1# coding: utf-8 2from __future__ import unicode_literals 3 4import re 5 6from .common import InfoExtractor 7from ..compat import ( 8 compat_HTTPError, 9 compat_parse_qs, 10 compat_urllib_parse_unquote, 11 compat_urllib_parse_urlparse, 12) 13from ..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 26from .periscope import ( 27 PeriscopeBaseIE, 28 PeriscopeIE, 29) 30 31 32class 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_and_subtitles( 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 subtitles = {} 62 urls = [] 63 for video_variant in vmap_data.findall('.//{http://twitter.com/schema/videoVMapV2.xsd}videoVariant'): 64 video_variant.attrib['url'] = compat_urllib_parse_unquote( 65 video_variant.attrib['url']) 66 urls.append(video_variant.attrib['url']) 67 fmts, subs = self._extract_variant_formats( 68 video_variant.attrib, video_id) 69 formats.extend(fmts) 70 subtitles = self._merge_subtitles(subtitles, subs) 71 video_url = strip_or_none(xpath_text(vmap_data, './/MediaFile')) 72 if video_url not in urls: 73 fmts, subs = self._extract_variant_formats({'url': video_url}, video_id) 74 formats.extend(fmts) 75 subtitles = self._merge_subtitles(subtitles, subs) 76 return formats, subtitles 77 78 @staticmethod 79 def _search_dimensions_in_video_url(a_format, video_url): 80 m = re.search(r'/(?P<width>\d+)x(?P<height>\d+)/', video_url) 81 if m: 82 a_format.update({ 83 'width': int(m.group('width')), 84 'height': int(m.group('height')), 85 }) 86 87 def _call_api(self, path, video_id, query={}): 88 headers = { 89 'Authorization': 'Bearer AAAAAAAAAAAAAAAAAAAAAPYXBAAAAAAACLXUNDekMxqa8h%2F40K4moUkGsoc%3DTYfbDKbT3jJPCEVnMYqilB28NHfOPqkca3qaAxGfsyKCs0wRbw', 90 } 91 if not self._GUEST_TOKEN: 92 self._GUEST_TOKEN = self._download_json( 93 self._API_BASE + 'guest/activate.json', video_id, 94 'Downloading guest token', data=b'', 95 headers=headers)['guest_token'] 96 headers['x-guest-token'] = self._GUEST_TOKEN 97 try: 98 return self._download_json( 99 self._API_BASE + path, video_id, headers=headers, query=query) 100 except ExtractorError as e: 101 if isinstance(e.cause, compat_HTTPError) and e.cause.code == 403: 102 raise ExtractorError(self._parse_json( 103 e.cause.read().decode(), 104 video_id)['errors'][0]['message'], expected=True) 105 raise 106 107 108class TwitterCardIE(InfoExtractor): 109 IE_NAME = 'twitter:card' 110 _VALID_URL = TwitterBaseIE._BASE_REGEX + r'i/(?:cards/tfw/v1|videos(?:/tweet)?)/(?P<id>\d+)' 111 _TESTS = [ 112 { 113 'url': 'https://twitter.com/i/cards/tfw/v1/560070183650213889', 114 # MD5 checksums are different in different places 115 'info_dict': { 116 'id': '560070183650213889', 117 'ext': 'mp4', 118 'title': "Twitter - You can now shoot, edit and share video on Twitter. Capture life's most moving moments from your perspective.", 119 'description': 'md5:18d3e24bb4f6e5007487dd546e53bd96', 120 'uploader': 'Twitter', 121 'uploader_id': 'Twitter', 122 'thumbnail': r're:^https?://.*\.jpg', 123 'duration': 30.033, 124 'timestamp': 1422366112, 125 'upload_date': '20150127', 126 }, 127 }, 128 { 129 'url': 'https://twitter.com/i/cards/tfw/v1/623160978427936768', 130 'md5': '7137eca597f72b9abbe61e5ae0161399', 131 'info_dict': { 132 'id': '623160978427936768', 133 'ext': 'mp4', 134 'title': "NASA - Fly over Pluto's icy Norgay Mountains and Sputnik Plain in this @NASANewHorizons #PlutoFlyby video.", 135 'description': "Fly over Pluto's icy Norgay Mountains and Sputnik Plain in this @NASANewHorizons #PlutoFlyby video. https://t.co/BJYgOjSeGA", 136 'uploader': 'NASA', 137 'uploader_id': 'NASA', 138 'timestamp': 1437408129, 139 'upload_date': '20150720', 140 }, 141 }, 142 { 143 'url': 'https://twitter.com/i/cards/tfw/v1/654001591733886977', 144 'md5': 'b6d9683dd3f48e340ded81c0e917ad46', 145 'info_dict': { 146 'id': 'dq4Oj5quskI', 147 'ext': 'mp4', 148 'title': 'Ubuntu 11.10 Overview', 149 'description': 'md5:a831e97fa384863d6e26ce48d1c43376', 150 'upload_date': '20111013', 151 'uploader': 'OMG! UBUNTU!', 152 'uploader_id': 'omgubuntu', 153 }, 154 'add_ie': ['Youtube'], 155 }, 156 { 157 'url': 'https://twitter.com/i/cards/tfw/v1/665289828897005568', 158 'md5': '6dabeaca9e68cbb71c99c322a4b42a11', 159 'info_dict': { 160 'id': 'iBb2x00UVlv', 161 'ext': 'mp4', 162 'upload_date': '20151113', 163 'uploader_id': '1189339351084113920', 164 'uploader': 'ArsenalTerje', 165 'title': 'Vine by ArsenalTerje', 166 'timestamp': 1447451307, 167 }, 168 'add_ie': ['Vine'], 169 }, { 170 'url': 'https://twitter.com/i/videos/tweet/705235433198714880', 171 'md5': '884812a2adc8aaf6fe52b15ccbfa3b88', 172 'info_dict': { 173 'id': '705235433198714880', 174 'ext': 'mp4', 175 'title': "Brent Yarina - Khalil Iverson's missed highlight dunk. And made highlight dunk. In one highlight.", 176 'description': "Khalil Iverson's missed highlight dunk. And made highlight dunk. In one highlight. https://t.co/OrxcJ28Bns", 177 'uploader': 'Brent Yarina', 178 'uploader_id': 'BTNBrentYarina', 179 'timestamp': 1456976204, 180 'upload_date': '20160303', 181 }, 182 'skip': 'This content is no longer available.', 183 }, { 184 'url': 'https://twitter.com/i/videos/752274308186120192', 185 'only_matching': True, 186 }, 187 ] 188 189 def _real_extract(self, url): 190 status_id = self._match_id(url) 191 return self.url_result( 192 'https://twitter.com/statuses/' + status_id, 193 TwitterIE.ie_key(), status_id) 194 195 196class TwitterIE(TwitterBaseIE): 197 IE_NAME = 'twitter' 198 _VALID_URL = TwitterBaseIE._BASE_REGEX + r'(?:(?:i/web|[^/]+)/status|statuses)/(?P<id>\d+)' 199 200 _TESTS = [{ 201 'url': 'https://twitter.com/freethenipple/status/643211948184596480', 202 'info_dict': { 203 'id': '643211948184596480', 204 'ext': 'mp4', 205 'title': 'FREE THE NIPPLE - FTN supporters on Hollywood Blvd today!', 206 'thumbnail': r're:^https?://.*\.jpg', 207 'description': 'FTN supporters on Hollywood Blvd today! http://t.co/c7jHH749xJ', 208 'uploader': 'FREE THE NIPPLE', 209 'uploader_id': 'freethenipple', 210 'duration': 12.922, 211 'timestamp': 1442188653, 212 'upload_date': '20150913', 213 'age_limit': 18, 214 }, 215 }, { 216 'url': 'https://twitter.com/giphz/status/657991469417025536/photo/1', 217 'md5': 'f36dcd5fb92bf7057f155e7d927eeb42', 218 'info_dict': { 219 'id': '657991469417025536', 220 'ext': 'mp4', 221 'title': 'Gifs - tu vai cai tu vai cai tu nao eh capaz disso tu vai cai', 222 'description': 'Gifs on Twitter: "tu vai cai tu vai cai tu nao eh capaz disso tu vai cai https://t.co/tM46VHFlO5"', 223 'thumbnail': r're:^https?://.*\.png', 224 'uploader': 'Gifs', 225 'uploader_id': 'giphz', 226 }, 227 'expected_warnings': ['height', 'width'], 228 'skip': 'Account suspended', 229 }, { 230 'url': 'https://twitter.com/starwars/status/665052190608723968', 231 'info_dict': { 232 'id': '665052190608723968', 233 'ext': 'mp4', 234 'title': 'Star Wars - A new beginning is coming December 18. Watch the official 60 second #TV spot for #StarWars: #TheForceAwakens.', 235 'description': 'A new beginning is coming December 18. Watch the official 60 second #TV spot for #StarWars: #TheForceAwakens. https://t.co/OkSqT2fjWJ', 236 'uploader_id': 'starwars', 237 'uploader': 'Star Wars', 238 'timestamp': 1447395772, 239 'upload_date': '20151113', 240 }, 241 }, { 242 'url': 'https://twitter.com/BTNBrentYarina/status/705235433198714880', 243 'info_dict': { 244 'id': '705235433198714880', 245 'ext': 'mp4', 246 'title': "Brent Yarina - Khalil Iverson's missed highlight dunk. And made highlight dunk. In one highlight.", 247 'description': "Khalil Iverson's missed highlight dunk. And made highlight dunk. In one highlight. https://t.co/OrxcJ28Bns", 248 'uploader_id': 'BTNBrentYarina', 249 'uploader': 'Brent Yarina', 250 'timestamp': 1456976204, 251 'upload_date': '20160303', 252 }, 253 'params': { 254 # The same video as https://twitter.com/i/videos/tweet/705235433198714880 255 # Test case of TwitterCardIE 256 'skip_download': True, 257 }, 258 }, { 259 'url': 'https://twitter.com/jaydingeer/status/700207533655363584', 260 'info_dict': { 261 'id': '700207533655363584', 262 'ext': 'mp4', 263 'title': 'simon vertugo - BEAT PROD: @suhmeduh #Damndaniel', 264 'description': 'BEAT PROD: @suhmeduh https://t.co/HBrQ4AfpvZ #Damndaniel https://t.co/byBooq2ejZ', 265 'thumbnail': r're:^https?://.*\.jpg', 266 'uploader': 'simon vertugo', 267 'uploader_id': 'simonvertugo', 268 'duration': 30.0, 269 'timestamp': 1455777459, 270 'upload_date': '20160218', 271 }, 272 }, { 273 'url': 'https://twitter.com/Filmdrunk/status/713801302971588609', 274 'md5': '89a15ed345d13b86e9a5a5e051fa308a', 275 'info_dict': { 276 'id': 'MIOxnrUteUd', 277 'ext': 'mp4', 278 'title': 'Dr.Pepperの飲み方 #japanese #バカ #ドクペ #電動ガン', 279 'uploader': 'TAKUMA', 280 'uploader_id': '1004126642786242560', 281 'timestamp': 1402826626, 282 'upload_date': '20140615', 283 }, 284 'add_ie': ['Vine'], 285 }, { 286 'url': 'https://twitter.com/captainamerica/status/719944021058060289', 287 'info_dict': { 288 'id': '719944021058060289', 289 'ext': 'mp4', 290 'title': 'Captain America - @King0fNerd Are you sure you made the right choice? Find out in theaters.', 291 'description': '@King0fNerd Are you sure you made the right choice? Find out in theaters. https://t.co/GpgYi9xMJI', 292 'uploader_id': 'CaptainAmerica', 293 'uploader': 'Captain America', 294 'duration': 3.17, 295 'timestamp': 1460483005, 296 'upload_date': '20160412', 297 }, 298 }, { 299 'url': 'https://twitter.com/OPP_HSD/status/779210622571536384', 300 'info_dict': { 301 'id': '1zqKVVlkqLaKB', 302 'ext': 'mp4', 303 'title': 'Sgt Kerry Schmidt - Ontario Provincial Police - Road rage, mischief, assault, rollover and fire in one occurrence', 304 'upload_date': '20160923', 305 'uploader_id': '1PmKqpJdOJQoY', 306 'uploader': 'Sgt Kerry Schmidt - Ontario Provincial Police', 307 'timestamp': 1474613214, 308 }, 309 'add_ie': ['Periscope'], 310 }, { 311 # has mp4 formats via mobile API 312 'url': 'https://twitter.com/news_al3alm/status/852138619213144067', 313 'info_dict': { 314 'id': '852138619213144067', 315 'ext': 'mp4', 316 'title': 'عالم الأخبار - كلمة تاريخية بجلسة الجناسي التاريخية.. النائب خالد مؤنس العتيبي للمعارضين : اتقوا الله .. الظلم ظلمات يوم القيامة', 317 'description': 'كلمة تاريخية بجلسة الجناسي التاريخية.. النائب خالد مؤنس العتيبي للمعارضين : اتقوا الله .. الظلم ظلمات يوم القيامة https://t.co/xg6OhpyKfN', 318 'uploader': 'عالم الأخبار', 319 'uploader_id': 'news_al3alm', 320 'duration': 277.4, 321 'timestamp': 1492000653, 322 'upload_date': '20170412', 323 }, 324 'skip': 'Account suspended', 325 }, { 326 'url': 'https://twitter.com/i/web/status/910031516746514432', 327 'info_dict': { 328 'id': '910031516746514432', 329 'ext': 'mp4', 330 '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.', 331 'thumbnail': r're:^https?://.*\.jpg', 332 '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', 333 'uploader': 'Préfet de Guadeloupe', 334 'uploader_id': 'Prefet971', 335 'duration': 47.48, 336 'timestamp': 1505803395, 337 'upload_date': '20170919', 338 }, 339 'params': { 340 'skip_download': True, # requires ffmpeg 341 }, 342 }, { 343 # card via api.twitter.com/1.1/videos/tweet/config 344 'url': 'https://twitter.com/LisPower1/status/1001551623938805763', 345 'info_dict': { 346 'id': '1001551623938805763', 347 'ext': 'mp4', 348 'title': 're:.*?Shep is on a roll today.*?', 349 'thumbnail': r're:^https?://.*\.jpg', 350 'description': 'md5:37b9f2ff31720cef23b2bd42ee8a0f09', 351 'uploader': 'Lis Power', 352 'uploader_id': 'LisPower1', 353 'duration': 111.278, 354 'timestamp': 1527623489, 355 'upload_date': '20180529', 356 }, 357 'params': { 358 'skip_download': True, # requires ffmpeg 359 }, 360 }, { 361 'url': 'https://twitter.com/foobar/status/1087791357756956680', 362 'info_dict': { 363 'id': '1087791357756956680', 364 'ext': 'mp4', 365 '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!', 366 'thumbnail': r're:^https?://.*\.jpg', 367 'description': 'md5:6dfd341a3310fb97d80d2bf7145df976', 368 'uploader': 'Twitter', 369 'uploader_id': 'Twitter', 370 'duration': 61.567, 371 'timestamp': 1548184644, 372 'upload_date': '20190122', 373 }, 374 }, { 375 # not available in Periscope 376 'url': 'https://twitter.com/ViviEducation/status/1136534865145286656', 377 'info_dict': { 378 'id': '1vOGwqejwoWxB', 379 'ext': 'mp4', 380 'title': 'Vivi - Vivi founder @lior_rauchy announcing our new student feedback tool live at @EduTECH_AU #EduTECH2019', 381 'uploader': 'Vivi', 382 'uploader_id': '1eVjYOLGkGrQL', 383 }, 384 'add_ie': ['TwitterBroadcast'], 385 }, { 386 # unified card 387 'url': 'https://twitter.com/BrooklynNets/status/1349794411333394432?s=20', 388 'info_dict': { 389 'id': '1349794411333394432', 390 'ext': 'mp4', 391 'title': 'md5:d1c4941658e4caaa6cb579260d85dcba', 392 'thumbnail': r're:^https?://.*\.jpg', 393 'description': 'md5:71ead15ec44cee55071547d6447c6a3e', 394 'uploader': 'Brooklyn Nets', 395 'uploader_id': 'BrooklynNets', 396 'duration': 324.484, 397 'timestamp': 1610651040, 398 'upload_date': '20210114', 399 }, 400 'params': { 401 'skip_download': True, 402 }, 403 }, { 404 # Twitch Clip Embed 405 'url': 'https://twitter.com/GunB1g/status/1163218564784017422', 406 'only_matching': True, 407 }, { 408 # promo_video_website card 409 'url': 'https://twitter.com/GunB1g/status/1163218564784017422', 410 'only_matching': True, 411 }, { 412 # promo_video_convo card 413 'url': 'https://twitter.com/poco_dandy/status/1047395834013384704', 414 'only_matching': True, 415 }, { 416 # appplayer card 417 'url': 'https://twitter.com/poco_dandy/status/1150646424461176832', 418 'only_matching': True, 419 }, { 420 # video_direct_message card 421 'url': 'https://twitter.com/qarev001/status/1348948114569269251', 422 'only_matching': True, 423 }, { 424 # poll2choice_video card 425 'url': 'https://twitter.com/CAF_Online/status/1349365911120195585', 426 'only_matching': True, 427 }, { 428 # poll3choice_video card 429 'url': 'https://twitter.com/SamsungMobileSA/status/1348609186725289984', 430 'only_matching': True, 431 }, { 432 # poll4choice_video card 433 'url': 'https://twitter.com/SouthamptonFC/status/1347577658079641604', 434 'only_matching': True, 435 }] 436 437 def _real_extract(self, url): 438 twid = self._match_id(url) 439 status = self._call_api( 440 'statuses/show/%s.json' % twid, twid, { 441 'cards_platform': 'Web-12', 442 'include_cards': 1, 443 'include_reply_count': 1, 444 'include_user_entities': 0, 445 'tweet_mode': 'extended', 446 }) 447 448 title = description = status['full_text'].replace('\n', ' ') 449 # strip 'https -_t.co_BJYgOjSeGA' junk from filenames 450 title = re.sub(r'\s+(https?://[^ ]+)', '', title) 451 user = status.get('user') or {} 452 uploader = user.get('name') 453 if uploader: 454 title = '%s - %s' % (uploader, title) 455 uploader_id = user.get('screen_name') 456 457 tags = [] 458 for hashtag in (try_get(status, lambda x: x['entities']['hashtags'], list) or []): 459 hashtag_text = hashtag.get('text') 460 if not hashtag_text: 461 continue 462 tags.append(hashtag_text) 463 464 info = { 465 'id': twid, 466 'title': title, 467 'description': description, 468 'uploader': uploader, 469 'timestamp': unified_timestamp(status.get('created_at')), 470 'uploader_id': uploader_id, 471 'uploader_url': 'https://twitter.com/' + uploader_id if uploader_id else None, 472 'like_count': int_or_none(status.get('favorite_count')), 473 'repost_count': int_or_none(status.get('retweet_count')), 474 'comment_count': int_or_none(status.get('reply_count')), 475 'age_limit': 18 if status.get('possibly_sensitive') else 0, 476 'tags': tags, 477 } 478 479 def extract_from_video_info(media): 480 video_info = media.get('video_info') or {} 481 482 formats = [] 483 subtitles = {} 484 for variant in video_info.get('variants', []): 485 fmts, subs = self._extract_variant_formats(variant, twid) 486 subtitles = self._merge_subtitles(subtitles, subs) 487 formats.extend(fmts) 488 self._sort_formats(formats, ('res', 'br', 'size', 'proto')) # The codec of http formats are unknown 489 490 thumbnails = [] 491 media_url = media.get('media_url_https') or media.get('media_url') 492 if media_url: 493 def add_thumbnail(name, size): 494 thumbnails.append({ 495 'id': name, 496 'url': update_url_query(media_url, {'name': name}), 497 'width': int_or_none(size.get('w') or size.get('width')), 498 'height': int_or_none(size.get('h') or size.get('height')), 499 }) 500 for name, size in media.get('sizes', {}).items(): 501 add_thumbnail(name, size) 502 add_thumbnail('orig', media.get('original_info') or {}) 503 504 info.update({ 505 'formats': formats, 506 'subtitles': subtitles, 507 'thumbnails': thumbnails, 508 'duration': float_or_none(video_info.get('duration_millis'), 1000), 509 }) 510 511 media = try_get(status, lambda x: x['extended_entities']['media'][0]) 512 if media and media.get('type') != 'photo': 513 extract_from_video_info(media) 514 else: 515 card = status.get('card') 516 if card: 517 binding_values = card['binding_values'] 518 519 def get_binding_value(k): 520 o = binding_values.get(k) or {} 521 return try_get(o, lambda x: x[x['type'].lower() + '_value']) 522 523 card_name = card['name'].split(':')[-1] 524 if card_name == 'player': 525 info.update({ 526 '_type': 'url', 527 'url': get_binding_value('player_url'), 528 }) 529 elif card_name == 'periscope_broadcast': 530 info.update({ 531 '_type': 'url', 532 'url': get_binding_value('url') or get_binding_value('player_url'), 533 'ie_key': PeriscopeIE.ie_key(), 534 }) 535 elif card_name == 'broadcast': 536 info.update({ 537 '_type': 'url', 538 'url': get_binding_value('broadcast_url'), 539 'ie_key': TwitterBroadcastIE.ie_key(), 540 }) 541 elif card_name == 'summary': 542 info.update({ 543 '_type': 'url', 544 'url': get_binding_value('card_url'), 545 }) 546 elif card_name == 'unified_card': 547 media_entities = self._parse_json(get_binding_value('unified_card'), twid)['media_entities'] 548 extract_from_video_info(next(iter(media_entities.values()))) 549 # amplify, promo_video_website, promo_video_convo, appplayer, 550 # video_direct_message, poll2choice_video, poll3choice_video, 551 # poll4choice_video, ... 552 else: 553 is_amplify = card_name == 'amplify' 554 vmap_url = get_binding_value('amplify_url_vmap') if is_amplify else get_binding_value('player_stream_url') 555 content_id = get_binding_value('%s_content_id' % (card_name if is_amplify else 'player')) 556 formats, subtitles = self._extract_formats_from_vmap_url(vmap_url, content_id or twid) 557 self._sort_formats(formats) 558 559 thumbnails = [] 560 for suffix in ('_small', '', '_large', '_x_large', '_original'): 561 image = get_binding_value('player_image' + suffix) or {} 562 image_url = image.get('url') 563 if not image_url or '/player-placeholder' in image_url: 564 continue 565 thumbnails.append({ 566 'id': suffix[1:] if suffix else 'medium', 567 'url': image_url, 568 'width': int_or_none(image.get('width')), 569 'height': int_or_none(image.get('height')), 570 }) 571 572 info.update({ 573 'formats': formats, 574 'subtitles': subtitles, 575 'thumbnails': thumbnails, 576 'duration': int_or_none(get_binding_value( 577 'content_duration_seconds')), 578 }) 579 else: 580 expanded_url = try_get(status, lambda x: x['entities']['urls'][0]['expanded_url']) 581 if not expanded_url: 582 raise ExtractorError("There's no video in this tweet.") 583 info.update({ 584 '_type': 'url', 585 'url': expanded_url, 586 }) 587 return info 588 589 590class TwitterAmplifyIE(TwitterBaseIE): 591 IE_NAME = 'twitter:amplify' 592 _VALID_URL = r'https?://amp\.twimg\.com/v/(?P<id>[0-9a-f\-]{36})' 593 594 _TEST = { 595 'url': 'https://amp.twimg.com/v/0ba0c3c7-0af3-4c0a-bed5-7efd1ffa2951', 596 'md5': '7df102d0b9fd7066b86f3159f8e81bf6', 597 'info_dict': { 598 'id': '0ba0c3c7-0af3-4c0a-bed5-7efd1ffa2951', 599 'ext': 'mp4', 600 'title': 'Twitter Video', 601 'thumbnail': 're:^https?://.*', 602 }, 603 } 604 605 def _real_extract(self, url): 606 video_id = self._match_id(url) 607 webpage = self._download_webpage(url, video_id) 608 609 vmap_url = self._html_search_meta( 610 'twitter:amplify:vmap', webpage, 'vmap url') 611 formats = self._extract_formats_from_vmap_url(vmap_url, video_id) 612 613 thumbnails = [] 614 thumbnail = self._html_search_meta( 615 'twitter:image:src', webpage, 'thumbnail', fatal=False) 616 617 def _find_dimension(target): 618 w = int_or_none(self._html_search_meta( 619 'twitter:%s:width' % target, webpage, fatal=False)) 620 h = int_or_none(self._html_search_meta( 621 'twitter:%s:height' % target, webpage, fatal=False)) 622 return w, h 623 624 if thumbnail: 625 thumbnail_w, thumbnail_h = _find_dimension('image') 626 thumbnails.append({ 627 'url': thumbnail, 628 'width': thumbnail_w, 629 'height': thumbnail_h, 630 }) 631 632 video_w, video_h = _find_dimension('player') 633 formats[0].update({ 634 'width': video_w, 635 'height': video_h, 636 }) 637 638 return { 639 'id': video_id, 640 'title': 'Twitter Video', 641 'formats': formats, 642 'thumbnails': thumbnails, 643 } 644 645 646class TwitterBroadcastIE(TwitterBaseIE, PeriscopeBaseIE): 647 IE_NAME = 'twitter:broadcast' 648 _VALID_URL = TwitterBaseIE._BASE_REGEX + r'i/broadcasts/(?P<id>[0-9a-zA-Z]{13})' 649 650 _TEST = { 651 # untitled Periscope video 652 'url': 'https://twitter.com/i/broadcasts/1yNGaQLWpejGj', 653 'info_dict': { 654 'id': '1yNGaQLWpejGj', 655 'ext': 'mp4', 656 'title': 'Andrea May Sahouri - Periscope Broadcast', 657 'uploader': 'Andrea May Sahouri', 658 'uploader_id': '1PXEdBZWpGwKe', 659 }, 660 } 661 662 def _real_extract(self, url): 663 broadcast_id = self._match_id(url) 664 broadcast = self._call_api( 665 'broadcasts/show.json', broadcast_id, 666 {'ids': broadcast_id})['broadcasts'][broadcast_id] 667 info = self._parse_broadcast_data(broadcast, broadcast_id) 668 media_key = broadcast['media_key'] 669 source = self._call_api( 670 'live_video_stream/status/' + media_key, media_key)['source'] 671 m3u8_url = source.get('noRedirectPlaybackUrl') or source['location'] 672 if '/live_video_stream/geoblocked/' in m3u8_url: 673 self.raise_geo_restricted() 674 m3u8_id = compat_parse_qs(compat_urllib_parse_urlparse( 675 m3u8_url).query).get('type', [None])[0] 676 state, width, height = self._extract_common_format_info(broadcast) 677 info['formats'] = self._extract_pscp_m3u8_formats( 678 m3u8_url, broadcast_id, m3u8_id, state, width, height) 679 return info 680 681 682class TwitterShortenerIE(TwitterBaseIE): 683 IE_NAME = 'twitter:shortener' 684 _VALID_URL = r'https?://t.co/(?P<id>[^?]+)|tco:(?P<eid>[^?]+)' 685 _BASE_URL = 'https://t.co/' 686 687 def _real_extract(self, url): 688 mobj = self._match_valid_url(url) 689 eid, id = mobj.group('eid', 'id') 690 if eid: 691 id = eid 692 url = self._BASE_URL + id 693 new_url = self._request_webpage(url, id, headers={'User-Agent': 'curl'}).geturl() 694 __UNSAFE_LINK = "https://twitter.com/safety/unsafe_link_warning?unsafe_link=" 695 if new_url.startswith(__UNSAFE_LINK): 696 new_url = new_url.replace(__UNSAFE_LINK, "") 697 return self.url_result(new_url) 698