twitch.py 33 KB
Newer Older
1
# coding: utf-8
2
3
from __future__ import unicode_literals

4
import collections
5
import itertools
6
import json
7
8
import random
import re
9
10

from .common import InfoExtractor
11
from ..compat import (
12
    compat_kwargs,
13
    compat_parse_qs,
14
    compat_str,
15
    compat_urlparse,
16
    compat_urllib_parse_urlencode,
17
    compat_urllib_parse_urlparse,
18
19
)
from ..utils import (
Sergey M․'s avatar
Sergey M․ committed
20
    clean_html,
21
    ExtractorError,
22
    float_or_none,
Sergey M․'s avatar
Sergey M․ committed
23
    int_or_none,
24
    parse_duration,
25
    parse_iso8601,
26
    qualities,
27
    str_or_none,
28
29
    try_get,
    unified_timestamp,
30
    update_url_query,
Sergey M․'s avatar
Sergey M․ committed
31
    url_or_none,
32
    urljoin,
33
34
35
)


36
class TwitchBaseIE(InfoExtractor):
37
    _VALID_URL_BASE = r'https?://(?:(?:www|go|m)\.)?twitch\.tv'
38

39
    _API_BASE = 'https://api.twitch.tv'
40
    _USHER_BASE = 'https://usher.ttvnw.net'
41
42
    _LOGIN_FORM_URL = 'https://www.twitch.tv/login'
    _LOGIN_POST_URL = 'https://passport.twitch.tv/login'
43
    _CLIENT_ID = 'kimne78kx3ncx6brgo4mv6wki5h1ko'
44
    _NETRC_MACHINE = 'twitch'
45

46
47
48
49
50
51
52
53
54
    def _handle_error(self, response):
        if not isinstance(response, dict):
            return
        error = response.get('error')
        if error:
            raise ExtractorError(
                '%s returned error: %s - %s' % (self.IE_NAME, error, response.get('message')),
                expected=True)

55
    def _call_api(self, path, item_id, *args, **kwargs):
56
        headers = kwargs.get('headers', {}).copy()
57
58
59
60
        headers.update({
            'Accept': 'application/vnd.twitchtv.v5+json; charset=UTF-8',
            'Client-ID': self._CLIENT_ID,
        })
61
62
63
64
        kwargs.update({
            'headers': headers,
            'expected_status': (400, 410),
        })
Sergey M․'s avatar
Sergey M․ committed
65
        response = self._download_json(
66
67
            '%s/%s' % (self._API_BASE, path), item_id,
            *args, **compat_kwargs(kwargs))
68
69
70
        self._handle_error(response)
        return response

71
72
73
74
    def _real_initialize(self):
        self._login()

    def _login(self):
75
        username, password = self._get_login_info()
76
77
78
        if username is None:
            return

Sergey M․'s avatar
Sergey M․ committed
79
80
81
82
        def fail(message):
            raise ExtractorError(
                'Unable to login. Twitch said: %s' % message, expected=True)

83
        def login_step(page, urlh, note, data):
84
85
86
87
88
89
            form = self._hidden_inputs(page)
            form.update(data)

            page_url = urlh.geturl()
            post_url = self._search_regex(
                r'<form[^>]+action=(["\'])(?P<url>.+?)\1', page,
90
                'post url', default=self._LOGIN_POST_URL, group='url')
91
92
            post_url = urljoin(page_url, post_url)

93
94
95
            headers = {
                'Referer': page_url,
                'Origin': page_url,
96
                'Content-Type': 'text/plain;charset=UTF-8',
97
            }
98

99
100
101
102
103
104
            response = self._download_json(
                post_url, None, note, data=json.dumps(form).encode(),
                headers=headers, expected_status=400)
            error = response.get('error_description') or response.get('error_code')
            if error:
                fail(error)
105

106
107
108
109
110
111
            if 'Authenticated successfully' in response.get('message', ''):
                return None, None

            redirect_url = urljoin(
                post_url,
                response.get('redirect') or response['redirect_path'])
112
113
114
            return self._download_webpage_handle(
                redirect_url, None, 'Downloading login redirect page',
                headers=headers)
115

Sergey M․'s avatar
Sergey M․ committed
116
        login_page, handle = self._download_webpage_handle(
117
            self._LOGIN_FORM_URL, None, 'Downloading login page')
118

Sergey M․'s avatar
Sergey M․ committed
119
120
121
122
        # Some TOR nodes and public proxies are blocked completely
        if 'blacklist_message' in login_page:
            fail(clean_html(login_page))

123
        redirect_page, handle = login_step(
124
            login_page, handle, 'Logging in', {
125
126
                'username': username,
                'password': password,
127
                'client_id': self._CLIENT_ID,
Yen Chi Hsuan's avatar
Yen Chi Hsuan committed
128
            })
129

130
131
132
133
        # Successful login
        if not redirect_page:
            return

134
135
136
        if re.search(r'(?i)<form[^>]+id="two-factor-submit"', redirect_page) is not None:
            # TODO: Add mechanism to request an SMS or phone call
            tfa_token = self._get_tfa_info('two-factor authentication token')
137
            login_step(redirect_page, handle, 'Submitting TFA token', {
138
139
                'authy_token': tfa_token,
                'remember_2fa': 'true',
140
            })
141

142
143
144
    def _prefer_source(self, formats):
        try:
            source = next(f for f in formats if f['format_id'] == 'Source')
145
            source['quality'] = 10
146
        except StopIteration:
147
148
149
            for f in formats:
                if '/chunked/' in f['url']:
                    f.update({
150
                        'quality': 10,
151
152
                        'format_note': 'Source',
                    })
153
154
        self._sort_formats(formats)

155
156
157
158
    def _download_access_token(self, channel_name):
        return self._call_api(
            'api/channels/%s/access_token' % channel_name, channel_name,
            'Downloading access token JSON')
159

160
161
    def _extract_channel_id(self, token, channel_name):
        return compat_str(self._parse_json(token, channel_name)['channel_id'])
162
163


164
class TwitchVodIE(TwitchBaseIE):
165
    IE_NAME = 'twitch:vod'
166
167
168
    _VALID_URL = r'''(?x)
                    https?://
                        (?:
169
                            (?:(?:www|go|m)\.)?twitch\.tv/(?:[^/]+/v(?:ideo)?|videos)/|
170
                            player\.twitch\.tv/\?.*?\bvideo=v?
171
172
173
                        )
                        (?P<id>\d+)
                    '''
174
175
176
    _ITEM_TYPE = 'vod'
    _ITEM_SHORTCUT = 'v'

Sergey M․'s avatar
Sergey M․ committed
177
    _TESTS = [{
178
        'url': 'http://www.twitch.tv/riotgames/v/6528877?t=5m10s',
179
        'info_dict': {
Yen Chi Hsuan's avatar
Yen Chi Hsuan committed
180
            'id': 'v6528877',
181
            'ext': 'mp4',
Yen Chi Hsuan's avatar
Yen Chi Hsuan committed
182
            'title': 'LCK Summer Split - Week 6 Day 1',
183
            'thumbnail': r're:^https?://.*\.jpg$',
Yen Chi Hsuan's avatar
Yen Chi Hsuan committed
184
185
186
187
188
            'duration': 17208,
            'timestamp': 1435131709,
            'upload_date': '20150624',
            'uploader': 'Riot Games',
            'uploader_id': 'riotgames',
189
            'view_count': int,
190
            'start_time': 310,
191
192
193
194
195
        },
        'params': {
            # m3u8 download
            'skip_download': True,
        },
Sergey M․'s avatar
Sergey M․ committed
196
197
198
199
200
201
202
    }, {
        # Untitled broadcast (title is None)
        'url': 'http://www.twitch.tv/belkao_o/v/11230755',
        'info_dict': {
            'id': 'v11230755',
            'ext': 'mp4',
            'title': 'Untitled Broadcast',
203
            'thumbnail': r're:^https?://.*\.jpg$',
Sergey M․'s avatar
Sergey M․ committed
204
205
206
207
208
209
210
211
212
213
214
            'duration': 1638,
            'timestamp': 1439746708,
            'upload_date': '20150816',
            'uploader': 'BelkAO_o',
            'uploader_id': 'belkao_o',
            'view_count': int,
        },
        'params': {
            # m3u8 download
            'skip_download': True,
        },
Yen Chi Hsuan's avatar
Yen Chi Hsuan committed
215
        'skip': 'HTTP Error 404: Not Found',
216
217
218
    }, {
        'url': 'http://player.twitch.tv/?t=5m10s&video=v6528877',
        'only_matching': True,
219
220
221
    }, {
        'url': 'https://www.twitch.tv/videos/6528877',
        'only_matching': True,
222
223
224
    }, {
        'url': 'https://m.twitch.tv/beagsandjam/v/247478721',
        'only_matching': True,
225
226
227
    }, {
        'url': 'https://www.twitch.tv/northernlion/video/291940395',
        'only_matching': True,
228
229
230
    }, {
        'url': 'https://player.twitch.tv/?video=480452374',
        'only_matching': True,
Sergey M․'s avatar
Sergey M․ committed
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
    def _download_info(self, item_id):
        return self._extract_info(
            self._call_api(
                'kraken/videos/%s' % item_id, item_id,
                'Downloading video info JSON'))

    @staticmethod
    def _extract_info(info):
        status = info.get('status')
        if status == 'recording':
            is_live = True
        elif status == 'recorded':
            is_live = False
        else:
            is_live = None
        _QUALITIES = ('small', 'medium', 'large')
        quality_key = qualities(_QUALITIES)
        thumbnails = []
        preview = info.get('preview')
        if isinstance(preview, dict):
            for thumbnail_id, thumbnail_url in preview.items():
                thumbnail_url = url_or_none(thumbnail_url)
                if not thumbnail_url:
                    continue
                if thumbnail_id not in _QUALITIES:
                    continue
                thumbnails.append({
                    'url': thumbnail_url,
                    'preference': quality_key(thumbnail_id),
                })
        return {
            'id': info['_id'],
            'title': info.get('title') or 'Untitled Broadcast',
            'description': info.get('description'),
            'duration': int_or_none(info.get('length')),
            'thumbnails': thumbnails,
            'uploader': info.get('channel', {}).get('display_name'),
            'uploader_id': info.get('channel', {}).get('name'),
            'timestamp': parse_iso8601(info.get('recorded_at')),
            'view_count': int_or_none(info.get('views')),
            'is_live': is_live,
        }

276
    def _real_extract(self, url):
277
        vod_id = self._match_id(url)
Sergey M․'s avatar
Sergey M․ committed
278

279
        info = self._download_info(vod_id)
Sergey M․'s avatar
Sergey M․ committed
280
        access_token = self._call_api(
281
            'api/vods/%s/access_token' % vod_id, vod_id,
Sergey M․'s avatar
Sergey M․ committed
282
            'Downloading %s access token' % self._ITEM_TYPE)
Sergey M․'s avatar
Sergey M․ committed
283

284
        formats = self._extract_m3u8_formats(
285
            '%s/vod/%s.m3u8?%s' % (
286
                self._USHER_BASE, vod_id,
287
                compat_urllib_parse_urlencode({
Sergey M․'s avatar
Sergey M․ committed
288
                    'allow_source': 'true',
289
                    'allow_audio_only': 'true',
Sergey M․'s avatar
Sergey M․ committed
290
291
                    'allow_spectre': 'true',
                    'player': 'twitchweb',
292
                    'playlist_include_framerate': 'true',
Sergey M․'s avatar
Sergey M․ committed
293
294
295
                    'nauth': access_token['token'],
                    'nauthsig': access_token['sig'],
                })),
296
            vod_id, 'mp4', entry_protocol='m3u8_native')
Sergey M․'s avatar
Sergey M․ committed
297

298
        self._prefer_source(formats)
299
        info['formats'] = formats
300
301
302
303
304
305

        parsed_url = compat_urllib_parse_urlparse(url)
        query = compat_parse_qs(parsed_url.query)
        if 't' in query:
            info['start_time'] = parse_duration(query['t'][0])

306
307
308
309
        if info.get('timestamp') is not None:
            info['subtitles'] = {
                'rechat': [{
                    'url': update_url_query(
310
                        'https://api.twitch.tv/v5/videos/%s/comments' % vod_id, {
311
                            'client_id': self._CLIENT_ID,
312
313
314
315
316
                        }),
                    'ext': 'json',
                }],
            }

317
318
319
        return info


320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
def _make_video_result(node):
    assert isinstance(node, dict)
    video_id = node.get('id')
    if not video_id:
        return
    return {
        '_type': 'url_transparent',
        'ie_key': TwitchVodIE.ie_key(),
        'id': video_id,
        'url': 'https://www.twitch.tv/videos/%s' % video_id,
        'title': node.get('title'),
        'thumbnail': node.get('previewThumbnailURL'),
        'duration': float_or_none(node.get('lengthSeconds')),
        'view_count': int_or_none(node.get('viewCount')),
    }


class TwitchGraphQLBaseIE(TwitchBaseIE):
338
    _PAGE_LIMIT = 100
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
    def _download_gql(self, video_id, op, variables, sha256_hash, note, fatal=True):
        return self._download_json(
            'https://gql.twitch.tv/gql', video_id, note,
            data=json.dumps({
                'operationName': op,
                'variables': variables,
                'extensions': {
                    'persistedQuery': {
                        'version': 1,
                        'sha256Hash': sha256_hash,
                    }
                }
            }).encode(),
            headers={
                'Content-Type': 'text/plain;charset=UTF-8',
                'Client-ID': self._CLIENT_ID,
            }, fatal=fatal)


class TwitchCollectionIE(TwitchGraphQLBaseIE):
    _VALID_URL = r'https?://(?:(?:www|go|m)\.)?twitch\.tv/collections/(?P<id>[^/]+)'

    _TESTS = [{
        'url': 'https://www.twitch.tv/collections/wlDCoH0zEBZZbQ',
        'info_dict': {
            'id': 'wlDCoH0zEBZZbQ',
            'title': 'Overthrow Nook, capitalism for children',
        },
        'playlist_mincount': 13,
    }]

    _OPERATION_NAME = 'CollectionSideBar'
    _SHA256_HASH = '27111f1b382effad0b6def325caef1909c733fe6a4fbabf54f8d491ef2cf2f14'

    def _real_extract(self, url):
        collection_id = self._match_id(url)
        collection = self._download_gql(
            collection_id, self._OPERATION_NAME,
            {'collectionID': collection_id}, self._SHA256_HASH,
            'Downloading collection GraphQL')['data']['collection']
        title = collection.get('title')
381
        entries = []
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
        for edge in collection['items']['edges']:
            if not isinstance(edge, dict):
                continue
            node = edge.get('node')
            if not isinstance(node, dict):
                continue
            video = _make_video_result(node)
            if video:
                entries.append(video)
        return self.playlist_result(
            entries, playlist_id=collection_id, playlist_title=title)


class TwitchPlaylistBaseIE(TwitchGraphQLBaseIE):
    def _entries(self, channel_name, *args):
        cursor = None
        variables_common = self._make_variables(channel_name, *args)
        entries_key = '%ss' % self._ENTRY_KIND
        for page_num in itertools.count(1):
            variables = variables_common.copy()
            variables['limit'] = self._PAGE_LIMIT
            if cursor:
                variables['cursor'] = cursor
            page = self._download_gql(
                channel_name, self._OPERATION_NAME, variables,
                self._SHA256_HASH,
                'Downloading %ss GraphQL page %s' % (self._NODE_KIND, page_num),
                fatal=False)
            if not page:
                break
            edges = try_get(
                page, lambda x: x['data']['user'][entries_key]['edges'], list)
            if not edges:
                break
            for edge in edges:
                if not isinstance(edge, dict):
                    continue
                if edge.get('__typename') != self._EDGE_KIND:
                    continue
                node = edge.get('node')
                if not isinstance(node, dict):
                    continue
                if node.get('__typename') != self._NODE_KIND:
                    continue
                entry = self._extract_entry(node)
                if entry:
                    cursor = edge.get('cursor')
                    yield entry
            if not cursor or not isinstance(cursor, compat_str):
                break

    # Deprecated kraken v5 API
    def _entries_kraken(self, channel_name, broadcast_type, sort):
        access_token = self._download_access_token(channel_name)
        channel_id = self._extract_channel_id(access_token['token'], channel_name)
437
        offset = 0
438
        counter_override = None
439
        for counter in itertools.count(1):
Sergey M․'s avatar
Sergey M․ committed
440
            response = self._call_api(
441
                'kraken/channels/%s/videos/' % channel_id,
442
                channel_id,
443
444
445
446
447
448
449
450
451
                'Downloading video JSON page %s' % (counter_override or counter),
                query={
                    'offset': offset,
                    'limit': self._PAGE_LIMIT,
                    'broadcast_type': broadcast_type,
                    'sort': sort,
                })
            videos = response.get('videos')
            if not isinstance(videos, list):
452
                break
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
            for video in videos:
                if not isinstance(video, dict):
                    continue
                video_url = url_or_none(video.get('url'))
                if not video_url:
                    continue
                yield {
                    '_type': 'url_transparent',
                    'ie_key': TwitchVodIE.ie_key(),
                    'id': video.get('_id'),
                    'url': video_url,
                    'title': video.get('title'),
                    'description': video.get('description'),
                    'timestamp': unified_timestamp(video.get('published_at')),
                    'duration': float_or_none(video.get('length')),
                    'view_count': int_or_none(video.get('views')),
                    'language': video.get('language'),
                }
            offset += self._PAGE_LIMIT
472
            total = int_or_none(response.get('_total'))
473
            if total and offset >= total:
474
                break
475
476


477
478
class TwitchVideosIE(TwitchPlaylistBaseIE):
    _VALID_URL = r'https?://(?:(?:www|go|m)\.)?twitch\.tv/(?P<id>[^/]+)/(?:videos|profile)'
479

480
    _TESTS = [{
481
482
        # All Videos sorted by Date
        'url': 'https://www.twitch.tv/spamfish/videos?filter=all',
483
        'info_dict': {
484
485
            'id': 'spamfish',
            'title': 'spamfish - All Videos sorted by Date',
486
        },
487
        'playlist_mincount': 924,
488
    }, {
489
490
        # All Videos sorted by Popular
        'url': 'https://www.twitch.tv/spamfish/videos?filter=all&sort=views',
491
492
        'info_dict': {
            'id': 'spamfish',
493
            'title': 'spamfish - All Videos sorted by Popular',
494
        },
495
        'playlist_mincount': 931,
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
        # Past Broadcasts sorted by Date
        'url': 'https://www.twitch.tv/spamfish/videos?filter=archives',
        'info_dict': {
            'id': 'spamfish',
            'title': 'spamfish - Past Broadcasts sorted by Date',
        },
        'playlist_mincount': 27,
    }, {
        # Highlights sorted by Date
        'url': 'https://www.twitch.tv/spamfish/videos?filter=highlights',
        'info_dict': {
            'id': 'spamfish',
            'title': 'spamfish - Highlights sorted by Date',
        },
        'playlist_mincount': 901,
    }, {
        # Uploads sorted by Date
        'url': 'https://www.twitch.tv/esl_csgo/videos?filter=uploads&sort=time',
        'info_dict': {
            'id': 'esl_csgo',
            'title': 'esl_csgo - Uploads sorted by Date',
        },
        'playlist_mincount': 5,
    }, {
        # Past Premieres sorted by Date
        'url': 'https://www.twitch.tv/spamfish/videos?filter=past_premieres',
523
524
        'info_dict': {
            'id': 'spamfish',
525
            'title': 'spamfish - Past Premieres sorted by Date',
526
        },
527
        'playlist_mincount': 1,
528
    }, {
529
530
531
532
533
534
535
        'url': 'https://www.twitch.tv/spamfish/videos/all',
        'only_matching': True,
    }, {
        'url': 'https://m.twitch.tv/spamfish/videos/all',
        'only_matching': True,
    }, {
        'url': 'https://www.twitch.tv/spamfish/videos',
536
537
        'only_matching': True,
    }]
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
    Broadcast = collections.namedtuple('Broadcast', ['type', 'label'])

    _DEFAULT_BROADCAST = Broadcast(None, 'All Videos')
    _BROADCASTS = {
        'archives': Broadcast('ARCHIVE', 'Past Broadcasts'),
        'highlights': Broadcast('HIGHLIGHT', 'Highlights'),
        'uploads': Broadcast('UPLOAD', 'Uploads'),
        'past_premieres': Broadcast('PAST_PREMIERE', 'Past Premieres'),
        'all': _DEFAULT_BROADCAST,
    }

    _DEFAULT_SORTED_BY = 'Date'
    _SORTED_BY = {
        'time': _DEFAULT_SORTED_BY,
        'views': 'Popular',
    }

    _SHA256_HASH = 'a937f1d22e269e39a03b509f65a7490f9fc247d7f83d6ac1421523e3b68042cb'
    _OPERATION_NAME = 'FilterableVideoTower_Videos'
    _ENTRY_KIND = 'video'
    _EDGE_KIND = 'VideoEdge'
    _NODE_KIND = 'Video'

    @classmethod
    def suitable(cls, url):
        return (False
                if any(ie.suitable(url) for ie in (
                    TwitchVideosClipsIE,
                    TwitchVideosCollectionsIE))
                else super(TwitchVideosIE, cls).suitable(url))

    @staticmethod
    def _make_variables(channel_name, broadcast_type, sort):
        return {
            'channelOwnerLogin': channel_name,
            'broadcastType': broadcast_type,
            'videoSort': sort.upper(),
        }

    @staticmethod
    def _extract_entry(node):
        return _make_video_result(node)
581

582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
    def _real_extract(self, url):
        channel_name = self._match_id(url)
        qs = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
        filter = qs.get('filter', ['all'])[0]
        sort = qs.get('sort', ['time'])[0]
        broadcast = self._BROADCASTS.get(filter, self._DEFAULT_BROADCAST)
        return self.playlist_result(
            self._entries(channel_name, broadcast.type, sort),
            playlist_id=channel_name,
            playlist_title='%s - %s sorted by %s'
            % (channel_name, broadcast.label,
               self._SORTED_BY.get(sort, self._DEFAULT_SORTED_BY)))


class TwitchVideosClipsIE(TwitchPlaylistBaseIE):
    _VALID_URL = r'https?://(?:(?:www|go|m)\.)?twitch\.tv/(?P<id>[^/]+)/(?:clips|videos/*?\?.*?\bfilter=clips)'
598

599
    _TESTS = [{
600
601
        # Clips
        'url': 'https://www.twitch.tv/vanillatv/clips?filter=clips&range=all',
602
        'info_dict': {
603
604
            'id': 'vanillatv',
            'title': 'vanillatv - Clips Top All',
605
        },
606
        'playlist_mincount': 1,
607
    }, {
608
        'url': 'https://www.twitch.tv/dota2ruhub/videos?filter=clips&range=7d',
609
610
        'only_matching': True,
    }]
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
    Clip = collections.namedtuple('Clip', ['filter', 'label'])

    _DEFAULT_CLIP = Clip('LAST_WEEK', 'Top 7D')
    _RANGE = {
        '24hr': Clip('LAST_DAY', 'Top 24H'),
        '7d': _DEFAULT_CLIP,
        '30d': Clip('LAST_MONTH', 'Top 30D'),
        'all': Clip('ALL_TIME', 'Top All'),
    }

    # NB: values other than 20 result in skipped videos
    _PAGE_LIMIT = 20

    _SHA256_HASH = 'b73ad2bfaecfd30a9e6c28fada15bd97032c83ec77a0440766a56fe0bd632777'
    _OPERATION_NAME = 'ClipsCards__User'
    _ENTRY_KIND = 'clip'
    _EDGE_KIND = 'ClipEdge'
    _NODE_KIND = 'Clip'

    @staticmethod
    def _make_variables(channel_name, filter):
        return {
            'login': channel_name,
            'criteria': {
                'filter': filter,
            },
        }

    @staticmethod
    def _extract_entry(node):
        assert isinstance(node, dict)
        clip_url = url_or_none(node.get('url'))
        if not clip_url:
            return
        return {
            '_type': 'url_transparent',
            'ie_key': TwitchClipsIE.ie_key(),
            'id': node.get('id'),
            'url': clip_url,
            'title': node.get('title'),
            'thumbnail': node.get('thumbnailURL'),
            'duration': float_or_none(node.get('durationSeconds')),
            'timestamp': unified_timestamp(node.get('createdAt')),
            'view_count': int_or_none(node.get('viewCount')),
            'language': node.get('language'),
        }

    def _real_extract(self, url):
        channel_name = self._match_id(url)
        qs = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
        range = qs.get('range', ['7d'])[0]
        clip = self._RANGE.get(range, self._DEFAULT_CLIP)
        return self.playlist_result(
            self._entries(channel_name, clip.filter),
            playlist_id=channel_name,
            playlist_title='%s - Clips %s' % (channel_name, clip.label))

669

670
671
class TwitchVideosCollectionsIE(TwitchPlaylistBaseIE):
    _VALID_URL = r'https?://(?:(?:www|go|m)\.)?twitch\.tv/(?P<id>[^/]+)/videos/*?\?.*?\bfilter=collections'
672

673
    _TESTS = [{
674
675
        # Collections
        'url': 'https://www.twitch.tv/spamfish/videos?filter=collections',
676
677
        'info_dict': {
            'id': 'spamfish',
678
            'title': 'spamfish - Collections',
679
        },
680
        'playlist_mincount': 3,
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
    _SHA256_HASH = '07e3691a1bad77a36aba590c351180439a40baefc1c275356f40fc7082419a84'
    _OPERATION_NAME = 'ChannelCollectionsContent'
    _ENTRY_KIND = 'collection'
    _EDGE_KIND = 'CollectionsItemEdge'
    _NODE_KIND = 'Collection'

    @staticmethod
    def _make_variables(channel_name):
        return {
            'ownerLogin': channel_name,
        }

    @staticmethod
    def _extract_entry(node):
        assert isinstance(node, dict)
        collection_id = node.get('id')
        if not collection_id:
            return
        return {
            '_type': 'url_transparent',
            'ie_key': TwitchCollectionIE.ie_key(),
            'id': collection_id,
            'url': 'https://www.twitch.tv/collections/%s' % collection_id,
            'title': node.get('title'),
            'thumbnail': node.get('thumbnailURL'),
            'duration': float_or_none(node.get('lengthSeconds')),
            'timestamp': unified_timestamp(node.get('updatedAt')),
            'view_count': int_or_none(node.get('viewCount')),
        }

    def _real_extract(self, url):
        channel_name = self._match_id(url)
        return self.playlist_result(
            self._entries(channel_name), playlist_id=channel_name,
            playlist_title='%s - Collections' % channel_name)

719
720
721

class TwitchStreamIE(TwitchBaseIE):
    IE_NAME = 'twitch:stream'
722
723
724
    _VALID_URL = r'''(?x)
                    https?://
                        (?:
725
                            (?:(?:www|go|m)\.)?twitch\.tv/|
726
727
728
729
                            player\.twitch\.tv/\?.*?\bchannel=
                        )
                        (?P<id>[^/#?]+)
                    '''
730

731
    _TESTS = [{
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
        'url': 'http://www.twitch.tv/shroomztv',
        'info_dict': {
            'id': '12772022048',
            'display_id': 'shroomztv',
            'ext': 'mp4',
            'title': 're:^ShroomzTV [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
            'description': 'H1Z1 - lonewolfing with ShroomzTV | A3 Battle Royale later - @ShroomzTV',
            'is_live': True,
            'timestamp': 1421928037,
            'upload_date': '20150122',
            'uploader': 'ShroomzTV',
            'uploader_id': 'shroomztv',
            'view_count': int,
        },
        'params': {
            # m3u8 download
            'skip_download': True,
        },
750
751
752
    }, {
        'url': 'http://www.twitch.tv/miracle_doto#profile-0',
        'only_matching': True,
753
754
755
    }, {
        'url': 'https://player.twitch.tv/?channel=lotsofs',
        'only_matching': True,
756
757
758
    }, {
        'url': 'https://go.twitch.tv/food',
        'only_matching': True,
759
760
761
    }, {
        'url': 'https://m.twitch.tv/food',
        'only_matching': True,
762
    }]
763

764
765
766
767
768
    @classmethod
    def suitable(cls, url):
        return (False
                if any(ie.suitable(url) for ie in (
                    TwitchVodIE,
769
770
771
772
                    TwitchCollectionIE,
                    TwitchVideosIE,
                    TwitchVideosClipsIE,
                    TwitchVideosCollectionsIE,
773
                    TwitchClipsIE))
774
775
                else super(TwitchStreamIE, cls).suitable(url))

776
    def _real_extract(self, url):
777
778
        channel_name = self._match_id(url)

779
        access_token = self._download_access_token(channel_name)
780
781

        token = access_token['token']
782
        channel_id = self._extract_channel_id(token, channel_name)
783

Sergey M․'s avatar
Sergey M․ committed
784
        stream = self._call_api(
785
            'kraken/streams/%s?stream_type=all' % channel_id,
786
            channel_id, 'Downloading stream JSON').get('stream')
787
788

        if not stream:
789
            raise ExtractorError('%s is offline' % channel_id, expected=True)
790

791
792
793
794
        # Channel name may be typed if different case than the original channel name
        # (e.g. http://www.twitch.tv/TWITCHPLAYSPOKEMON) that will lead to constructing
        # an invalid m3u8 URL. Working around by use of original channel name from stream
        # JSON and fallback to lowercase if it's not available.
795
796
797
        channel_name = try_get(
            stream, lambda x: x['channel']['name'],
            compat_str) or channel_name.lower()
798
799
800

        query = {
            'allow_source': 'true',
801
            'allow_audio_only': 'true',
802
            'allow_spectre': 'true',
Sergey M․'s avatar
Sergey M․ committed
803
            'p': random.randint(1000000, 10000000),
804
            'player': 'twitchweb',
805
            'playlist_include_framerate': 'true',
806
            'segment_preference': '4',
807
            'sig': access_token['sig'].encode('utf-8'),
808
            'token': token.encode('utf-8'),
809
810
811
        }
        formats = self._extract_m3u8_formats(
            '%s/api/channel/hls/%s.m3u8?%s'
812
            % (self._USHER_BASE, channel_name, compat_urllib_parse_urlencode(query)),
813
            channel_id, 'mp4')
814
        self._prefer_source(formats)
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834

        view_count = stream.get('viewers')
        timestamp = parse_iso8601(stream.get('created_at'))

        channel = stream['channel']
        title = self._live_title(channel.get('display_name') or channel.get('name'))
        description = channel.get('status')

        thumbnails = []
        for thumbnail_key, thumbnail_url in stream['preview'].items():
            m = re.search(r'(?P<width>\d+)x(?P<height>\d+)\.jpg$', thumbnail_key)
            if not m:
                continue
            thumbnails.append({
                'url': thumbnail_url,
                'width': int(m.group('width')),
                'height': int(m.group('height')),
            })

        return {
835
836
            'id': str_or_none(stream.get('_id')) or channel_id,
            'display_id': channel_name,
837
838
839
840
841
842
843
844
845
            'title': title,
            'description': description,
            'thumbnails': thumbnails,
            'uploader': channel.get('display_name'),
            'uploader_id': channel.get('name'),
            'timestamp': timestamp,
            'view_count': view_count,
            'formats': formats,
            'is_live': True,
Jaime Marquínez Ferrándiz's avatar
Jaime Marquínez Ferrándiz committed
846
        }
847
848


849
class TwitchClipsIE(TwitchBaseIE):
850
    IE_NAME = 'twitch:clips'
851
852
853
854
855
856
857
858
    _VALID_URL = r'''(?x)
                    https?://
                        (?:
                            clips\.twitch\.tv/(?:embed\?.*?\bclip=|(?:[^/]+/)*)|
                            (?:(?:www|go|m)\.)?twitch\.tv/[^/]+/clip/
                        )
                        (?P<id>[^/?#&]+)
                    '''
859

860
    _TESTS = [{
861
        'url': 'https://clips.twitch.tv/FaintLightGullWholeWheat',
862
863
        'md5': '761769e1eafce0ffebfb4089cb3847cd',
        'info_dict': {
864
            'id': '42850523',
865
866
            'ext': 'mp4',
            'title': 'EA Play 2016 Live from the Novo Theatre',
867
            'thumbnail': r're:^https?://.*\.jpg',
868
869
            'timestamp': 1465767393,
            'upload_date': '20160612',
870
871
            'creator': 'EA',
            'uploader': 'stereotype_',
872
            'uploader_id': '43566419',
873
        },
874
875
876
877
    }, {
        # multiple formats
        'url': 'https://clips.twitch.tv/rflegendary/UninterestedBeeDAESuppy',
        'only_matching': True,
878
879
880
    }, {
        'url': 'https://www.twitch.tv/sergeynixon/clip/StormyThankfulSproutFutureMan',
        'only_matching': True,
881
882
883
    }, {
        'url': 'https://clips.twitch.tv/embed?clip=InquisitiveBreakableYogurtJebaited',
        'only_matching': True,
884
885
886
887
888
889
    }, {
        'url': 'https://m.twitch.tv/rossbroadcast/clip/ConfidentBraveHumanChefFrank',
        'only_matching': True,
    }, {
        'url': 'https://go.twitch.tv/rossbroadcast/clip/ConfidentBraveHumanChefFrank',
        'only_matching': True,
890
    }]
891
892
893
894

    def _real_extract(self, url):
        video_id = self._match_id(url)

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
        clip = self._download_json(
            'https://gql.twitch.tv/gql', video_id, data=json.dumps({
                'query': '''{
  clip(slug: "%s") {
    broadcaster {
      displayName
    }
    createdAt
    curator {
      displayName
      id
    }
    durationSeconds
    id
    tiny: thumbnailURL(width: 86, height: 45)
    small: thumbnailURL(width: 260, height: 147)
    medium: thumbnailURL(width: 480, height: 272)
    title
    videoQualities {
      frameRate
      quality
      sourceURL
    }
    viewCount
  }
}''' % video_id,
            }).encode(), headers={
                'Client-ID': self._CLIENT_ID,
            })['data']['clip']

        if not clip:
            raise ExtractorError(
                'This clip is no longer available', expected=True)
928

929
        formats = []
930
        for option in clip.get('videoQualities', []):
931
932
            if not isinstance(option, dict):
                continue
933
            source = url_or_none(option.get('sourceURL'))
Sergey M․'s avatar
Sergey M․ committed
934
            if not source:
935
936
937
938
939
                continue
            formats.append({
                'url': source,
                'format_id': option.get('quality'),
                'height': int_or_none(option.get('quality')),
940
                'fps': int_or_none(option.get('frameRate')),
941
            })
Sergey M․'s avatar
Sergey M․ committed
942
943
        self._sort_formats(formats)

944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
        thumbnails = []
        for thumbnail_id in ('tiny', 'small', 'medium'):
            thumbnail_url = clip.get(thumbnail_id)
            if not thumbnail_url:
                continue
            thumb = {
                'id': thumbnail_id,
                'url': thumbnail_url,
            }
            mobj = re.search(r'-(\d+)x(\d+)\.', thumbnail_url)
            if mobj:
                thumb.update({
                    'height': int(mobj.group(2)),
                    'width': int(mobj.group(1)),
                })
            thumbnails.append(thumb)

        return {
            'id': clip.get('id') or video_id,
            'title': clip.get('title') or video_id,
964
            'formats': formats,
965
966
967
968
969
970
971
            'duration': int_or_none(clip.get('durationSeconds')),
            'views': int_or_none(clip.get('viewCount')),
            'timestamp': unified_timestamp(clip.get('createdAt')),
            'thumbnails': thumbnails,
            'creator': try_get(clip, lambda x: x['broadcaster']['displayName'], compat_str),
            'uploader': try_get(clip, lambda x: x['curator']['displayName'], compat_str),
            'uploader_id': try_get(clip, lambda x: x['curator']['id'], compat_str),
972
        }