2 from __future__ import unicode_literals
6 from .common import InfoExtractor
7 from ..compat import compat_urlparse
19 from .periscope import PeriscopeIE
22 class TwitterBaseIE(InfoExtractor):
23 def _extract_formats_from_vmap_url(self, vmap_url, video_id):
24 vmap_data = self._download_xml(vmap_url, video_id)
25 video_url = xpath_text(vmap_data, './/MediaFile').strip()
26 if determine_ext(video_url) == 'm3u8':
27 return self._extract_m3u8_formats(
28 video_url, video_id, ext='mp4', m3u8_id='hls',
29 entry_protocol='m3u8_native')
35 def _search_dimensions_in_video_url(a_format, video_url):
36 m = re.search(r'/(?P<width>\d+)x(?P<height>\d+)/', video_url)
39 'width': int(m.group('width')),
40 'height': int(m.group('height')),
44 class TwitterCardIE(TwitterBaseIE):
45 IE_NAME = 'twitter:card'
46 _VALID_URL = r'https?://(?:www\.)?twitter\.com/i/(?:cards/tfw/v1|videos(?:/tweet)?)/(?P<id>\d+)'
49 'url': 'https://twitter.com/i/cards/tfw/v1/560070183650213889',
50 # MD5 checksums are different in different places
52 'id': '560070183650213889',
54 'title': 'Twitter Card',
55 'thumbnail': r're:^https?://.*\.jpg$',
61 'url': 'https://twitter.com/i/cards/tfw/v1/623160978427936768',
62 'md5': '7ee2a553b63d1bccba97fbed97d9e1c8',
64 'id': '623160978427936768',
66 'title': 'Twitter Card',
67 'thumbnail': r're:^https?://.*\.jpg',
73 'url': 'https://twitter.com/i/cards/tfw/v1/654001591733886977',
74 'md5': 'b6d9683dd3f48e340ded81c0e917ad46',
78 'title': 'Ubuntu 11.10 Overview',
79 'description': 'md5:a831e97fa384863d6e26ce48d1c43376',
80 'upload_date': '20111013',
81 'uploader': 'OMG! Ubuntu!',
82 'uploader_id': 'omgubuntu',
84 'add_ie': ['Youtube'],
87 'url': 'https://twitter.com/i/cards/tfw/v1/665289828897005568',
88 'md5': '6dabeaca9e68cbb71c99c322a4b42a11',
92 'upload_date': '20151113',
93 'uploader_id': '1189339351084113920',
94 'uploader': 'ArsenalTerje',
95 'title': 'Vine by ArsenalTerje',
96 'timestamp': 1447451307,
100 'url': 'https://twitter.com/i/videos/tweet/705235433198714880',
101 'md5': '884812a2adc8aaf6fe52b15ccbfa3b88',
103 'id': '705235433198714880',
105 'title': 'Twitter web player',
106 'thumbnail': r're:^https?://.*',
109 'url': 'https://twitter.com/i/videos/752274308186120192',
110 'only_matching': True,
114 def _parse_media_info(self, media_info, video_id):
116 for media_variant in media_info.get('variants', []):
117 media_url = media_variant['url']
118 if media_url.endswith('.m3u8'):
119 formats.extend(self._extract_m3u8_formats(media_url, video_id, ext='mp4', m3u8_id='hls'))
120 elif media_url.endswith('.mpd'):
121 formats.extend(self._extract_mpd_formats(media_url, video_id, mpd_id='dash'))
123 vbr = int_or_none(dict_get(media_variant, ('bitRate', 'bitrate')), scale=1000)
126 'format_id': 'http-%d' % vbr if vbr else 'http',
129 # Reported bitRate may be zero
130 if not a_format['vbr']:
133 self._search_dimensions_in_video_url(a_format, media_url)
135 formats.append(a_format)
138 def _extract_mobile_formats(self, username, video_id):
139 webpage = self._download_webpage(
140 'https://mobile.twitter.com/%s/status/%s' % (username, video_id),
141 video_id, 'Downloading mobile webpage',
143 # A recent mobile UA is necessary for `gt` cookie
144 'User-Agent': 'Mozilla/5.0 (Android 6.0.1; Mobile; rv:54.0) Gecko/54.0 Firefox/54.0',
146 main_script_url = self._html_search_regex(
147 r'<script[^>]+src="([^"]+main\.[^"]+)"', webpage, 'main script URL')
148 main_script = self._download_webpage(
149 main_script_url, video_id, 'Downloading main script')
150 bearer_token = self._search_regex(
151 r'BEARER_TOKEN\s*:\s*"([^"]+)"',
152 main_script, 'bearer token')
153 guest_token = self._search_regex(
154 r'document\.cookie\s*=\s*decodeURIComponent\("gt=(\d+)',
155 webpage, 'guest token')
156 api_data = self._download_json(
157 'https://api.twitter.com/2/timeline/conversation/%s.json' % video_id,
158 video_id, 'Downloading mobile API data',
160 'Authorization': 'Bearer ' + bearer_token,
161 'x-guest-token': guest_token,
163 media_info = try_get(api_data, lambda o: o['globalObjects']['tweets'][video_id]
164 ['extended_entities']['media'][0]['video_info']) or {}
165 return self._parse_media_info(media_info, video_id)
167 def _real_extract(self, url):
168 video_id = self._match_id(url)
174 webpage = self._download_webpage(url, video_id)
176 iframe_url = self._html_search_regex(
177 r'<iframe[^>]+src="((?:https?:)?//(?:www.youtube.com/embed/[^"]+|(?:www\.)?vine\.co/v/\w+/card))"',
178 webpage, 'video iframe', default=None)
180 return self.url_result(iframe_url)
182 config = self._parse_json(self._html_search_regex(
183 r'data-(?:player-)?config="([^"]+)"', webpage,
184 'data player config', default='{}'),
187 if config.get('source_type') == 'vine':
188 return self.url_result(config['player_url'], 'Vine')
190 periscope_url = PeriscopeIE._extract_url(webpage)
192 return self.url_result(periscope_url, PeriscopeIE.ie_key())
194 video_url = config.get('video_url') or config.get('playlist', [{}])[0].get('source')
197 if determine_ext(video_url) == 'm3u8':
198 formats.extend(self._extract_m3u8_formats(video_url, video_id, ext='mp4', m3u8_id='hls'))
204 self._search_dimensions_in_video_url(f, video_url)
208 vmap_url = config.get('vmapUrl') or config.get('vmap_url')
211 self._extract_formats_from_vmap_url(vmap_url, video_id))
215 for entity in config.get('status', {}).get('entities', []):
216 if 'mediaInfo' in entity:
217 media_info = entity['mediaInfo']
220 formats.extend(self._parse_media_info(media_info, video_id))
221 duration = float_or_none(media_info.get('duration', {}).get('nanos'), scale=1e9)
223 username = config.get('user', {}).get('screen_name')
225 formats.extend(self._extract_mobile_formats(username, video_id))
227 self._remove_duplicate_formats(formats)
228 self._sort_formats(formats)
230 title = self._search_regex(r'<title>([^<]+)</title>', webpage, 'title')
231 thumbnail = config.get('posterImageUrl') or config.get('image_src')
232 duration = float_or_none(config.get('duration')) or duration
237 'thumbnail': thumbnail,
238 'duration': duration,
243 class TwitterIE(InfoExtractor):
245 _VALID_URL = r'https?://(?:www\.|m\.|mobile\.)?twitter\.com/(?P<user_id>[^/]+)/status/(?P<id>\d+)'
246 _TEMPLATE_URL = 'https://twitter.com/%s/status/%s'
249 'url': 'https://twitter.com/freethenipple/status/643211948184596480',
251 'id': '643211948184596480',
253 'title': 'FREE THE NIPPLE - FTN supporters on Hollywood Blvd today!',
254 'thumbnail': r're:^https?://.*\.jpg',
255 'description': 'FREE THE NIPPLE on Twitter: "FTN supporters on Hollywood Blvd today! http://t.co/c7jHH749xJ"',
256 'uploader': 'FREE THE NIPPLE',
257 'uploader_id': 'freethenipple',
260 'skip_download': True, # requires ffmpeg
263 'url': 'https://twitter.com/giphz/status/657991469417025536/photo/1',
264 'md5': 'f36dcd5fb92bf7057f155e7d927eeb42',
266 'id': '657991469417025536',
268 'title': 'Gifs - tu vai cai tu vai cai tu nao eh capaz disso tu vai cai',
269 'description': 'Gifs on Twitter: "tu vai cai tu vai cai tu nao eh capaz disso tu vai cai https://t.co/tM46VHFlO5"',
270 'thumbnail': r're:^https?://.*\.png',
272 'uploader_id': 'giphz',
274 'expected_warnings': ['height', 'width'],
275 'skip': 'Account suspended',
277 'url': 'https://twitter.com/starwars/status/665052190608723968',
278 'md5': '39b7199856dee6cd4432e72c74bc69d4',
280 'id': '665052190608723968',
282 'title': 'Star Wars - A new beginning is coming December 18. Watch the official 60 second #TV spot for #StarWars: #TheForceAwakens.',
283 'description': 'Star Wars on Twitter: "A new beginning is coming December 18. Watch the official 60 second #TV spot for #StarWars: #TheForceAwakens."',
284 'uploader_id': 'starwars',
285 'uploader': 'Star Wars',
288 'url': 'https://twitter.com/BTNBrentYarina/status/705235433198714880',
290 'id': '705235433198714880',
292 'title': 'Brent Yarina - Khalil Iverson\'s missed highlight dunk. And made highlight dunk. In one highlight.',
293 'description': 'Brent Yarina on Twitter: "Khalil Iverson\'s missed highlight dunk. And made highlight dunk. In one highlight."',
294 'uploader_id': 'BTNBrentYarina',
295 'uploader': 'Brent Yarina',
298 # The same video as https://twitter.com/i/videos/tweet/705235433198714880
299 # Test case of TwitterCardIE
300 'skip_download': True,
303 'url': 'https://twitter.com/jaydingeer/status/700207533655363584',
306 'id': '700207533655363584',
308 'title': 'Donte - BEAT PROD: @suhmeduh #Damndaniel',
309 'description': 'Donte on Twitter: "BEAT PROD: @suhmeduh https://t.co/HBrQ4AfpvZ #Damndaniel https://t.co/byBooq2ejZ"',
310 'thumbnail': r're:^https?://.*\.jpg',
312 'uploader_id': 'jaydingeer',
315 'skip_download': True, # requires ffmpeg
318 'url': 'https://twitter.com/Filmdrunk/status/713801302971588609',
319 'md5': '89a15ed345d13b86e9a5a5e051fa308a',
323 'title': 'FilmDrunk - Vine of the day',
324 'description': 'FilmDrunk on Twitter: "Vine of the day https://t.co/xmTvRdqxWf"',
325 'uploader': 'FilmDrunk',
326 'uploader_id': 'Filmdrunk',
327 'timestamp': 1402826626,
328 'upload_date': '20140615',
332 'url': 'https://twitter.com/captainamerica/status/719944021058060289',
334 'id': '719944021058060289',
336 'title': 'Captain America - @King0fNerd Are you sure you made the right choice? Find out in theaters.',
337 'description': 'Captain America on Twitter: "@King0fNerd Are you sure you made the right choice? Find out in theaters. https://t.co/GpgYi9xMJI"',
338 'uploader_id': 'captainamerica',
339 'uploader': 'Captain America',
342 'skip_download': True, # requires ffmpeg
345 'url': 'https://twitter.com/OPP_HSD/status/779210622571536384',
347 'id': '1zqKVVlkqLaKB',
349 'title': 'Sgt Kerry Schmidt - LIVE on #Periscope: Road rage, mischief, assault, rollover and fire in one occurrence',
350 'description': 'Sgt Kerry Schmidt on Twitter: "LIVE on #Periscope: Road rage, mischief, assault, rollover and fire in one occurrence https://t.co/EKrVgIXF3s"',
351 'upload_date': '20160923',
352 'uploader_id': 'OPP_HSD',
353 'uploader': 'Sgt Kerry Schmidt',
354 'timestamp': 1474613214,
356 'add_ie': ['Periscope'],
358 # has mp4 formats via mobile API
359 'url': 'https://twitter.com/news_al3alm/status/852138619213144067',
361 'id': '852138619213144067',
363 'title': 'عالم الأخبار - كلمة تاريخية بجلسة الجناسي التاريخية.. النائب خالد مؤنس العتيبي للمعارضين : اتقوا الله .. الظلم ظلمات يوم القيامة',
364 'description': 'عالم الأخبار on Twitter: "كلمة تاريخية بجلسة الجناسي التاريخية.. النائب خالد مؤنس العتيبي للمعارضين : اتقوا الله .. الظلم ظلمات يوم القيامة https://t.co/xg6OhpyKfN"',
365 'uploader': 'عالم الأخبار',
366 'uploader_id': 'news_al3alm',
369 'format': 'best[format_id^=http-]',
373 def _real_extract(self, url):
374 mobj = re.match(self._VALID_URL, url)
375 user_id = mobj.group('user_id')
376 twid = mobj.group('id')
378 webpage, urlh = self._download_webpage_handle(
379 self._TEMPLATE_URL % (user_id, twid), twid)
381 if 'twitter.com/account/suspended' in urlh.geturl():
382 raise ExtractorError('Account suspended by Twitter.', expected=True)
384 username = remove_end(self._og_search_title(webpage), ' on Twitter')
386 title = description = self._og_search_description(webpage).strip('').replace('\n', ' ').strip('“”')
388 # strip 'https -_t.co_BJYgOjSeGA' junk from filenames
389 title = re.sub(r'\s+(https?://[^ ]+)', '', title)
392 'uploader_id': user_id,
393 'uploader': username,
395 'description': '%s on Twitter: "%s"' % (username, description),
396 'title': username + ' - ' + title,
399 mobj = re.search(r'''(?x)
400 <video[^>]+class="animated-gif"(?P<more_info>[^>]+)>\s*
401 <source[^>]+video-src="(?P<url>[^"]+)"
405 more_info = mobj.group('more_info')
406 height = int_or_none(self._search_regex(
407 r'data-height="(\d+)"', more_info, 'height', fatal=False))
408 width = int_or_none(self._search_regex(
409 r'data-width="(\d+)"', more_info, 'width', fatal=False))
410 thumbnail = self._search_regex(
411 r'poster="([^"]+)"', more_info, 'poster', fatal=False)
414 'url': mobj.group('url'),
417 'thumbnail': thumbnail,
421 twitter_card_url = None
422 if 'class="PlayableMedia' in webpage:
423 twitter_card_url = '%s//twitter.com/i/videos/tweet/%s' % (self.http_scheme(), twid)
425 twitter_card_iframe_url = self._search_regex(
426 r'data-full-card-iframe-url=([\'"])(?P<url>(?:(?!\1).)+)\1',
427 webpage, 'Twitter card iframe URL', default=None, group='url')
428 if twitter_card_iframe_url:
429 twitter_card_url = compat_urlparse.urljoin(url, twitter_card_iframe_url)
433 '_type': 'url_transparent',
434 'ie_key': 'TwitterCard',
435 'url': twitter_card_url,
439 raise ExtractorError('There\'s no video in this tweet.')
442 class TwitterAmplifyIE(TwitterBaseIE):
443 IE_NAME = 'twitter:amplify'
444 _VALID_URL = r'https?://amp\.twimg\.com/v/(?P<id>[0-9a-f\-]{36})'
447 'url': 'https://amp.twimg.com/v/0ba0c3c7-0af3-4c0a-bed5-7efd1ffa2951',
448 'md5': '7df102d0b9fd7066b86f3159f8e81bf6',
450 'id': '0ba0c3c7-0af3-4c0a-bed5-7efd1ffa2951',
452 'title': 'Twitter Video',
453 'thumbnail': 're:^https?://.*',
457 def _real_extract(self, url):
458 video_id = self._match_id(url)
459 webpage = self._download_webpage(url, video_id)
461 vmap_url = self._html_search_meta(
462 'twitter:amplify:vmap', webpage, 'vmap url')
463 formats = self._extract_formats_from_vmap_url(vmap_url, video_id)
466 thumbnail = self._html_search_meta(
467 'twitter:image:src', webpage, 'thumbnail', fatal=False)
469 def _find_dimension(target):
470 w = int_or_none(self._html_search_meta(
471 'twitter:%s:width' % target, webpage, fatal=False))
472 h = int_or_none(self._html_search_meta(
473 'twitter:%s:height' % target, webpage, fatal=False))
477 thumbnail_w, thumbnail_h = _find_dimension('image')
480 'width': thumbnail_w,
481 'height': thumbnail_h,
484 video_w, video_h = _find_dimension('player')
492 'title': 'Twitter Video',
494 'thumbnails': thumbnails,