Streaming Protocol,
            https://msdn.microsoft.com/en-us/library/ff469518.aspx
        """
        if ism_doc.get('IsLive') == 'TRUE' or ism_doc.find('Protection') is not None:
            return []

        duration = int(ism_doc.attrib['Duration'])
        timescale = int_or_none(ism_doc.get('TimeScale')) or 10000000

        formats = []
        for stream in ism_doc.findall('StreamIndex'):
            stream_type = stream.get('Type')
            if stream_type not in ('video', 'audio'):
                continue
            url_pattern = stream.attrib['Url']
            stream_timescale = int_or_none(stream.get('TimeScale')) or timescale
            stream_name = stream.get('Name')
            for track in stream.findall('QualityLevel'):
                fourcc = track.get('FourCC', 'AACL' if track.get('AudioTag') == '255' else None)
                # TODO: add support for WVC1 and WMAP
                if fourcc not in ('H264', 'AVC1', 'AACL'):
                    self.report_warning('%s is not a supported codec' % fourcc)
                    continue
                tbr = int(track.attrib['Bitrate']) // 1000
                # [1] does not mention Width and Height attributes. However,
                # they're often present while MaxWidth and MaxHeight are
                # missing, so should be used as fallbacks
                width = int_or_none(track.get('MaxWidth') or track.get('Width'))
                height = int_or_none(track.get('MaxHeight') or track.get('Height'))
                sampling_rate = int_or_none(track.get('SamplingRate'))

                track_url_pattern = re.sub(r'{[Bb]itrate}', track.attrib['Bitrate'], url_pattern)
                track_url_pattern = compat_urlparse.urljoin(ism_url, track_url_pattern)

                fragments = []
                fragment_ctx = {
                    'time': 0,
                }
                stream_fragments = stream.findall('c')
                for stream_fragment_index, stream_fragment in enumerate(stream_fragments):
                    fragment_ctx['time'] = int_or_none(stream_fragment.get('t')) or fragment_ctx['time']
                    fragment_repeat = int_or_none(stream_fragment.get('r')) or 1
                    fragment_ctx['duration'] = int_or_none(stream_fragment.get('d'))
                    if not fragment_ctx['duration']:
                        try:
                            next_fragment_time = int(stream_fragment[stream_fragment_index + 1].attrib['t'])
                        except IndexError:
                            next_fragment_time = duration
                        fragment_ctx['duration'] = (next_fragment_time - fragment_ctx['time']) / fragment_repeat
                    for _ in range(fragment_repeat):
                        fragments.append({
                            'url': re.sub(r'{start[ _]time}', compat_str(fragment_ctx['time']), track_url_pattern),
                            'duration': fragment_ctx['duration'] / stream_timescale,
                        })
                        fragment_ctx['time'] += fragment_ctx['duration']

                format_id = []
                if ism_id:
                    format_id.append(ism_id)
                if stream_name:
                    format_id.append(stream_name)
                format_id.append(compat_str(tbr))

                formats.append({
                    'format_id': '-'.join(format_id),
                    'url': ism_url,
                    'manifest_url': ism_url,
                    'ext': 'ismv' if stream_type == 'video' else 'isma',
                    'width': width,
                    'height': height,
                    'tbr': tbr,
                    'asr': sampling_rate,
                    'vcodec': 'none' if stream_type == 'audio' else fourcc,
                    'acodec': 'none' if stream_type == 'video' else fourcc,
                    'protocol': 'ism',
                    'fragments': fragments,
                    '_download_params': {
                        'duration': duration,
                        'timescale': stream_timescale,
                        'width': width or 0,
                        'height': height or 0,
                        'fourcc': fourcc,
                        'codec_private_data': track.get('CodecPrivateData'),
                        'sampling_rate': sampling_rate,
                        'channels': int_or_none(track.get('Channels', 2)),
                        'bits_per_sample': int_or_none(track.get('BitsPerSample', 16)),
                        'nal_unit_length_field': int_or_none(track.get('NALUnitLengthField', 4)),
                    },
                })
        return formats

    def _parse_html5_media_entries(self, base_url, webpage, video_id, m3u8_id=None, m3u8_entry_protocol='m3u8', mpd_id=None, preference=None):
        def absolute_url(video_url):
            return compat_urlparse.urljoin(base_url, video_url)

        def parse_content_type(content_type):
            if not content_type:
                return {}
            ctr = re.search(r'(?P<mimetype>[^/]+/[^;]+)(?:;\s*codecs="?(?P<codecs>[^"]+))?', content_type)
            if ctr:
                mimetype, codecs = ctr.groups()
                f = parse_codecs(codecs)
                f['ext'] = mimetype2ext(mimetype)
                return f
            return {}

        def _media_formats(src, cur_media_type, type_info={}):
            full_url = absolute_url(src)
            ext = type_info.get('ext') or determine_ext(full_url)
            if ext == 'm3u8':
                is_plain_url = False
                formats = self._extract_m3u8_formats(
                    full_url, video_id, ext='mp4',
                    entry_protocol=m3u8_entry_protocol, m3u8_id=m3u8_id,
                    preference=preference, fatal=False)
            elif ext == 'mpd':
                is_plain_url = False
                formats = self._extract_mpd_formats(
                    full_url, video_id, mpd_id=mpd_id, fatal=False)
            else:
                is_plain_url = True
                formats = [{
                    'url': full_url,
                    'vcodec': 'none' if cur_media_type == 'audio' else None,
                }]
            return is_plain_url, formats

        entries = []
        # amp-video and amp-audio are very similar to their HTML5 counterparts
        # so we wll include them right here (see
        # https://www.ampproject.org/docs/reference/components/amp-video)
        media_tags = [(media_tag, media_type, '')
                      for media_tag, media_type
                      in re.findall(r'(?s)(<(?:amp-)?(video|audio)[^>]*/>)', webpage)]
        media_tags.extend(re.findall(
            # We only allow video|audio followed by a whitespace or '>'.
            # Allowing more characters may end up in significant slow down (see
            # https://github.com/rg3/youtube-dl/issues/11979, example URL:
            # http://www.porntrex.com/maps/videositemap.xml).
            r'(?s)(<(?P<tag>(?:amp-)?(?:video|audio))(?:\s+[^>]*)?>)(.*?)</(?P=tag)>', webpage))
        for media_tag, media_type, media_content in media_tags:
            media_info = {
                'formats': [],
                'subtitles': {},
            }
            media_attributes = extract_attributes(media_tag)
            src = media_attributes.get('src')
            if src:
                _, formats = _media_formats(src, media_type)
                media_info['formats'].extend(formats)
            media_info['thumbnail'] = media_attributes.get('poster')
            if media_content:
                for source_tag in re.findall(r'<source[^>]+>', media_content):
                    source_attributes = extract_attributes(source_tag)
                    src = source_attributes.get('src')
                    if not src:
                        continue
                    f = parse_content_type(source_attributes.get('type'))
                    is_plain_url, formats = _media_formats(src, media_type, f)
                    if is_plain_url:
                        # res attribute is not standard but seen several times
                        # in the wild
                        f.update({
                            'height': int_or_none(source_attributes.get('res')),
                            'format_id': source_attributes.get('label'),
                        })
                        f.update(formats[0])
                        media_info['formats'].append(f)
                    else:
                        media_info['formats'].extend(formats)
                for track_tag in re.findall(r'<track[^>]+>', media_content):
                    track_attributes = extract_attributes(track_tag)
                    kind = track_attributes.get('kind')
                    if not kind or kind in ('subtitles', 'captions'):
                        src = track_attributes.get('src')
                        if not src:
                            continue
                        lang = track_attributes.get('srclang') or track_attributes.get('lang') or track_attributes.get('label')
                        media_info['subtitles'].setdefault(lang, []).append({
                            'url': absolute_url(src),
                        })
            if media_info['formats'] or media_info['subtitles']:
                entries.append(media_info)
        return entries

    def _extract_akamai_formats(self, manifest_url, video_id, hosts={}):
        formats = []
        hdcore_sign = 'hdcore=3.7.0'
        f4m_url = re.sub(r'(https?://[^/]+)/i/', r'\1/z/', manifest_url).replace('/master.m3u8', '/manifest.f4m')
        hds_host = hosts.get('hds')
        if hds_host:
            f4m_url = re.sub(r'(https?://)[^/]+', r'\1' + hds_host, f4m_url)
        if 'hdcore=' not in f4m_url:
            f4m_url += ('&' if '?' in f4m_url else '?') + hdcore_sign
        f4m_formats = self._extract_f4m_formats(
            f4m_url, video_id, f4m_id='hds', fatal=False)
        for entry in f4m_formats:
            entry.update({'extra_param_to_segment_url': hdcore_sign})
        formats.extend(f4m_formats)
        m3u8_url = re.sub(r'(https?://[^/]+)/z/', r'\1/i/', manifest_url).replace('/manifest.f4m', '/master.m3u8')
        hls_host = hosts.get('hls')
        if hls_host:
            m3u8_url = re.sub(r'(https?://)[^/]+', r'\1' + hls_host, m3u8_url)
        formats.extend(self._extract_m3u8_formats(
            m3u8_url, video_id, 'mp4', 'm3u8_native',
            m3u8_id='hls', fatal=False))
        return formats

    def _extract_wowza_formats(self, url, video_id, m3u8_entry_protocol='m3u8_native', skip_protocols=[]):
        query = compat_urlparse.urlparse(url).query
        url = re.sub(r'/(?:manifest|playlist|jwplayer)\.(?:m3u8|f4m|mpd|smil)', '', url)
        url_base = self._search_regex(
            r'(?:(?:https?|rtmp|rtsp):)?(//[^?]+)', url, 'format url')
        http_base_url = '%s:%s' % ('http', url_base)
        formats = []

        def manifest_url(manifest):
            m_url = '%s/%s' % (http_base_url, manifest)
            if query:
                m_url += '?%s' % query
            return m_url

        if 'm3u8' not in skip_protocols:
            formats.extend(self._extract_m3u8_formats(
                manifest_url('playlist.m3u8'), video_id, 'mp4',
                m3u8_entry_protocol, m3u8_id='hls', fatal=False))
        if 'f4m' not in skip_protocols:
            formats.extend(self._extract_f4m_formats(
                manifest_url('manifest.f4m'),
                video_id, f4m_id='hds', fatal=False))
        if 'dash' not in skip_protocols:
            formats.extend(self._extract_mpd_formats(
                manifest_url('manifest.mpd'),
                video_id, mpd_id='dash', fatal=False))
        if re.search(r'(?:/smil:|\.smil)', url_base):
            if 'smil' not in skip_protocols:
                rtmp_formats = self._extract_smil_formats(
                    manifest_url('jwplayer.smil'),
                    video_id, fatal=False)
                for rtmp_format in rtmp_formats:
                    rtsp_format = rtmp_format.copy()
                    rtsp_format['url'] = '%s/%s' % (rtmp_format['url'], rtmp_format['play_path'])
                    del rtsp_format['play_path']
                    del rtsp_format['ext']
                    rtsp_format.update({
                        'url': rtsp_format['url'].replace('rtmp://', 'rtsp://'),
                        'format_id': rtmp_format['format_id'].replace('rtmp', 'rtsp'),
                        'protocol': 'rtsp',
                    })
                    formats.extend([rtmp_format, rtsp_format])
        else:
            for protocol in ('rtmp', 'rtsp'):
                if protocol not in skip_protocols:
                    formats.append({
                        'url': '%s:%s' % (protocol, url_base),
                        'format_id': protocol,
                        'protocol': protocol,
                    })
        return formats

    def _find_jwplayer_data(self, webpage, video_id=None, transform_source=js_to_json):
        mobj = re.search(
            r'(?s)jwplayer\((?P<quote>[\'"])[^\'" ]+(?P=quote)\)(?!</script>).*?\.setup\s*\((?P<options>[^)]+)\)',
            webpage)
        if mobj:
            try:
                jwplayer_data = self._parse_json(mobj.group('options'),
                                                 video_id=video_id,
                                                 transform_source=transform_source)
            except ExtractorError:
                pass
            else:
                if isinstance(jwplayer_data, dict):
                    return jwplayer_data

    def _extract_jwplayer_data(self, webpage, video_id, *args, **kwargs):
        jwplayer_data = self._find_jwplayer_data(
            webpage, video_id, transform_source=js_to_json)
        return self._parse_jwplayer_data(
            jwplayer_data, video_id, *args, **kwargs)

    def _parse_jwplayer_data(self, jwplayer_data, video_id=None, require_title=True,
                             m3u8_id=None, mpd_id=None, rtmp_params=None, base_url=None):
        # JWPlayer backward compatibility: flattened playlists
        # https://github.com/jwplayer/jwplayer/blob/v7.4.3/src/js/api/config.js#L81-L96
        if 'playlist' not in jwplayer_data:
            jwplayer_data = {'playlist': [jwplayer_data]}

        entries = []

        # JWPlayer backward compatibility: single playlist item
        # https://github.com/jwplayer/jwplayer/blob/v7.7.0/src/js/playlist/playlist.js#L10
        if not isinstance(jwplayer_data['playlist'], list):
            jwplayer_data['playlist'] = [jwplayer_data['playlist']]

        for video_data in jwplayer_data['playlist']:
            # JWPlayer backward compatibility: flattened sources
            # https://github.com/jwplayer/jwplayer/blob/v7.4.3/src/js/playlist/item.js#L29-L35
            if 'sources' not in video_data:
                video_data['sources'] = [video_data]

            this_video_id = video_id or video_data['mediaid']

            formats = self._parse_jwplayer_formats(
                video_data['sources'], video_id=this_video_id, m3u8_id=m3u8_id,
                mpd_id=mpd_id, rtmp_params=rtmp_params, base_url=base_url)

            subtitles = {}
            tracks = video_data.get('tracks')
            if tracks and isinstance(tracks, list):
                for track in tracks:
                    if not isinstance(track, dict):
                        continue
                    if track.get('kind') != 'captions':
                        continue
                    track_url = urljoin(base_url, track.get('file'))
                    if not track_url:
                        continue
                    subtitles.setdefault(track.get('label') or 'en', []).append({
                        'url': self._proto_relative_url(track_url)
                    })

            entry = {
                'id': this_video_id,
                'title': unescapeHTML(video_data['title'] if require_title else video_data.get('title')),
                'description': video_data.get('description'),
                'thumbnail': self._proto_relative_url(video_data.get('image')),
                'timestamp': int_or_none(video_data.get('pubdate')),
                'duration': float_or_none(jwplayer_data.get('duration') or video_data.get('duration')),
                'subtitles': subtitles,
            }
            # https://github.com/jwplayer/jwplayer/blob/master/src/js/utils/validator.js#L32
            if len(formats) == 1 and re.search(r'^(?:http|//).*(?:youtube\.com|youtu\.be)/.+', formats[0]['url']):
                entry.update({
                    '_type': 'url_transparent',
                    'url': formats[0]['url'],
                })
            else:
                self._sort_formats(formats)
                entry['formats'] = formats
            entries.append(entry)
        if len(entries) == 1:
            return entries[0]
        else:
            return self.playlist_result(entries)

    def _parse_jwplayer_formats(self, jwplayer_sources_data, video_id=None,
                                m3u8_id=None, mpd_id=None, rtmp_params=None, base_url=None):
        urls = []
        formats = []
        for source in jwplayer_sources_data:
            if not isinstance(source, dict):
                continue
            source_url = self._proto_relative_url(source.get('file'))
            if not source_url:
                continue
            if base_url:
                source_url = compat_urlparse.urljoin(base_url, source_url)
            if source_url in urls:
                continue
            urls.append(source_url)
            source_type = source.get('type') or ''
            ext = mimetype2ext(source_type) or determine_ext(source_url)
            if source_type == 'hls' or ext == 'm3u8':
                formats.extend(self._extract_m3u8_formats(
                    source_url, video_id, 'mp4', entry_protocol='m3u8_native',
                    m3u8_id=m3u8_id, fatal=False))
            elif ext == 'mpd':
                formats.extend(self._extract_mpd_formats(
                    source_url, video_id, mpd_id=mpd_id, fatal=False))
            elif ext == 'smil':
                formats.extend(self._extract_smil_formats(
                    source_url, video_id, fatal=False))
            # https://github.com/jwplayer/jwplayer/blob/master/src/js/providers/default.js#L67
            elif source_type.startswith('audio') or ext in (
                    'oga', 'aac', 'mp3', 'mpeg', 'vorbis'):
                formats.append({
                    'url': source_url,
                    'vcodec': 'none',
                    'ext': ext,
                })
            else:
                height = int_or_none(source.get('height'))
                if height is None:
                    # Often no height is provided but there is a label in
                    # format like "1080p", "720p SD", or 1080.
                    height = int_or_none(self._search_regex(
                        r'^(\d{3,4})[pP]?(?:\b|$)', compat_str(source.get('label') or ''),
                        'height', default=None))
                a_format = {
                    'url': source_url,
                    'width': int_or_none(source.get('width')),
                    'height': height,
                    'tbr': int_or_none(source.get('bitrate')),
                    'ext': ext,
                }
                if source_url.startswith('rtmp'):
                    a_format['ext'] = 'flv'
                    # See com/longtailvideo/jwplayer/media/RTMPMediaProvider.as
                    # of jwplayer.flash.swf
                    rtmp_url_parts = re.split(
                        r'((?:mp4|mp3|flv):)', source_url, 1)
                    if len(rtmp_url_parts) == 3:
                        rtmp_url, prefix, play_path = rtmp_url_parts
                        a_format.update({
                            'url': rtmp_url,
                            'play_path': prefix + play_path,
                        })
                    if rtmp_params:
                        a_format.update(rtmp_params)
                formats.append(a_format)
        return formats

    def _live_title(self, name):
        """ Generate the title for a live video """
        now = datetime.datetime.now()
        now_str = now.strftime('%Y-%m-%d %H:%M')
        return name + ' ' + now_str

    def _int(self, v, name, fatal=False, **kwargs):
        res = int_or_none(v, **kwargs)
        if 'get_attr' in kwargs:
            print(getattr(v, kwargs['get_attr']))
        if res is None:
            msg = 'Failed to extract %s: Could not parse value %r' % (name, v)
            if fatal:
                raise ExtractorError(msg)
            else:
                self._downloader.report_warning(msg)
        return res

    def _float(self, v, name, fatal=False, **kwargs):
        res = float_or_none(v, **kwargs)
        if res is None:
            msg = 'Failed to extract %s: Could not parse value %r' % (name, v)
            if fatal:
                raise ExtractorError(msg)
            else:
                self._downloader.report_warning(msg)
        return res

    def _set_cookie(self, domain, name, value, expire_time=None, port=None,
                    path='/', secure=False, discard=False, rest={}, **kwargs):
        cookie = compat_cookiejar.Cookie(
            0, name, value, port, port is not None, domain, True,
            domain.startswith('.'), path, True, secure, expire_time,
            discard, None, None, rest)
        self._downloader.cookiejar.set_cookie(cookie)

    def _get_cookies(self, url):
        """ Return a compat_cookies.SimpleCookie with the cookies for the url """
        req = sanitized_Request(url)
        self._downloader.cookiejar.add_cookie_header(req)
        return compat_cookies.SimpleCookie(req.get_header('Cookie'))

    def get_testcases(self, include_onlymatching=False):
        t = getattr(self, '_TEST', None)
        if t:
            assert not hasattr(self, '_TESTS'), \
                '%s has _TEST and _TESTS' % type(self).__name__
            tests = [t]
        else:
            tests = getattr(self, '_TESTS', [])
        for t in tests:
            if not include_onlymatching and t.get('only_matching', False):
                continue
            t['name'] = type(self).__name__[:-len('IE')]
            yield t

    def is_suitable(self, age_limit):
        """ Test whether the extractor is generally suitable for the given
        age limit (i.e. pornographic sites are not, all others usually are) """

        any_restricted = False
        for tc in self.get_testcases(include_onlymatching=False):
            if tc.get('playlist', []):
                tc = tc['playlist'][0]
            is_restricted = age_restricted(
                tc.get('info_dict', {}).get('age_limit'), age_limit)
            if not is_restricted:
                return True
            any_restricted = any_restricted or is_restricted
        return not any_restricted

    def extract_subtitles(self, *args, **kwargs):
        if (self._downloader.params.get('writesubtitles', False) or
                self._downloader.params.get('listsubtitles')):
            return self._get_subtitles(*args, **kwargs)
        return {}

    def _get_subtitles(self, *args, **kwargs):
        raise NotImplementedError('This method must be implemented by subclasses')

    @staticmethod
    def _merge_subtitle_items(subtitle_list1, subtitle_list2):
        """ Merge subtitle items for one language. Items with duplicated URLs
        will be dropped. """
        list1_urls = set([item['url'] for item in subtitle_list1])
        ret = list(subtitle_list1)
        ret.extend([item for item in subtitle_list2 if item['url'] not in list1_urls])
        return ret

    @classmethod
    def _merge_subtitles(cls, subtitle_dict1, subtitle_dict2):
        """ Merge two subtitle dictionaries, language by language. """
        ret = dict(subtitle_dict1)
        for lang in subtitle_dict2:
            ret[lang] = cls._merge_subtitle_items(subtitle_dict1.get(lang, []), subtitle_dict2[lang])
        return ret

    def extract_automatic_captions(self, *args, **kwargs):
        if (self._downloader.params.get('writeautomaticsub', False) or
                self._downloader.params.get('listsubtitles')):
            return self._get_automatic_captions(*args, **kwargs)
        return {}

    def _get_automatic_captions(self, *args, **kwargs):
        raise NotImplementedError('This method must be implemented by subclasses')

    def mark_watched(self, *args, **kwargs):
        if (self._downloader.params.get('mark_watched', False) and
                (self._get_login_info()[0] is not None or
                    self._downloader.params.get('cookiefile') is not None)):
            self._mark_watched(*args, **kwargs)

    def _mark_watched(self, *args, **kwargs):
        raise NotImplementedError('This method must be implemented by subclasses')

    def geo_verification_headers(self):
        headers = {}
        geo_verification_proxy = self._downloader.params.get('geo_verification_proxy')
        if geo_verification_proxy:
            headers['Ytdl-request-proxy'] = geo_verification_proxy
        return headers

    def _generic_id(self, url):
        return compat_urllib_parse_unquote(os.path.splitext(url.rstrip('/').split('/')[-1])[0])

    def _generic_title(self, url):
        return compat_urllib_parse_unquote(os.path.splitext(url_basename(url))[0])


class SearchInfoExtractor(InfoExtractor):
    """
    Base class for paged search queries extractors.
    They accept URLs in the format _SEARCH_KEY(|all|[0-9]):{query}
    Instances should define _SEARCH_KEY and _MAX_RESULTS.
    """

    @classmethod
    def _make_valid_url(cls):
        return r'%s(?P<prefix>|[1-9][0-9]*|all):(?P<query>[\s\S]+)' % cls._SEARCH_KEY

    @classmethod
    def suitable(cls, url):
        return re.match(cls._make_valid_url(), url) is not None

    def _real_extract(self, query):
        mobj = re.match(self._make_valid_url(), query)
        if mobj is None:
            raise ExtractorError('Invalid search query "%s"' % query)

        prefix = mobj.group('prefix')
        query = mobj.group('query')
        if prefix == '':
            return self._get_n_results(query, 1)
        elif prefix == 'all':
            return self._get_n_results(query, self._MAX_RESULTS)
        else:
            n = int(prefix)
            if n <= 0:
                raise ExtractorError('invalid download number %s for query "%s"' % (n, query))
            elif n > self._MAX_RESULTS:
                self._downloader.report_warning('%s returns max %i results (you requested %i)' % (self._SEARCH_KEY, self._MAX_RESULTS, n))
                n = self._MAX_RESULTS
            return self._get_n_results(query, n)

    def _get_n_results(self, query, n):
        """Get a specified number of results for a query"""
        raise NotImplementedError('This method must be implemented by subclasses')

    @property
    def SEARCH_KEY(self):
        return self._SEARCH_KEY