]> jfr.im git - yt-dlp.git/blame - yt_dlp/extractor/twitch.py
[cleanup] Misc fixes
[yt-dlp.git] / yt_dlp / extractor / twitch.py
CommitLineData
04d02a9d 1# coding: utf-8
ee1e1996
PH
2from __future__ import unicode_literals
3
841b6838 4import collections
3182f3e2 5import itertools
48afc6ca 6import json
841b6838
S
7import random
8import re
79e93125
PH
9
10from .common import InfoExtractor
1cc79574 11from ..compat import (
e704f87f 12 compat_parse_qs,
240b9b7a 13 compat_str,
15707c7e 14 compat_urllib_parse_urlencode,
e704f87f 15 compat_urllib_parse_urlparse,
1cc79574
PH
16)
17from ..utils import (
efe470e2 18 clean_html,
30a074c2 19 dict_get,
79e93125 20 ExtractorError,
841b6838 21 float_or_none,
7a6e8a1b 22 int_or_none,
e704f87f 23 parse_duration,
355d074f 24 parse_iso8601,
4dfbf869 25 parse_qs,
a0455d0f 26 qualities,
639f80c1 27 str_or_none,
28 traverse_obj,
49fa7de3
S
29 try_get,
30 unified_timestamp,
264e77c4 31 update_url_query,
3052a30d 32 url_or_none,
c64c03be 33 urljoin,
79e93125
PH
34)
35
36
c5db6bb3 37class TwitchBaseIE(InfoExtractor):
fce79626 38 _VALID_URL_BASE = r'https?://(?:(?:www|go|m)\.)?twitch\.tv'
c5db6bb3 39
46fd0dd5 40 _API_BASE = 'https://api.twitch.tv'
9f4576a7 41 _USHER_BASE = 'https://usher.ttvnw.net'
48afc6ca
TB
42 _LOGIN_FORM_URL = 'https://www.twitch.tv/login'
43 _LOGIN_POST_URL = 'https://passport.twitch.tv/login'
cb1c3a3c 44 _CLIENT_ID = 'kimne78kx3ncx6brgo4mv6wki5h1ko'
499bfcbf 45 _NETRC_MACHINE = 'twitch'
79e93125 46
00dd0cd5 47 _OPERATION_HASHES = {
48 'CollectionSideBar': '27111f1b382effad0b6def325caef1909c733fe6a4fbabf54f8d491ef2cf2f14',
49 'FilterableVideoTower_Videos': 'a937f1d22e269e39a03b509f65a7490f9fc247d7f83d6ac1421523e3b68042cb',
50 'ClipsCards__User': 'b73ad2bfaecfd30a9e6c28fada15bd97032c83ec77a0440766a56fe0bd632777',
51 'ChannelCollectionsContent': '07e3691a1bad77a36aba590c351180439a40baefc1c275356f40fc7082419a84',
52 'StreamMetadata': '1c719a40e481453e5c48d9bb585d971b8b372f8ebb105b17076722264dfa5b3e',
53 'ComscoreStreamingQuery': 'e1edae8122517d013405f237ffcc124515dc6ded82480a88daef69c83b53ac01',
14eb1ee1 54 'VideoAccessToken_Clip': '36b89d2507fce29e5ca551df756d27c1cfe079e2609642b4390aa4c35796eb11',
00dd0cd5 55 'VideoPreviewOverlay': '3006e77e51b128d838fa4e835723ca4dc9a05c5efd4466c1085215c6e437e65c',
56 'VideoMetadata': '226edb3e692509f727fd56821f5653c05740242c82b0388883e0c0e75dcbf687',
639f80c1 57 'VideoPlayer_ChapterSelectButtonVideo': '8d2793384aac3773beab5e59bd5d6f585aedb923d292800119e03d40cd0f9b41',
00dd0cd5 58 }
355d074f 59
52efa4b3 60 def _perform_login(self, username, password):
efe470e2
S
61 def fail(message):
62 raise ExtractorError(
63 'Unable to login. Twitch said: %s' % message, expected=True)
64
5316566e 65 def login_step(page, urlh, note, data):
c64c03be
XDG
66 form = self._hidden_inputs(page)
67 form.update(data)
68
69 page_url = urlh.geturl()
70 post_url = self._search_regex(
71 r'<form[^>]+action=(["\'])(?P<url>.+?)\1', page,
48afc6ca 72 'post url', default=self._LOGIN_POST_URL, group='url')
c64c03be
XDG
73 post_url = urljoin(page_url, post_url)
74
48afc6ca
TB
75 headers = {
76 'Referer': page_url,
30a074c2 77 'Origin': 'https://www.twitch.tv',
cb1c3a3c 78 'Content-Type': 'text/plain;charset=UTF-8',
48afc6ca 79 }
c64c03be 80
cb1c3a3c
S
81 response = self._download_json(
82 post_url, None, note, data=json.dumps(form).encode(),
83 headers=headers, expected_status=400)
30a074c2 84 error = dict_get(response, ('error', 'error_description', 'error_code'))
cb1c3a3c
S
85 if error:
86 fail(error)
c64c03be 87
1370dba5
S
88 if 'Authenticated successfully' in response.get('message', ''):
89 return None, None
90
91 redirect_url = urljoin(
92 post_url,
93 response.get('redirect') or response['redirect_path'])
5316566e
S
94 return self._download_webpage_handle(
95 redirect_url, None, 'Downloading login redirect page',
96 headers=headers)
c64c03be 97
fbd9f6ea 98 login_page, handle = self._download_webpage_handle(
48afc6ca 99 self._LOGIN_FORM_URL, None, 'Downloading login page')
c5db6bb3 100
efe470e2
S
101 # Some TOR nodes and public proxies are blocked completely
102 if 'blacklist_message' in login_page:
103 fail(clean_html(login_page))
104
5316566e 105 redirect_page, handle = login_step(
e4d95865 106 login_page, handle, 'Logging in', {
5316566e
S
107 'username': username,
108 'password': password,
48afc6ca 109 'client_id': self._CLIENT_ID,
fe646a2f 110 })
c64c03be 111
1370dba5
S
112 # Successful login
113 if not redirect_page:
114 return
115
c64c03be
XDG
116 if re.search(r'(?i)<form[^>]+id="two-factor-submit"', redirect_page) is not None:
117 # TODO: Add mechanism to request an SMS or phone call
118 tfa_token = self._get_tfa_info('two-factor authentication token')
5316566e 119 login_step(redirect_page, handle, 'Submitting TFA token', {
c64c03be
XDG
120 'authy_token': tfa_token,
121 'remember_2fa': 'true',
5316566e 122 })
17b41a33 123
d0e958c7
PH
124 def _prefer_source(self, formats):
125 try:
126 source = next(f for f in formats if f['format_id'] == 'Source')
60e67c5b 127 source['quality'] = 10
d0e958c7 128 except StopIteration:
985637cb
S
129 for f in formats:
130 if '/chunked/' in f['url']:
131 f.update({
60e67c5b 132 'quality': 10,
985637cb
S
133 'format_note': 'Source',
134 })
d0e958c7
PH
135 self._sort_formats(formats)
136
00dd0cd5 137 def _download_base_gql(self, video_id, ops, note, fatal=True):
30a074c2 138 headers = {
139 'Content-Type': 'text/plain;charset=UTF-8',
140 'Client-ID': self._CLIENT_ID,
141 }
142 gql_auth = self._get_cookies('https://gql.twitch.tv').get('auth-token')
143 if gql_auth:
144 headers['Authorization'] = 'OAuth ' + gql_auth.value
00dd0cd5 145 return self._download_json(
146 'https://gql.twitch.tv/gql', video_id, note,
147 data=json.dumps(ops).encode(),
30a074c2 148 headers=headers, fatal=fatal)
c5db6bb3 149
00dd0cd5 150 def _download_gql(self, video_id, ops, note, fatal=True):
151 for op in ops:
152 op['extensions'] = {
153 'persistedQuery': {
154 'version': 1,
155 'sha256Hash': self._OPERATION_HASHES[op['operationName']],
156 }
157 }
158 return self._download_base_gql(video_id, ops, note)
159
160 def _download_access_token(self, video_id, token_kind, param_name):
161 method = '%sPlaybackAccessToken' % token_kind
162 ops = {
163 'query': '''{
164 %s(
165 %s: "%s",
166 params: {
167 platform: "web",
168 playerBackend: "mediaplayer",
169 playerType: "site"
170 }
171 )
172 {
173 value
174 signature
175 }
176 }''' % (method, param_name, video_id),
177 }
178 return self._download_base_gql(
179 video_id, ops,
180 'Downloading %s access token GraphQL' % token_kind)['data'][method]
04d02a9d 181
04d02a9d 182
841b6838 183class TwitchVodIE(TwitchBaseIE):
c5db6bb3 184 IE_NAME = 'twitch:vod'
3f1ce168
S
185 _VALID_URL = r'''(?x)
186 https?://
187 (?:
644921b3 188 (?:(?:www|go|m)\.)?twitch\.tv/(?:[^/]+/v(?:ideo)?|videos)/|
bc48773e 189 player\.twitch\.tv/\?.*?\bvideo=v?
3f1ce168
S
190 )
191 (?P<id>\d+)
192 '''
c5db6bb3 193
9c724a98 194 _TESTS = [{
e704f87f 195 'url': 'http://www.twitch.tv/riotgames/v/6528877?t=5m10s',
c5db6bb3 196 'info_dict': {
ac0474f8 197 'id': 'v6528877',
c5db6bb3 198 'ext': 'mp4',
ac0474f8 199 'title': 'LCK Summer Split - Week 6 Day 1',
ec85ded8 200 'thumbnail': r're:^https?://.*\.jpg$',
ac0474f8 201 'duration': 17208,
00dd0cd5 202 'timestamp': 1435131734,
ac0474f8
YCH
203 'upload_date': '20150624',
204 'uploader': 'Riot Games',
205 'uploader_id': 'riotgames',
c5db6bb3 206 'view_count': int,
e704f87f 207 'start_time': 310,
c5db6bb3
S
208 },
209 'params': {
210 # m3u8 download
211 'skip_download': True,
212 },
9c724a98
S
213 }, {
214 # Untitled broadcast (title is None)
215 'url': 'http://www.twitch.tv/belkao_o/v/11230755',
216 'info_dict': {
217 'id': 'v11230755',
218 'ext': 'mp4',
219 'title': 'Untitled Broadcast',
ec85ded8 220 'thumbnail': r're:^https?://.*\.jpg$',
9c724a98
S
221 'duration': 1638,
222 'timestamp': 1439746708,
223 'upload_date': '20150816',
224 'uploader': 'BelkAO_o',
225 'uploader_id': 'belkao_o',
226 'view_count': int,
227 },
228 'params': {
229 # m3u8 download
230 'skip_download': True,
231 },
9bd7bd0b 232 'skip': 'HTTP Error 404: Not Found',
3f1ce168
S
233 }, {
234 'url': 'http://player.twitch.tv/?t=5m10s&video=v6528877',
235 'only_matching': True,
0b23c222
S
236 }, {
237 'url': 'https://www.twitch.tv/videos/6528877',
238 'only_matching': True,
fce79626
S
239 }, {
240 'url': 'https://m.twitch.tv/beagsandjam/v/247478721',
241 'only_matching': True,
644921b3
S
242 }, {
243 'url': 'https://www.twitch.tv/northernlion/video/291940395',
244 'only_matching': True,
bc48773e
M
245 }, {
246 'url': 'https://player.twitch.tv/?video=480452374',
247 'only_matching': True,
639f80c1 248 }, {
249 'url': 'https://www.twitch.tv/videos/635475444',
250 'info_dict': {
251 'id': 'v635475444',
252 'ext': 'mp4',
253 'title': 'Riot Games',
254 'duration': 11643,
255 'uploader': 'Riot Games',
256 'uploader_id': 'riotgames',
257 'timestamp': 1590770569,
258 'upload_date': '20200529',
259 'chapters': [
260 {
261 'start_time': 0,
262 'end_time': 573,
263 'title': 'League of Legends'
264 },
265 {
266 'start_time': 573,
267 'end_time': 3922,
268 'title': 'Legends of Runeterra'
269 },
270 {
271 'start_time': 3922,
272 'end_time': 11643,
273 'title': 'Art'
274 }
275 ],
276 },
277 'params': {
278 'skip_download': True
279 }
9c724a98 280 }]
04d02a9d 281
841b6838 282 def _download_info(self, item_id):
00dd0cd5 283 data = self._download_gql(
284 item_id, [{
285 'operationName': 'VideoMetadata',
286 'variables': {
287 'channelLogin': '',
288 'videoID': item_id,
289 },
639f80c1 290 }, {
291 'operationName': 'VideoPlayer_ChapterSelectButtonVideo',
292 'variables': {
293 'includePrivate': False,
294 'videoID': item_id,
295 },
00dd0cd5 296 }],
639f80c1 297 'Downloading stream metadata GraphQL')
298
299 video = traverse_obj(data, (0, 'data', 'video'))
300 video['moments'] = traverse_obj(data, (1, 'data', 'video', 'moments', 'edges', ..., 'node'))
301
00dd0cd5 302 if video is None:
303 raise ExtractorError(
304 'Video %s does not exist' % item_id, expected=True)
305 return self._extract_info_gql(video, item_id)
841b6838 306
639f80c1 307 def _extract_info(self, info):
841b6838
S
308 status = info.get('status')
309 if status == 'recording':
310 is_live = True
311 elif status == 'recorded':
312 is_live = False
313 else:
314 is_live = None
315 _QUALITIES = ('small', 'medium', 'large')
316 quality_key = qualities(_QUALITIES)
317 thumbnails = []
318 preview = info.get('preview')
319 if isinstance(preview, dict):
320 for thumbnail_id, thumbnail_url in preview.items():
321 thumbnail_url = url_or_none(thumbnail_url)
322 if not thumbnail_url:
323 continue
324 if thumbnail_id not in _QUALITIES:
325 continue
326 thumbnails.append({
327 'url': thumbnail_url,
328 'preference': quality_key(thumbnail_id),
329 })
330 return {
331 'id': info['_id'],
332 'title': info.get('title') or 'Untitled Broadcast',
333 'description': info.get('description'),
334 'duration': int_or_none(info.get('length')),
335 'thumbnails': thumbnails,
336 'uploader': info.get('channel', {}).get('display_name'),
337 'uploader_id': info.get('channel', {}).get('name'),
338 'timestamp': parse_iso8601(info.get('recorded_at')),
339 'view_count': int_or_none(info.get('views')),
340 'is_live': is_live,
a33c0d9c 341 'was_live': True,
841b6838
S
342 }
343
639f80c1 344 def _extract_moments(self, info, item_id):
345 for moment in info.get('moments') or []:
346 start_time = int_or_none(moment.get('positionMilliseconds'), 1000)
347 duration = int_or_none(moment.get('durationMilliseconds'), 1000)
348 name = str_or_none(moment.get('description'))
349
350 if start_time is None or duration is None:
351 self.report_warning(f'Important chapter information missing for chapter {name}', item_id)
352 continue
353 yield {
354 'start_time': start_time,
355 'end_time': start_time + duration,
356 'title': name,
357 }
358
359 def _extract_info_gql(self, info, item_id):
00dd0cd5 360 vod_id = info.get('id') or item_id
361 # id backward compatibility for download archives
362 if vod_id[0] != 'v':
363 vod_id = 'v%s' % vod_id
364 thumbnail = url_or_none(info.get('previewThumbnailURL'))
a33c0d9c 365 is_live = None
00dd0cd5 366 if thumbnail:
a33c0d9c
D
367 if thumbnail.endswith('/404_processing_{width}x{height}.png'):
368 is_live, thumbnail = True, None
369 else:
370 is_live = False
371 for p in ('width', 'height'):
372 thumbnail = thumbnail.replace('{%s}' % p, '0')
639f80c1 373
00dd0cd5 374 return {
375 'id': vod_id,
376 'title': info.get('title') or 'Untitled Broadcast',
377 'description': info.get('description'),
378 'duration': int_or_none(info.get('lengthSeconds')),
379 'thumbnail': thumbnail,
380 'uploader': try_get(info, lambda x: x['owner']['displayName'], compat_str),
381 'uploader_id': try_get(info, lambda x: x['owner']['login'], compat_str),
382 'timestamp': unified_timestamp(info.get('publishedAt')),
383 'view_count': int_or_none(info.get('viewCount')),
639f80c1 384 'chapters': list(self._extract_moments(info, item_id)),
a33c0d9c
D
385 'is_live': is_live,
386 'was_live': True,
00dd0cd5 387 }
388
79e93125 389 def _real_extract(self, url):
841b6838 390 vod_id = self._match_id(url)
e5e99661 391
841b6838 392 info = self._download_info(vod_id)
00dd0cd5 393 access_token = self._download_access_token(vod_id, 'video', 'id')
e5e99661 394
c5db6bb3 395 formats = self._extract_m3u8_formats(
2da4316e 396 '%s/vod/%s.m3u8?%s' % (
841b6838 397 self._USHER_BASE, vod_id,
15707c7e 398 compat_urllib_parse_urlencode({
e5e99661 399 'allow_source': 'true',
ac455055 400 'allow_audio_only': 'true',
e5e99661
S
401 'allow_spectre': 'true',
402 'player': 'twitchweb',
1d31b7ca 403 'playlist_include_framerate': 'true',
00dd0cd5 404 'nauth': access_token['value'],
405 'nauthsig': access_token['signature'],
e5e99661 406 })),
841b6838 407 vod_id, 'mp4', entry_protocol='m3u8_native')
e5e99661 408
d0e958c7 409 self._prefer_source(formats)
c5db6bb3 410 info['formats'] = formats
e704f87f
NH
411
412 parsed_url = compat_urllib_parse_urlparse(url)
413 query = compat_parse_qs(parsed_url.query)
414 if 't' in query:
415 info['start_time'] = parse_duration(query['t'][0])
416
264e77c4
S
417 if info.get('timestamp') is not None:
418 info['subtitles'] = {
419 'rechat': [{
420 'url': update_url_query(
841b6838 421 'https://api.twitch.tv/v5/videos/%s/comments' % vod_id, {
ce112a8c 422 'client_id': self._CLIENT_ID,
264e77c4
S
423 }),
424 'ext': 'json',
425 }],
426 }
427
c5db6bb3
S
428 return info
429
430
841b6838
S
431def _make_video_result(node):
432 assert isinstance(node, dict)
433 video_id = node.get('id')
434 if not video_id:
435 return
436 return {
437 '_type': 'url_transparent',
438 'ie_key': TwitchVodIE.ie_key(),
c76eb41b 439 'id': 'v' + video_id,
841b6838
S
440 'url': 'https://www.twitch.tv/videos/%s' % video_id,
441 'title': node.get('title'),
442 'thumbnail': node.get('previewThumbnailURL'),
443 'duration': float_or_none(node.get('lengthSeconds')),
444 'view_count': int_or_none(node.get('viewCount')),
445 }
446
447
00dd0cd5 448class TwitchCollectionIE(TwitchBaseIE):
841b6838
S
449 _VALID_URL = r'https?://(?:(?:www|go|m)\.)?twitch\.tv/collections/(?P<id>[^/]+)'
450
451 _TESTS = [{
452 'url': 'https://www.twitch.tv/collections/wlDCoH0zEBZZbQ',
453 'info_dict': {
454 'id': 'wlDCoH0zEBZZbQ',
455 'title': 'Overthrow Nook, capitalism for children',
456 },
457 'playlist_mincount': 13,
458 }]
459
460 _OPERATION_NAME = 'CollectionSideBar'
841b6838
S
461
462 def _real_extract(self, url):
463 collection_id = self._match_id(url)
464 collection = self._download_gql(
defc820b
S
465 collection_id, [{
466 'operationName': self._OPERATION_NAME,
467 'variables': {'collectionID': collection_id},
468 }],
469 'Downloading collection GraphQL')[0]['data']['collection']
841b6838 470 title = collection.get('title')
c5db6bb3 471 entries = []
841b6838
S
472 for edge in collection['items']['edges']:
473 if not isinstance(edge, dict):
474 continue
475 node = edge.get('node')
476 if not isinstance(node, dict):
477 continue
478 video = _make_video_result(node)
479 if video:
480 entries.append(video)
481 return self.playlist_result(
482 entries, playlist_id=collection_id, playlist_title=title)
483
484
00dd0cd5 485class TwitchPlaylistBaseIE(TwitchBaseIE):
486 _PAGE_LIMIT = 100
487
841b6838
S
488 def _entries(self, channel_name, *args):
489 cursor = None
490 variables_common = self._make_variables(channel_name, *args)
491 entries_key = '%ss' % self._ENTRY_KIND
492 for page_num in itertools.count(1):
493 variables = variables_common.copy()
494 variables['limit'] = self._PAGE_LIMIT
495 if cursor:
496 variables['cursor'] = cursor
497 page = self._download_gql(
defc820b
S
498 channel_name, [{
499 'operationName': self._OPERATION_NAME,
500 'variables': variables,
501 }],
841b6838
S
502 'Downloading %ss GraphQL page %s' % (self._NODE_KIND, page_num),
503 fatal=False)
504 if not page:
505 break
506 edges = try_get(
defc820b 507 page, lambda x: x[0]['data']['user'][entries_key]['edges'], list)
841b6838
S
508 if not edges:
509 break
510 for edge in edges:
511 if not isinstance(edge, dict):
512 continue
513 if edge.get('__typename') != self._EDGE_KIND:
514 continue
515 node = edge.get('node')
516 if not isinstance(node, dict):
517 continue
518 if node.get('__typename') != self._NODE_KIND:
519 continue
520 entry = self._extract_entry(node)
521 if entry:
522 cursor = edge.get('cursor')
523 yield entry
524 if not cursor or not isinstance(cursor, compat_str):
525 break
526
c5db6bb3 527
841b6838
S
528class TwitchVideosIE(TwitchPlaylistBaseIE):
529 _VALID_URL = r'https?://(?:(?:www|go|m)\.)?twitch\.tv/(?P<id>[^/]+)/(?:videos|profile)'
c5db6bb3 530
fce79626 531 _TESTS = [{
841b6838
S
532 # All Videos sorted by Date
533 'url': 'https://www.twitch.tv/spamfish/videos?filter=all',
c5db6bb3 534 'info_dict': {
841b6838
S
535 'id': 'spamfish',
536 'title': 'spamfish - All Videos sorted by Date',
c5db6bb3 537 },
841b6838 538 'playlist_mincount': 924,
fce79626 539 }, {
841b6838
S
540 # All Videos sorted by Popular
541 'url': 'https://www.twitch.tv/spamfish/videos?filter=all&sort=views',
93753aad
S
542 'info_dict': {
543 'id': 'spamfish',
841b6838 544 'title': 'spamfish - All Videos sorted by Popular',
93753aad 545 },
841b6838 546 'playlist_mincount': 931,
fce79626 547 }, {
841b6838
S
548 # Past Broadcasts sorted by Date
549 'url': 'https://www.twitch.tv/spamfish/videos?filter=archives',
550 'info_dict': {
551 'id': 'spamfish',
552 'title': 'spamfish - Past Broadcasts sorted by Date',
553 },
554 'playlist_mincount': 27,
555 }, {
556 # Highlights sorted by Date
557 'url': 'https://www.twitch.tv/spamfish/videos?filter=highlights',
558 'info_dict': {
559 'id': 'spamfish',
560 'title': 'spamfish - Highlights sorted by Date',
561 },
562 'playlist_mincount': 901,
563 }, {
564 # Uploads sorted by Date
565 'url': 'https://www.twitch.tv/esl_csgo/videos?filter=uploads&sort=time',
566 'info_dict': {
567 'id': 'esl_csgo',
568 'title': 'esl_csgo - Uploads sorted by Date',
569 },
570 'playlist_mincount': 5,
571 }, {
572 # Past Premieres sorted by Date
573 'url': 'https://www.twitch.tv/spamfish/videos?filter=past_premieres',
93753aad
S
574 'info_dict': {
575 'id': 'spamfish',
841b6838 576 'title': 'spamfish - Past Premieres sorted by Date',
93753aad 577 },
841b6838 578 'playlist_mincount': 1,
fce79626 579 }, {
841b6838
S
580 'url': 'https://www.twitch.tv/spamfish/videos/all',
581 'only_matching': True,
582 }, {
583 'url': 'https://m.twitch.tv/spamfish/videos/all',
584 'only_matching': True,
585 }, {
586 'url': 'https://www.twitch.tv/spamfish/videos',
fce79626
S
587 'only_matching': True,
588 }]
93753aad 589
841b6838
S
590 Broadcast = collections.namedtuple('Broadcast', ['type', 'label'])
591
592 _DEFAULT_BROADCAST = Broadcast(None, 'All Videos')
593 _BROADCASTS = {
594 'archives': Broadcast('ARCHIVE', 'Past Broadcasts'),
595 'highlights': Broadcast('HIGHLIGHT', 'Highlights'),
596 'uploads': Broadcast('UPLOAD', 'Uploads'),
597 'past_premieres': Broadcast('PAST_PREMIERE', 'Past Premieres'),
598 'all': _DEFAULT_BROADCAST,
599 }
600
601 _DEFAULT_SORTED_BY = 'Date'
602 _SORTED_BY = {
603 'time': _DEFAULT_SORTED_BY,
604 'views': 'Popular',
605 }
606
841b6838
S
607 _OPERATION_NAME = 'FilterableVideoTower_Videos'
608 _ENTRY_KIND = 'video'
609 _EDGE_KIND = 'VideoEdge'
610 _NODE_KIND = 'Video'
611
612 @classmethod
613 def suitable(cls, url):
614 return (False
615 if any(ie.suitable(url) for ie in (
616 TwitchVideosClipsIE,
617 TwitchVideosCollectionsIE))
618 else super(TwitchVideosIE, cls).suitable(url))
619
620 @staticmethod
621 def _make_variables(channel_name, broadcast_type, sort):
622 return {
623 'channelOwnerLogin': channel_name,
624 'broadcastType': broadcast_type,
625 'videoSort': sort.upper(),
626 }
627
628 @staticmethod
629 def _extract_entry(node):
630 return _make_video_result(node)
93753aad 631
841b6838
S
632 def _real_extract(self, url):
633 channel_name = self._match_id(url)
4dfbf869 634 qs = parse_qs(url)
841b6838
S
635 filter = qs.get('filter', ['all'])[0]
636 sort = qs.get('sort', ['time'])[0]
637 broadcast = self._BROADCASTS.get(filter, self._DEFAULT_BROADCAST)
638 return self.playlist_result(
639 self._entries(channel_name, broadcast.type, sort),
640 playlist_id=channel_name,
641 playlist_title='%s - %s sorted by %s'
642 % (channel_name, broadcast.label,
643 self._SORTED_BY.get(sort, self._DEFAULT_SORTED_BY)))
644
645
646class TwitchVideosClipsIE(TwitchPlaylistBaseIE):
647 _VALID_URL = r'https?://(?:(?:www|go|m)\.)?twitch\.tv/(?P<id>[^/]+)/(?:clips|videos/*?\?.*?\bfilter=clips)'
c5db6bb3 648
fce79626 649 _TESTS = [{
841b6838
S
650 # Clips
651 'url': 'https://www.twitch.tv/vanillatv/clips?filter=clips&range=all',
93753aad 652 'info_dict': {
841b6838
S
653 'id': 'vanillatv',
654 'title': 'vanillatv - Clips Top All',
93753aad 655 },
841b6838 656 'playlist_mincount': 1,
fce79626 657 }, {
841b6838 658 'url': 'https://www.twitch.tv/dota2ruhub/videos?filter=clips&range=7d',
fce79626
S
659 'only_matching': True,
660 }]
93753aad 661
841b6838
S
662 Clip = collections.namedtuple('Clip', ['filter', 'label'])
663
664 _DEFAULT_CLIP = Clip('LAST_WEEK', 'Top 7D')
665 _RANGE = {
666 '24hr': Clip('LAST_DAY', 'Top 24H'),
667 '7d': _DEFAULT_CLIP,
668 '30d': Clip('LAST_MONTH', 'Top 30D'),
669 'all': Clip('ALL_TIME', 'Top All'),
670 }
671
672 # NB: values other than 20 result in skipped videos
673 _PAGE_LIMIT = 20
674
841b6838
S
675 _OPERATION_NAME = 'ClipsCards__User'
676 _ENTRY_KIND = 'clip'
677 _EDGE_KIND = 'ClipEdge'
678 _NODE_KIND = 'Clip'
679
680 @staticmethod
681 def _make_variables(channel_name, filter):
682 return {
683 'login': channel_name,
684 'criteria': {
685 'filter': filter,
686 },
687 }
688
689 @staticmethod
690 def _extract_entry(node):
691 assert isinstance(node, dict)
692 clip_url = url_or_none(node.get('url'))
693 if not clip_url:
694 return
695 return {
696 '_type': 'url_transparent',
697 'ie_key': TwitchClipsIE.ie_key(),
698 'id': node.get('id'),
699 'url': clip_url,
700 'title': node.get('title'),
701 'thumbnail': node.get('thumbnailURL'),
702 'duration': float_or_none(node.get('durationSeconds')),
703 'timestamp': unified_timestamp(node.get('createdAt')),
704 'view_count': int_or_none(node.get('viewCount')),
705 'language': node.get('language'),
706 }
707
708 def _real_extract(self, url):
709 channel_name = self._match_id(url)
4dfbf869 710 qs = parse_qs(url)
841b6838
S
711 range = qs.get('range', ['7d'])[0]
712 clip = self._RANGE.get(range, self._DEFAULT_CLIP)
713 return self.playlist_result(
714 self._entries(channel_name, clip.filter),
715 playlist_id=channel_name,
716 playlist_title='%s - Clips %s' % (channel_name, clip.label))
717
93753aad 718
841b6838
S
719class TwitchVideosCollectionsIE(TwitchPlaylistBaseIE):
720 _VALID_URL = r'https?://(?:(?:www|go|m)\.)?twitch\.tv/(?P<id>[^/]+)/videos/*?\?.*?\bfilter=collections'
93753aad 721
fce79626 722 _TESTS = [{
841b6838
S
723 # Collections
724 'url': 'https://www.twitch.tv/spamfish/videos?filter=collections',
c5db6bb3
S
725 'info_dict': {
726 'id': 'spamfish',
841b6838 727 'title': 'spamfish - Collections',
c5db6bb3 728 },
841b6838 729 'playlist_mincount': 3,
fce79626 730 }]
240b9b7a 731
841b6838
S
732 _OPERATION_NAME = 'ChannelCollectionsContent'
733 _ENTRY_KIND = 'collection'
734 _EDGE_KIND = 'CollectionsItemEdge'
735 _NODE_KIND = 'Collection'
736
737 @staticmethod
738 def _make_variables(channel_name):
739 return {
740 'ownerLogin': channel_name,
741 }
742
743 @staticmethod
744 def _extract_entry(node):
745 assert isinstance(node, dict)
746 collection_id = node.get('id')
747 if not collection_id:
748 return
749 return {
750 '_type': 'url_transparent',
751 'ie_key': TwitchCollectionIE.ie_key(),
752 'id': collection_id,
753 'url': 'https://www.twitch.tv/collections/%s' % collection_id,
754 'title': node.get('title'),
755 'thumbnail': node.get('thumbnailURL'),
756 'duration': float_or_none(node.get('lengthSeconds')),
757 'timestamp': unified_timestamp(node.get('updatedAt')),
758 'view_count': int_or_none(node.get('viewCount')),
759 }
760
761 def _real_extract(self, url):
762 channel_name = self._match_id(url)
763 return self.playlist_result(
764 self._entries(channel_name), playlist_id=channel_name,
765 playlist_title='%s - Collections' % channel_name)
766
240b9b7a 767
00dd0cd5 768class TwitchStreamIE(TwitchBaseIE):
240b9b7a 769 IE_NAME = 'twitch:stream'
9db8f6c5
S
770 _VALID_URL = r'''(?x)
771 https?://
772 (?:
fce79626 773 (?:(?:www|go|m)\.)?twitch\.tv/|
9db8f6c5
S
774 player\.twitch\.tv/\?.*?\bchannel=
775 )
776 (?P<id>[^/#?]+)
777 '''
240b9b7a 778
faa1f83a 779 _TESTS = [{
240b9b7a
S
780 'url': 'http://www.twitch.tv/shroomztv',
781 'info_dict': {
782 'id': '12772022048',
783 'display_id': 'shroomztv',
784 'ext': 'mp4',
785 'title': 're:^ShroomzTV [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
786 'description': 'H1Z1 - lonewolfing with ShroomzTV | A3 Battle Royale later - @ShroomzTV',
787 'is_live': True,
788 'timestamp': 1421928037,
789 'upload_date': '20150122',
790 'uploader': 'ShroomzTV',
791 'uploader_id': 'shroomztv',
792 'view_count': int,
793 },
794 'params': {
795 # m3u8 download
796 'skip_download': True,
797 },
faa1f83a
S
798 }, {
799 'url': 'http://www.twitch.tv/miracle_doto#profile-0',
800 'only_matching': True,
9db8f6c5
S
801 }, {
802 'url': 'https://player.twitch.tv/?channel=lotsofs',
803 'only_matching': True,
b763e1d6
S
804 }, {
805 'url': 'https://go.twitch.tv/food',
806 'only_matching': True,
fce79626
S
807 }, {
808 'url': 'https://m.twitch.tv/food',
809 'only_matching': True,
faa1f83a 810 }]
240b9b7a 811
9db8f6c5
S
812 @classmethod
813 def suitable(cls, url):
814 return (False
815 if any(ie.suitable(url) for ie in (
9db8f6c5 816 TwitchVodIE,
841b6838
S
817 TwitchCollectionIE,
818 TwitchVideosIE,
819 TwitchVideosClipsIE,
820 TwitchVideosCollectionsIE,
db348e88 821 TwitchClipsIE))
9db8f6c5
S
822 else super(TwitchStreamIE, cls).suitable(url))
823
240b9b7a 824 def _real_extract(self, url):
defc820b
S
825 channel_name = self._match_id(url).lower()
826
827 gql = self._download_gql(
828 channel_name, [{
829 'operationName': 'StreamMetadata',
830 'variables': {'channelLogin': channel_name},
831 }, {
832 'operationName': 'ComscoreStreamingQuery',
833 'variables': {
834 'channel': channel_name,
835 'clipSlug': '',
836 'isClip': False,
837 'isLive': True,
838 'isVodOrCollection': False,
839 'vodID': '',
840 },
841 }, {
842 'operationName': 'VideoPreviewOverlay',
843 'variables': {'login': channel_name},
844 }],
845 'Downloading stream GraphQL')
846
847 user = gql[0]['data']['user']
848
849 if not user:
850 raise ExtractorError(
851 '%s does not exist' % channel_name, expected=True)
240b9b7a 852
defc820b 853 stream = user['stream']
240b9b7a 854
240b9b7a 855 if not stream:
defc820b 856 raise ExtractorError('%s is offline' % channel_name, expected=True)
240b9b7a 857
00dd0cd5 858 access_token = self._download_access_token(
859 channel_name, 'stream', 'channelName')
860 token = access_token['value']
240b9b7a 861
defc820b 862 stream_id = stream.get('id') or channel_name
240b9b7a
S
863 query = {
864 'allow_source': 'true',
ac455055 865 'allow_audio_only': 'true',
9aa929d3 866 'allow_spectre': 'true',
f353cbdb 867 'p': random.randint(1000000, 10000000),
240b9b7a 868 'player': 'twitchweb',
1d31b7ca 869 'playlist_include_framerate': 'true',
240b9b7a 870 'segment_preference': '4',
00dd0cd5 871 'sig': access_token['signature'].encode('utf-8'),
ce3735df 872 'token': token.encode('utf-8'),
240b9b7a 873 }
240b9b7a 874 formats = self._extract_m3u8_formats(
defc820b
S
875 '%s/api/channel/hls/%s.m3u8' % (self._USHER_BASE, channel_name),
876 stream_id, 'mp4', query=query)
d0e958c7 877 self._prefer_source(formats)
240b9b7a
S
878
879 view_count = stream.get('viewers')
defc820b 880 timestamp = unified_timestamp(stream.get('createdAt'))
240b9b7a 881
defc820b
S
882 sq_user = try_get(gql, lambda x: x[1]['data']['user'], dict) or {}
883 uploader = sq_user.get('displayName')
884 description = try_get(
885 sq_user, lambda x: x['broadcastSettings']['title'], compat_str)
240b9b7a 886
defc820b
S
887 thumbnail = url_or_none(try_get(
888 gql, lambda x: x[2]['data']['user']['stream']['previewImageURL'],
889 compat_str))
890
891 title = uploader or channel_name
892 stream_type = stream.get('type')
893 if stream_type in ['rerun', 'live']:
894 title += ' (%s)' % stream_type
240b9b7a
S
895
896 return {
defc820b 897 'id': stream_id,
ce3735df 898 'display_id': channel_name,
39ca3b5c 899 'title': title,
240b9b7a 900 'description': description,
defc820b
S
901 'thumbnail': thumbnail,
902 'uploader': uploader,
903 'uploader_id': channel_name,
240b9b7a
S
904 'timestamp': timestamp,
905 'view_count': view_count,
906 'formats': formats,
defc820b 907 'is_live': stream_type == 'live',
12d1fb5a 908 }
778f9694
S
909
910
49fa7de3 911class TwitchClipsIE(TwitchBaseIE):
778f9694 912 IE_NAME = 'twitch:clips'
c9595ee7
FS
913 _VALID_URL = r'''(?x)
914 https?://
915 (?:
916 clips\.twitch\.tv/(?:embed\?.*?\bclip=|(?:[^/]+/)*)|
917 (?:(?:www|go|m)\.)?twitch\.tv/[^/]+/clip/
918 )
919 (?P<id>[^/?#&]+)
920 '''
778f9694 921
74ba450a 922 _TESTS = [{
49fa7de3 923 'url': 'https://clips.twitch.tv/FaintLightGullWholeWheat',
778f9694
S
924 'md5': '761769e1eafce0ffebfb4089cb3847cd',
925 'info_dict': {
49fa7de3 926 'id': '42850523',
678da2f2 927 'display_id': 'FaintLightGullWholeWheat',
778f9694
S
928 'ext': 'mp4',
929 'title': 'EA Play 2016 Live from the Novo Theatre',
ec85ded8 930 'thumbnail': r're:^https?://.*\.jpg',
49fa7de3
S
931 'timestamp': 1465767393,
932 'upload_date': '20160612',
778f9694
S
933 'creator': 'EA',
934 'uploader': 'stereotype_',
49fa7de3 935 'uploader_id': '43566419',
778f9694 936 },
74ba450a
S
937 }, {
938 # multiple formats
939 'url': 'https://clips.twitch.tv/rflegendary/UninterestedBeeDAESuppy',
940 'only_matching': True,
db348e88
S
941 }, {
942 'url': 'https://www.twitch.tv/sergeynixon/clip/StormyThankfulSproutFutureMan',
943 'only_matching': True,
0b16b3c2
RA
944 }, {
945 'url': 'https://clips.twitch.tv/embed?clip=InquisitiveBreakableYogurtJebaited',
946 'only_matching': True,
c9595ee7
FS
947 }, {
948 'url': 'https://m.twitch.tv/rossbroadcast/clip/ConfidentBraveHumanChefFrank',
949 'only_matching': True,
950 }, {
951 'url': 'https://go.twitch.tv/rossbroadcast/clip/ConfidentBraveHumanChefFrank',
952 'only_matching': True,
74ba450a 953 }]
778f9694
S
954
955 def _real_extract(self, url):
956 video_id = self._match_id(url)
957
14eb1ee1 958 clip = self._download_gql(
959 video_id, [{
960 'operationName': 'VideoAccessToken_Clip',
961 'variables': {
962 'slug': video_id,
963 },
964 }],
965 'Downloading clip access token GraphQL')[0]['data']['clip']
966
967 if not clip:
968 raise ExtractorError(
969 'This clip is no longer available', expected=True)
970
971 access_query = {
972 'sig': clip['playbackAccessToken']['signature'],
973 'token': clip['playbackAccessToken']['value'],
974 }
975
976 data = self._download_base_gql(
00dd0cd5 977 video_id, {
232ed8e6
RA
978 'query': '''{
979 clip(slug: "%s") {
980 broadcaster {
981 displayName
982 }
983 createdAt
984 curator {
985 displayName
986 id
987 }
988 durationSeconds
989 id
990 tiny: thumbnailURL(width: 86, height: 45)
991 small: thumbnailURL(width: 260, height: 147)
992 medium: thumbnailURL(width: 480, height: 272)
993 title
994 videoQualities {
995 frameRate
996 quality
997 sourceURL
998 }
999 viewCount
1000 }
14eb1ee1 1001}''' % video_id}, 'Downloading clip GraphQL', fatal=False)
232ed8e6 1002
14eb1ee1 1003 if data:
1004 clip = try_get(data, lambda x: x['data']['clip'], dict) or clip
778f9694 1005
49fa7de3 1006 formats = []
232ed8e6 1007 for option in clip.get('videoQualities', []):
49fa7de3
S
1008 if not isinstance(option, dict):
1009 continue
232ed8e6 1010 source = url_or_none(option.get('sourceURL'))
3052a30d 1011 if not source:
49fa7de3
S
1012 continue
1013 formats.append({
14eb1ee1 1014 'url': update_url_query(source, access_query),
49fa7de3
S
1015 'format_id': option.get('quality'),
1016 'height': int_or_none(option.get('quality')),
232ed8e6 1017 'fps': int_or_none(option.get('frameRate')),
49fa7de3 1018 })
07acdc5a
S
1019 self._sort_formats(formats)
1020
232ed8e6
RA
1021 thumbnails = []
1022 for thumbnail_id in ('tiny', 'small', 'medium'):
1023 thumbnail_url = clip.get(thumbnail_id)
1024 if not thumbnail_url:
1025 continue
1026 thumb = {
1027 'id': thumbnail_id,
1028 'url': thumbnail_url,
1029 }
1030 mobj = re.search(r'-(\d+)x(\d+)\.', thumbnail_url)
1031 if mobj:
1032 thumb.update({
1033 'height': int(mobj.group(2)),
1034 'width': int(mobj.group(1)),
1035 })
1036 thumbnails.append(thumb)
1037
1038 return {
1039 'id': clip.get('id') or video_id,
678da2f2 1040 'display_id': video_id,
232ed8e6 1041 'title': clip.get('title') or video_id,
49fa7de3 1042 'formats': formats,
232ed8e6 1043 'duration': int_or_none(clip.get('durationSeconds')),
be8cd3cb 1044 'view_count': int_or_none(clip.get('viewCount')),
232ed8e6
RA
1045 'timestamp': unified_timestamp(clip.get('createdAt')),
1046 'thumbnails': thumbnails,
1047 'creator': try_get(clip, lambda x: x['broadcaster']['displayName'], compat_str),
1048 'uploader': try_get(clip, lambda x: x['curator']['displayName'], compat_str),
1049 'uploader_id': try_get(clip, lambda x: x['curator']['id'], compat_str),
49fa7de3 1050 }