twitter.py (28035B)
1 # coding: utf-8 2 from __future__ import unicode_literals 3 4 import re 5 6 from .common import InfoExtractor 7 from ..compat import ( 8 compat_HTTPError, 9 compat_parse_qs, 10 compat_urllib_parse_unquote, 11 compat_urllib_parse_urlparse, 12 ) 13 from ..utils import ( 14 dict_get, 15 ExtractorError, 16 float_or_none, 17 int_or_none, 18 try_get, 19 strip_or_none, 20 unified_timestamp, 21 update_url_query, 22 url_or_none, 23 xpath_text, 24 ) 25 26 from .periscope import ( 27 PeriscopeBaseIE, 28 PeriscopeIE, 29 ) 30 31 32 class TwitterBaseIE(InfoExtractor): 33 _API_BASE = 'https://api.twitter.com/1.1/' 34 _BASE_REGEX = r'https?://(?:(?:www|m(?:obile)?)\.)?twitter\.com/' 35 _GUEST_TOKEN = None 36 37 def _extract_variant_formats(self, variant, video_id): 38 variant_url = variant.get('url') 39 if not variant_url: 40 return [] 41 elif '.m3u8' in variant_url: 42 return self._extract_m3u8_formats( 43 variant_url, video_id, 'mp4', 'm3u8_native', 44 m3u8_id='hls', fatal=False) 45 else: 46 tbr = int_or_none(dict_get(variant, ('bitrate', 'bit_rate')), 1000) or None 47 f = { 48 'url': variant_url, 49 'format_id': 'http' + ('-%d' % tbr if tbr else ''), 50 'tbr': tbr, 51 } 52 self._search_dimensions_in_video_url(f, variant_url) 53 return [f] 54 55 def _extract_formats_from_vmap_url(self, vmap_url, video_id): 56 vmap_url = url_or_none(vmap_url) 57 if not vmap_url: 58 return [] 59 vmap_data = self._download_xml(vmap_url, video_id) 60 formats = [] 61 urls = [] 62 for video_variant in vmap_data.findall('.//{http://twitter.com/schema/videoVMapV2.xsd}videoVariant'): 63 video_variant.attrib['url'] = compat_urllib_parse_unquote( 64 video_variant.attrib['url']) 65 urls.append(video_variant.attrib['url']) 66 formats.extend(self._extract_variant_formats( 67 video_variant.attrib, video_id)) 68 video_url = strip_or_none(xpath_text(vmap_data, './/MediaFile')) 69 if video_url not in urls: 70 formats.extend(self._extract_variant_formats({'url': video_url}, video_id)) 71 return formats 72 73 @staticmethod 74 def _search_dimensions_in_video_url(a_format, video_url): 75 m = re.search(r'/(?P<width>\d+)x(?P<height>\d+)/', video_url) 76 if m: 77 a_format.update({ 78 'width': int(m.group('width')), 79 'height': int(m.group('height')), 80 }) 81 82 def _call_api(self, path, video_id, query={}): 83 headers = { 84 'Authorization': 'Bearer AAAAAAAAAAAAAAAAAAAAAPYXBAAAAAAACLXUNDekMxqa8h%2F40K4moUkGsoc%3DTYfbDKbT3jJPCEVnMYqilB28NHfOPqkca3qaAxGfsyKCs0wRbw', 85 } 86 if not self._GUEST_TOKEN: 87 self._GUEST_TOKEN = self._download_json( 88 self._API_BASE + 'guest/activate.json', video_id, 89 'Downloading guest token', data=b'', 90 headers=headers)['guest_token'] 91 headers['x-guest-token'] = self._GUEST_TOKEN 92 try: 93 return self._download_json( 94 self._API_BASE + path, video_id, headers=headers, query=query) 95 except ExtractorError as e: 96 if isinstance(e.cause, compat_HTTPError) and e.cause.code == 403: 97 raise ExtractorError(self._parse_json( 98 e.cause.read().decode(), 99 video_id)['errors'][0]['message'], expected=True) 100 raise 101 102 103 class TwitterCardIE(InfoExtractor): 104 IE_NAME = 'twitter:card' 105 _VALID_URL = TwitterBaseIE._BASE_REGEX + r'i/(?:cards/tfw/v1|videos(?:/tweet)?)/(?P<id>\d+)' 106 _TESTS = [ 107 { 108 'url': 'https://twitter.com/i/cards/tfw/v1/560070183650213889', 109 # MD5 checksums are different in different places 110 'info_dict': { 111 'id': '560070183650213889', 112 'ext': 'mp4', 113 'title': "Twitter - You can now shoot, edit and share video on Twitter. Capture life's most moving moments from your perspective.", 114 'description': 'md5:18d3e24bb4f6e5007487dd546e53bd96', 115 'uploader': 'Twitter', 116 'uploader_id': 'Twitter', 117 'thumbnail': r're:^https?://.*\.jpg', 118 'duration': 30.033, 119 'timestamp': 1422366112, 120 'upload_date': '20150127', 121 }, 122 }, 123 { 124 'url': 'https://twitter.com/i/cards/tfw/v1/623160978427936768', 125 'md5': '7137eca597f72b9abbe61e5ae0161399', 126 'info_dict': { 127 'id': '623160978427936768', 128 'ext': 'mp4', 129 'title': "NASA - Fly over Pluto's icy Norgay Mountains and Sputnik Plain in this @NASANewHorizons #PlutoFlyby video.", 130 'description': "Fly over Pluto's icy Norgay Mountains and Sputnik Plain in this @NASANewHorizons #PlutoFlyby video. https://t.co/BJYgOjSeGA", 131 'uploader': 'NASA', 132 'uploader_id': 'NASA', 133 'timestamp': 1437408129, 134 'upload_date': '20150720', 135 }, 136 }, 137 { 138 'url': 'https://twitter.com/i/cards/tfw/v1/654001591733886977', 139 'md5': 'b6d9683dd3f48e340ded81c0e917ad46', 140 'info_dict': { 141 'id': 'dq4Oj5quskI', 142 'ext': 'mp4', 143 'title': 'Ubuntu 11.10 Overview', 144 'description': 'md5:a831e97fa384863d6e26ce48d1c43376', 145 'upload_date': '20111013', 146 'uploader': 'OMG! UBUNTU!', 147 'uploader_id': 'omgubuntu', 148 }, 149 'add_ie': ['Youtube'], 150 }, 151 { 152 'url': 'https://twitter.com/i/cards/tfw/v1/665289828897005568', 153 'md5': '6dabeaca9e68cbb71c99c322a4b42a11', 154 'info_dict': { 155 'id': 'iBb2x00UVlv', 156 'ext': 'mp4', 157 'upload_date': '20151113', 158 'uploader_id': '1189339351084113920', 159 'uploader': 'ArsenalTerje', 160 'title': 'Vine by ArsenalTerje', 161 'timestamp': 1447451307, 162 }, 163 'add_ie': ['Vine'], 164 }, { 165 'url': 'https://twitter.com/i/videos/tweet/705235433198714880', 166 'md5': '884812a2adc8aaf6fe52b15ccbfa3b88', 167 'info_dict': { 168 'id': '705235433198714880', 169 'ext': 'mp4', 170 'title': "Brent Yarina - Khalil Iverson's missed highlight dunk. And made highlight dunk. In one highlight.", 171 'description': "Khalil Iverson's missed highlight dunk. And made highlight dunk. In one highlight. https://t.co/OrxcJ28Bns", 172 'uploader': 'Brent Yarina', 173 'uploader_id': 'BTNBrentYarina', 174 'timestamp': 1456976204, 175 'upload_date': '20160303', 176 }, 177 'skip': 'This content is no longer available.', 178 }, { 179 'url': 'https://twitter.com/i/videos/752274308186120192', 180 'only_matching': True, 181 }, 182 ] 183 184 def _real_extract(self, url): 185 status_id = self._match_id(url) 186 return self.url_result( 187 'https://twitter.com/statuses/' + status_id, 188 TwitterIE.ie_key(), status_id) 189 190 191 class TwitterIE(TwitterBaseIE): 192 IE_NAME = 'twitter' 193 _VALID_URL = TwitterBaseIE._BASE_REGEX + r'(?:(?:i/web|[^/]+)/status|statuses)/(?P<id>\d+)' 194 195 _TESTS = [{ 196 'url': 'https://twitter.com/freethenipple/status/643211948184596480', 197 'info_dict': { 198 'id': '643211948184596480', 199 'ext': 'mp4', 200 'title': 'FREE THE NIPPLE - FTN supporters on Hollywood Blvd today!', 201 'thumbnail': r're:^https?://.*\.jpg', 202 'description': 'FTN supporters on Hollywood Blvd today! http://t.co/c7jHH749xJ', 203 'uploader': 'FREE THE NIPPLE', 204 'uploader_id': 'freethenipple', 205 'duration': 12.922, 206 'timestamp': 1442188653, 207 'upload_date': '20150913', 208 'age_limit': 18, 209 }, 210 }, { 211 'url': 'https://twitter.com/giphz/status/657991469417025536/photo/1', 212 'md5': 'f36dcd5fb92bf7057f155e7d927eeb42', 213 'info_dict': { 214 'id': '657991469417025536', 215 'ext': 'mp4', 216 'title': 'Gifs - tu vai cai tu vai cai tu nao eh capaz disso tu vai cai', 217 'description': 'Gifs on Twitter: "tu vai cai tu vai cai tu nao eh capaz disso tu vai cai https://t.co/tM46VHFlO5"', 218 'thumbnail': r're:^https?://.*\.png', 219 'uploader': 'Gifs', 220 'uploader_id': 'giphz', 221 }, 222 'expected_warnings': ['height', 'width'], 223 'skip': 'Account suspended', 224 }, { 225 'url': 'https://twitter.com/starwars/status/665052190608723968', 226 'info_dict': { 227 'id': '665052190608723968', 228 'ext': 'mp4', 229 'title': 'Star Wars - A new beginning is coming December 18. Watch the official 60 second #TV spot for #StarWars: #TheForceAwakens.', 230 'description': 'A new beginning is coming December 18. Watch the official 60 second #TV spot for #StarWars: #TheForceAwakens. https://t.co/OkSqT2fjWJ', 231 'uploader_id': 'starwars', 232 'uploader': 'Star Wars', 233 'timestamp': 1447395772, 234 'upload_date': '20151113', 235 }, 236 }, { 237 'url': 'https://twitter.com/BTNBrentYarina/status/705235433198714880', 238 'info_dict': { 239 'id': '705235433198714880', 240 'ext': 'mp4', 241 'title': "Brent Yarina - Khalil Iverson's missed highlight dunk. And made highlight dunk. In one highlight.", 242 'description': "Khalil Iverson's missed highlight dunk. And made highlight dunk. In one highlight. https://t.co/OrxcJ28Bns", 243 'uploader_id': 'BTNBrentYarina', 244 'uploader': 'Brent Yarina', 245 'timestamp': 1456976204, 246 'upload_date': '20160303', 247 }, 248 'params': { 249 # The same video as https://twitter.com/i/videos/tweet/705235433198714880 250 # Test case of TwitterCardIE 251 'skip_download': True, 252 }, 253 }, { 254 'url': 'https://twitter.com/jaydingeer/status/700207533655363584', 255 'info_dict': { 256 'id': '700207533655363584', 257 'ext': 'mp4', 258 'title': 'simon vertugo - BEAT PROD: @suhmeduh #Damndaniel', 259 'description': 'BEAT PROD: @suhmeduh https://t.co/HBrQ4AfpvZ #Damndaniel https://t.co/byBooq2ejZ', 260 'thumbnail': r're:^https?://.*\.jpg', 261 'uploader': 'simon vertugo', 262 'uploader_id': 'simonvertugo', 263 'duration': 30.0, 264 'timestamp': 1455777459, 265 'upload_date': '20160218', 266 }, 267 }, { 268 'url': 'https://twitter.com/Filmdrunk/status/713801302971588609', 269 'md5': '89a15ed345d13b86e9a5a5e051fa308a', 270 'info_dict': { 271 'id': 'MIOxnrUteUd', 272 'ext': 'mp4', 273 'title': 'Dr.Pepperの飲み方 #japanese #バカ #ドクペ #電動ガン', 274 'uploader': 'TAKUMA', 275 'uploader_id': '1004126642786242560', 276 'timestamp': 1402826626, 277 'upload_date': '20140615', 278 }, 279 'add_ie': ['Vine'], 280 }, { 281 'url': 'https://twitter.com/captainamerica/status/719944021058060289', 282 'info_dict': { 283 'id': '719944021058060289', 284 'ext': 'mp4', 285 'title': 'Captain America - @King0fNerd Are you sure you made the right choice? Find out in theaters.', 286 'description': '@King0fNerd Are you sure you made the right choice? Find out in theaters. https://t.co/GpgYi9xMJI', 287 'uploader_id': 'CaptainAmerica', 288 'uploader': 'Captain America', 289 'duration': 3.17, 290 'timestamp': 1460483005, 291 'upload_date': '20160412', 292 }, 293 }, { 294 'url': 'https://twitter.com/OPP_HSD/status/779210622571536384', 295 'info_dict': { 296 'id': '1zqKVVlkqLaKB', 297 'ext': 'mp4', 298 'title': 'Sgt Kerry Schmidt - Ontario Provincial Police - Road rage, mischief, assault, rollover and fire in one occurrence', 299 'upload_date': '20160923', 300 'uploader_id': '1PmKqpJdOJQoY', 301 'uploader': 'Sgt Kerry Schmidt - Ontario Provincial Police', 302 'timestamp': 1474613214, 303 }, 304 'add_ie': ['Periscope'], 305 }, { 306 # has mp4 formats via mobile API 307 'url': 'https://twitter.com/news_al3alm/status/852138619213144067', 308 'info_dict': { 309 'id': '852138619213144067', 310 'ext': 'mp4', 311 'title': 'عالم الأخبار - كلمة تاريخية بجلسة الجناسي التاريخية.. النائب خالد مؤنس العتيبي للمعارضين : اتقوا الله .. الظلم ظلمات يوم القيامة', 312 'description': 'كلمة تاريخية بجلسة الجناسي التاريخية.. النائب خالد مؤنس العتيبي للمعارضين : اتقوا الله .. الظلم ظلمات يوم القيامة https://t.co/xg6OhpyKfN', 313 'uploader': 'عالم الأخبار', 314 'uploader_id': 'news_al3alm', 315 'duration': 277.4, 316 'timestamp': 1492000653, 317 'upload_date': '20170412', 318 }, 319 'skip': 'Account suspended', 320 }, { 321 'url': 'https://twitter.com/i/web/status/910031516746514432', 322 'info_dict': { 323 'id': '910031516746514432', 324 'ext': 'mp4', 325 'title': 'Préfet de Guadeloupe - [Direct] #Maria Le centre se trouve actuellement au sud de Basse-Terre. Restez confinés. Réfugiez-vous dans la pièce la + sûre.', 326 'thumbnail': r're:^https?://.*\.jpg', 327 'description': '[Direct] #Maria Le centre se trouve actuellement au sud de Basse-Terre. Restez confinés. Réfugiez-vous dans la pièce la + sûre. https://t.co/mwx01Rs4lo', 328 'uploader': 'Préfet de Guadeloupe', 329 'uploader_id': 'Prefet971', 330 'duration': 47.48, 331 'timestamp': 1505803395, 332 'upload_date': '20170919', 333 }, 334 'params': { 335 'skip_download': True, # requires ffmpeg 336 }, 337 }, { 338 # card via api.twitter.com/1.1/videos/tweet/config 339 'url': 'https://twitter.com/LisPower1/status/1001551623938805763', 340 'info_dict': { 341 'id': '1001551623938805763', 342 'ext': 'mp4', 343 'title': 're:.*?Shep is on a roll today.*?', 344 'thumbnail': r're:^https?://.*\.jpg', 345 'description': 'md5:37b9f2ff31720cef23b2bd42ee8a0f09', 346 'uploader': 'Lis Power', 347 'uploader_id': 'LisPower1', 348 'duration': 111.278, 349 'timestamp': 1527623489, 350 'upload_date': '20180529', 351 }, 352 'params': { 353 'skip_download': True, # requires ffmpeg 354 }, 355 }, { 356 'url': 'https://twitter.com/foobar/status/1087791357756956680', 357 'info_dict': { 358 'id': '1087791357756956680', 359 'ext': 'mp4', 360 'title': 'Twitter - A new is coming. Some of you got an opt-in to try it now. Check out the emoji button, quick keyboard shortcuts, upgraded trends, advanced search, and more. Let us know your thoughts!', 361 'thumbnail': r're:^https?://.*\.jpg', 362 'description': 'md5:6dfd341a3310fb97d80d2bf7145df976', 363 'uploader': 'Twitter', 364 'uploader_id': 'Twitter', 365 'duration': 61.567, 366 'timestamp': 1548184644, 367 'upload_date': '20190122', 368 }, 369 }, { 370 # not available in Periscope 371 'url': 'https://twitter.com/ViviEducation/status/1136534865145286656', 372 'info_dict': { 373 'id': '1vOGwqejwoWxB', 374 'ext': 'mp4', 375 'title': 'Vivi - Vivi founder @lior_rauchy announcing our new student feedback tool live at @EduTECH_AU #EduTECH2019', 376 'uploader': 'Vivi', 377 'uploader_id': '1eVjYOLGkGrQL', 378 }, 379 'add_ie': ['TwitterBroadcast'], 380 }, { 381 # unified card 382 'url': 'https://twitter.com/BrooklynNets/status/1349794411333394432?s=20', 383 'info_dict': { 384 'id': '1349794411333394432', 385 'ext': 'mp4', 386 'title': 'md5:d1c4941658e4caaa6cb579260d85dcba', 387 'thumbnail': r're:^https?://.*\.jpg', 388 'description': 'md5:71ead15ec44cee55071547d6447c6a3e', 389 'uploader': 'Brooklyn Nets', 390 'uploader_id': 'BrooklynNets', 391 'duration': 324.484, 392 'timestamp': 1610651040, 393 'upload_date': '20210114', 394 }, 395 'params': { 396 'skip_download': True, 397 }, 398 }, { 399 # Twitch Clip Embed 400 'url': 'https://twitter.com/GunB1g/status/1163218564784017422', 401 'only_matching': True, 402 }, { 403 # promo_video_website card 404 'url': 'https://twitter.com/GunB1g/status/1163218564784017422', 405 'only_matching': True, 406 }, { 407 # promo_video_convo card 408 'url': 'https://twitter.com/poco_dandy/status/1047395834013384704', 409 'only_matching': True, 410 }, { 411 # appplayer card 412 'url': 'https://twitter.com/poco_dandy/status/1150646424461176832', 413 'only_matching': True, 414 }, { 415 # video_direct_message card 416 'url': 'https://twitter.com/qarev001/status/1348948114569269251', 417 'only_matching': True, 418 }, { 419 # poll2choice_video card 420 'url': 'https://twitter.com/CAF_Online/status/1349365911120195585', 421 'only_matching': True, 422 }, { 423 # poll3choice_video card 424 'url': 'https://twitter.com/SamsungMobileSA/status/1348609186725289984', 425 'only_matching': True, 426 }, { 427 # poll4choice_video card 428 'url': 'https://twitter.com/SouthamptonFC/status/1347577658079641604', 429 'only_matching': True, 430 }] 431 432 def _real_extract(self, url): 433 twid = self._match_id(url) 434 status = self._call_api( 435 'statuses/show/%s.json' % twid, twid, { 436 'cards_platform': 'Web-12', 437 'include_cards': 1, 438 'include_reply_count': 1, 439 'include_user_entities': 0, 440 'tweet_mode': 'extended', 441 }) 442 443 title = description = status['full_text'].replace('\n', ' ') 444 # strip 'https -_t.co_BJYgOjSeGA' junk from filenames 445 title = re.sub(r'\s+(https?://[^ ]+)', '', title) 446 user = status.get('user') or {} 447 uploader = user.get('name') 448 if uploader: 449 title = '%s - %s' % (uploader, title) 450 uploader_id = user.get('screen_name') 451 452 tags = [] 453 for hashtag in (try_get(status, lambda x: x['entities']['hashtags'], list) or []): 454 hashtag_text = hashtag.get('text') 455 if not hashtag_text: 456 continue 457 tags.append(hashtag_text) 458 459 info = { 460 'id': twid, 461 'title': title, 462 'description': description, 463 'uploader': uploader, 464 'timestamp': unified_timestamp(status.get('created_at')), 465 'uploader_id': uploader_id, 466 'uploader_url': 'https://twitter.com/' + uploader_id if uploader_id else None, 467 'like_count': int_or_none(status.get('favorite_count')), 468 'repost_count': int_or_none(status.get('retweet_count')), 469 'comment_count': int_or_none(status.get('reply_count')), 470 'age_limit': 18 if status.get('possibly_sensitive') else 0, 471 'tags': tags, 472 } 473 474 def extract_from_video_info(media): 475 video_info = media.get('video_info') or {} 476 477 formats = [] 478 for variant in video_info.get('variants', []): 479 formats.extend(self._extract_variant_formats(variant, twid)) 480 self._sort_formats(formats) 481 482 thumbnails = [] 483 media_url = media.get('media_url_https') or media.get('media_url') 484 if media_url: 485 def add_thumbnail(name, size): 486 thumbnails.append({ 487 'id': name, 488 'url': update_url_query(media_url, {'name': name}), 489 'width': int_or_none(size.get('w') or size.get('width')), 490 'height': int_or_none(size.get('h') or size.get('height')), 491 }) 492 for name, size in media.get('sizes', {}).items(): 493 add_thumbnail(name, size) 494 add_thumbnail('orig', media.get('original_info') or {}) 495 496 info.update({ 497 'formats': formats, 498 'thumbnails': thumbnails, 499 'duration': float_or_none(video_info.get('duration_millis'), 1000), 500 }) 501 502 media = try_get(status, lambda x: x['extended_entities']['media'][0]) 503 if media and media.get('type') != 'photo': 504 extract_from_video_info(media) 505 else: 506 card = status.get('card') 507 if card: 508 binding_values = card['binding_values'] 509 510 def get_binding_value(k): 511 o = binding_values.get(k) or {} 512 return try_get(o, lambda x: x[x['type'].lower() + '_value']) 513 514 card_name = card['name'].split(':')[-1] 515 if card_name == 'player': 516 info.update({ 517 '_type': 'url', 518 'url': get_binding_value('player_url'), 519 }) 520 elif card_name == 'periscope_broadcast': 521 info.update({ 522 '_type': 'url', 523 'url': get_binding_value('url') or get_binding_value('player_url'), 524 'ie_key': PeriscopeIE.ie_key(), 525 }) 526 elif card_name == 'broadcast': 527 info.update({ 528 '_type': 'url', 529 'url': get_binding_value('broadcast_url'), 530 'ie_key': TwitterBroadcastIE.ie_key(), 531 }) 532 elif card_name == 'summary': 533 info.update({ 534 '_type': 'url', 535 'url': get_binding_value('card_url'), 536 }) 537 elif card_name == 'unified_card': 538 media_entities = self._parse_json(get_binding_value('unified_card'), twid)['media_entities'] 539 extract_from_video_info(next(iter(media_entities.values()))) 540 # amplify, promo_video_website, promo_video_convo, appplayer, 541 # video_direct_message, poll2choice_video, poll3choice_video, 542 # poll4choice_video, ... 543 else: 544 is_amplify = card_name == 'amplify' 545 vmap_url = get_binding_value('amplify_url_vmap') if is_amplify else get_binding_value('player_stream_url') 546 content_id = get_binding_value('%s_content_id' % (card_name if is_amplify else 'player')) 547 formats = self._extract_formats_from_vmap_url(vmap_url, content_id or twid) 548 self._sort_formats(formats) 549 550 thumbnails = [] 551 for suffix in ('_small', '', '_large', '_x_large', '_original'): 552 image = get_binding_value('player_image' + suffix) or {} 553 image_url = image.get('url') 554 if not image_url or '/player-placeholder' in image_url: 555 continue 556 thumbnails.append({ 557 'id': suffix[1:] if suffix else 'medium', 558 'url': image_url, 559 'width': int_or_none(image.get('width')), 560 'height': int_or_none(image.get('height')), 561 }) 562 563 info.update({ 564 'formats': formats, 565 'thumbnails': thumbnails, 566 'duration': int_or_none(get_binding_value( 567 'content_duration_seconds')), 568 }) 569 else: 570 expanded_url = try_get(status, lambda x: x['entities']['urls'][0]['expanded_url']) 571 if not expanded_url: 572 raise ExtractorError("There's no video in this tweet.") 573 info.update({ 574 '_type': 'url', 575 'url': expanded_url, 576 }) 577 return info 578 579 580 class TwitterAmplifyIE(TwitterBaseIE): 581 IE_NAME = 'twitter:amplify' 582 _VALID_URL = r'https?://amp\.twimg\.com/v/(?P<id>[0-9a-f\-]{36})' 583 584 _TEST = { 585 'url': 'https://amp.twimg.com/v/0ba0c3c7-0af3-4c0a-bed5-7efd1ffa2951', 586 'md5': '7df102d0b9fd7066b86f3159f8e81bf6', 587 'info_dict': { 588 'id': '0ba0c3c7-0af3-4c0a-bed5-7efd1ffa2951', 589 'ext': 'mp4', 590 'title': 'Twitter Video', 591 'thumbnail': 're:^https?://.*', 592 }, 593 } 594 595 def _real_extract(self, url): 596 video_id = self._match_id(url) 597 webpage = self._download_webpage(url, video_id) 598 599 vmap_url = self._html_search_meta( 600 'twitter:amplify:vmap', webpage, 'vmap url') 601 formats = self._extract_formats_from_vmap_url(vmap_url, video_id) 602 603 thumbnails = [] 604 thumbnail = self._html_search_meta( 605 'twitter:image:src', webpage, 'thumbnail', fatal=False) 606 607 def _find_dimension(target): 608 w = int_or_none(self._html_search_meta( 609 'twitter:%s:width' % target, webpage, fatal=False)) 610 h = int_or_none(self._html_search_meta( 611 'twitter:%s:height' % target, webpage, fatal=False)) 612 return w, h 613 614 if thumbnail: 615 thumbnail_w, thumbnail_h = _find_dimension('image') 616 thumbnails.append({ 617 'url': thumbnail, 618 'width': thumbnail_w, 619 'height': thumbnail_h, 620 }) 621 622 video_w, video_h = _find_dimension('player') 623 formats[0].update({ 624 'width': video_w, 625 'height': video_h, 626 }) 627 628 return { 629 'id': video_id, 630 'title': 'Twitter Video', 631 'formats': formats, 632 'thumbnails': thumbnails, 633 } 634 635 636 class TwitterBroadcastIE(TwitterBaseIE, PeriscopeBaseIE): 637 IE_NAME = 'twitter:broadcast' 638 _VALID_URL = TwitterBaseIE._BASE_REGEX + r'i/broadcasts/(?P<id>[0-9a-zA-Z]{13})' 639 640 _TEST = { 641 # untitled Periscope video 642 'url': 'https://twitter.com/i/broadcasts/1yNGaQLWpejGj', 643 'info_dict': { 644 'id': '1yNGaQLWpejGj', 645 'ext': 'mp4', 646 'title': 'Andrea May Sahouri - Periscope Broadcast', 647 'uploader': 'Andrea May Sahouri', 648 'uploader_id': '1PXEdBZWpGwKe', 649 }, 650 } 651 652 def _real_extract(self, url): 653 broadcast_id = self._match_id(url) 654 broadcast = self._call_api( 655 'broadcasts/show.json', broadcast_id, 656 {'ids': broadcast_id})['broadcasts'][broadcast_id] 657 info = self._parse_broadcast_data(broadcast, broadcast_id) 658 media_key = broadcast['media_key'] 659 source = self._call_api( 660 'live_video_stream/status/' + media_key, media_key)['source'] 661 m3u8_url = source.get('noRedirectPlaybackUrl') or source['location'] 662 if '/live_video_stream/geoblocked/' in m3u8_url: 663 self.raise_geo_restricted() 664 m3u8_id = compat_parse_qs(compat_urllib_parse_urlparse( 665 m3u8_url).query).get('type', [None])[0] 666 state, width, height = self._extract_common_format_info(broadcast) 667 info['formats'] = self._extract_pscp_m3u8_formats( 668 m3u8_url, broadcast_id, m3u8_id, state, width, height) 669 return info