Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/twitter.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..utils import (
   8     determine_ext,
   9     float_or_none,
  10     xpath_text,
  11     remove_end,
  12     int_or_none,
  13     ExtractorError,
  14 )
  15
  16
  17 class TwitterBaseIE(InfoExtractor):
  18     def _get_vmap_video_url(self, vmap_url, video_id):
  19         vmap_data = self._download_xml(vmap_url, video_id)
  20         return xpath_text(vmap_data, './/MediaFile').strip()
  21
  22
  23 class TwitterCardIE(TwitterBaseIE):
  24     IE_NAME = 'twitter:card'
  25     _VALID_URL = r'https?://(?:www\.)?twitter\.com/i/(?:cards/tfw/v1|videos/tweet)/(?P<id>\d+)'
  26     _TESTS = [
  27         {
  28             'url': 'https://twitter.com/i/cards/tfw/v1/560070183650213889',
  29             # MD5 checksums are different in different places
  30             'info_dict': {
  31                 'id': '560070183650213889',
  32                 'ext': 'mp4',
  33                 'title': 'Twitter Card',
  34                 'thumbnail': 're:^https?://.*\.jpg$',
  35                 'duration': 30.033,
  36             }
  37         },
  38         {
  39             'url': 'https://twitter.com/i/cards/tfw/v1/623160978427936768',
  40             'md5': '7ee2a553b63d1bccba97fbed97d9e1c8',
  41             'info_dict': {
  42                 'id': '623160978427936768',
  43                 'ext': 'mp4',
  44                 'title': 'Twitter Card',
  45                 'thumbnail': 're:^https?://.*\.jpg',
  46                 'duration': 80.155,
  47             },
  48         },
  49         {
  50             'url': 'https://twitter.com/i/cards/tfw/v1/654001591733886977',
  51             'md5': 'd4724ffe6d2437886d004fa5de1043b3',
  52             'info_dict': {
  53                 'id': 'dq4Oj5quskI',
  54                 'ext': 'mp4',
  55                 'title': 'Ubuntu 11.10 Overview',
  56                 'description': 'Take a quick peek at what\'s new and improved in Ubuntu 11.10.\n\nOnce installed take a look at 10 Things to Do After Installing: http://www.omgubuntu.co.uk/2011/10/10...',
  57                 'upload_date': '20111013',
  58                 'uploader': 'OMG! Ubuntu!',
  59                 'uploader_id': 'omgubuntu',
  60             },
  61             'add_ie': ['Youtube'],
  62         },
  63         {
  64             'url': 'https://twitter.com/i/cards/tfw/v1/665289828897005568',
  65             'md5': 'ab2745d0b0ce53319a534fccaa986439',
  66             'info_dict': {
  67                 'id': 'iBb2x00UVlv',
  68                 'ext': 'mp4',
  69                 'upload_date': '20151113',
  70                 'uploader_id': '1189339351084113920',
  71                 'uploader': 'ArsenalTerje',
  72                 'title': 'Vine by ArsenalTerje',
  73             },
  74             'add_ie': ['Vine'],
  75         }, {
  76             'url': 'https://twitter.com/i/videos/tweet/705235433198714880',
  77             'md5': '3846d0a07109b5ab622425449b59049d',
  78             'info_dict': {
  79                 'id': '705235433198714880',
  80                 'ext': 'mp4',
  81                 'title': 'Twitter web player',
  82                 'thumbnail': 're:^https?://.*\.jpg',
  83             },
  84         },
  85     ]
  86
  87     def _real_extract(self, url):
  88         video_id = self._match_id(url)
  89
  90         config = None
  91         formats = []
  92         duration = None
  93
  94         webpage = self._download_webpage(url, video_id)
  95
  96         iframe_url = self._html_search_regex(
  97             r'<iframe[^>]+src="((?:https?:)?//(?:www.youtube.com/embed/[^"]+|(?:www\.)?vine\.co/v/\w+/card))"',
  98             webpage, 'video iframe', default=None)
  99         if iframe_url:
 100             return self.url_result(iframe_url)
 101
 102         config = self._parse_json(self._html_search_regex(
 103             r'data-(?:player-)?config="([^"]+)"', webpage, 'data player config'),
 104             video_id)
 105
 106         if config.get('source_type') == 'vine':
 107             return self.url_result(config['player_url'], 'Vine')
 108
 109         def _search_dimensions_in_video_url(a_format, video_url):
 110             m = re.search(r'/(?P<width>\d+)x(?P<height>\d+)/', video_url)
 111             if m:
 112                 a_format.update({
 113                     'width': int(m.group('width')),
 114                     'height': int(m.group('height')),
 115                 })
 116
 117         video_url = config.get('video_url') or config.get('playlist', [{}])[0].get('source')
 118
 119         if video_url:
 120             if determine_ext(video_url) == 'm3u8':
 121                 formats.extend(self._extract_m3u8_formats(video_url, video_id, ext='mp4', m3u8_id='hls'))
 122             else:
 123                 f = {
 124                     'url': video_url,
 125                 }
 126
 127                 _search_dimensions_in_video_url(f, video_url)
 128
 129                 formats.append(f)
 130
 131         vmap_url = config.get('vmapUrl') or config.get('vmap_url')
 132         if vmap_url:
 133             formats.append({
 134                 'url': self._get_vmap_video_url(vmap_url, video_id),
 135             })
 136
 137         media_info = None
 138
 139         for entity in config.get('status', {}).get('entities', []):
 140             if 'mediaInfo' in entity:
 141                 media_info = entity['mediaInfo']
 142
 143         if media_info:
 144             for media_variant in media_info['variants']:
 145                 media_url = media_variant['url']
 146                 if media_url.endswith('.m3u8'):
 147                     formats.extend(self._extract_m3u8_formats(media_url, video_id, ext='mp4', m3u8_id='hls'))
 148                 elif media_url.endswith('.mpd'):
 149                     formats.extend(self._extract_mpd_formats(media_url, video_id, mpd_id='dash'))
 150                 else:
 151                     vbr = int_or_none(media_variant.get('bitRate'), scale=1000)
 152                     a_format = {
 153                         'url': media_url,
 154                         'format_id': 'http-%d' % vbr if vbr else 'http',
 155                         'vbr': vbr,
 156                     }
 157                     # Reported bitRate may be zero
 158                     if not a_format['vbr']:
 159                         del a_format['vbr']
 160
 161                     _search_dimensions_in_video_url(a_format, media_url)
 162
 163                     formats.append(a_format)
 164
 165             duration = float_or_none(media_info.get('duration', {}).get('nanos'), scale=1e9)
 166
 167         self._sort_formats(formats)
 168
 169         title = self._search_regex(r'<title>([^<]+)</title>', webpage, 'title')
 170         thumbnail = config.get('posterImageUrl') or config.get('image_src')
 171         duration = float_or_none(config.get('duration')) or duration
 172
 173         return {
 174             'id': video_id,
 175             'title': title,
 176             'thumbnail': thumbnail,
 177             'duration': duration,
 178             'formats': formats,
 179         }
 180
 181
 182 class TwitterIE(InfoExtractor):
 183     IE_NAME = 'twitter'
 184     _VALID_URL = r'https?://(?:www\.|m\.|mobile\.)?twitter\.com/(?P<user_id>[^/]+)/status/(?P<id>\d+)'
 185     _TEMPLATE_URL = 'https://twitter.com/%s/status/%s'
 186
 187     _TESTS = [{
 188         'url': 'https://twitter.com/freethenipple/status/643211948184596480',
 189         'info_dict': {
 190             'id': '643211948184596480',
 191             'ext': 'mp4',
 192             'title': 'FREE THE NIPPLE - FTN supporters on Hollywood Blvd today!',
 193             'thumbnail': 're:^https?://.*\.jpg',
 194             'description': 'FREE THE NIPPLE on Twitter: "FTN supporters on Hollywood Blvd today! http://t.co/c7jHH749xJ"',
 195             'uploader': 'FREE THE NIPPLE',
 196             'uploader_id': 'freethenipple',
 197         },
 198         'params': {
 199             'skip_download': True,  # requires ffmpeg
 200         },
 201     }, {
 202         'url': 'https://twitter.com/giphz/status/657991469417025536/photo/1',
 203         'md5': 'f36dcd5fb92bf7057f155e7d927eeb42',
 204         'info_dict': {
 205             'id': '657991469417025536',
 206             'ext': 'mp4',
 207             'title': 'Gifs - tu vai cai tu vai cai tu nao eh capaz disso tu vai cai',
 208             'description': 'Gifs on Twitter: "tu vai cai tu vai cai tu nao eh capaz disso tu vai cai https://t.co/tM46VHFlO5"',
 209             'thumbnail': 're:^https?://.*\.png',
 210             'uploader': 'Gifs',
 211             'uploader_id': 'giphz',
 212         },
 213         'expected_warnings': ['height', 'width'],
 214         'skip': 'Account suspended',
 215     }, {
 216         'url': 'https://twitter.com/starwars/status/665052190608723968',
 217         'md5': '39b7199856dee6cd4432e72c74bc69d4',
 218         'info_dict': {
 219             'id': '665052190608723968',
 220             'ext': 'mp4',
 221             'title': 'Star Wars - A new beginning is coming December 18. Watch the official 60 second #TV spot for #StarWars: #TheForceAwakens.',
 222             'description': 'Star Wars on Twitter: "A new beginning is coming December 18. Watch the official 60 second #TV spot for #StarWars: #TheForceAwakens."',
 223             'uploader_id': 'starwars',
 224             'uploader': 'Star Wars',
 225         },
 226     }, {
 227         'url': 'https://twitter.com/BTNBrentYarina/status/705235433198714880',
 228         'info_dict': {
 229             'id': '705235433198714880',
 230             'ext': 'mp4',
 231             'title': 'Brent Yarina - Khalil Iverson\'s missed highlight dunk. And made highlight dunk. In one highlight.',
 232             'description': 'Brent Yarina on Twitter: "Khalil Iverson\'s missed highlight dunk. And made highlight dunk. In one highlight."',
 233             'uploader_id': 'BTNBrentYarina',
 234             'uploader': 'Brent Yarina',
 235         },
 236         'params': {
 237             # The same video as https://twitter.com/i/videos/tweet/705235433198714880
 238             # Test case of TwitterCardIE
 239             'skip_download': True,
 240         },
 241     }, {
 242         'url': 'https://twitter.com/jaydingeer/status/700207533655363584',
 243         'md5': '',
 244         'info_dict': {
 245             'id': '700207533655363584',
 246             'ext': 'mp4',
 247             'title': 'Donte The Dumbass - BEAT PROD: @suhmeduh #Damndaniel',
 248             'description': 'Donte The Dumbass on Twitter: "BEAT PROD: @suhmeduh  https://t.co/HBrQ4AfpvZ #Damndaniel https://t.co/byBooq2ejZ"',
 249             'thumbnail': 're:^https?://.*\.jpg',
 250             'uploader': 'Donte The Dumbass',
 251             'uploader_id': 'jaydingeer',
 252         },
 253         'params': {
 254             'skip_download': True,  # requires ffmpeg
 255         },
 256     }, {
 257         'url': 'https://twitter.com/Filmdrunk/status/713801302971588609',
 258         'md5': '89a15ed345d13b86e9a5a5e051fa308a',
 259         'info_dict': {
 260             'id': 'MIOxnrUteUd',
 261             'ext': 'mp4',
 262             'title': 'Dr.Pepperの飲み方 #japanese #バカ #ドクペ #電動ガン',
 263             'uploader': 'TAKUMA',
 264             'uploader_id': '1004126642786242560',
 265             'upload_date': '20140615',
 266         },
 267         'add_ie': ['Vine'],
 268     }, {
 269         'url': 'https://twitter.com/captainamerica/status/719944021058060289',
 270         'info_dict': {
 271             'id': '719944021058060289',
 272             'ext': 'mp4',
 273             'title': 'Captain America - @King0fNerd Are you sure you made the right choice? Find out in theaters.',
 274             'description': 'Captain America on Twitter: "@King0fNerd Are you sure you made the right choice? Find out in theaters. https://t.co/GpgYi9xMJI"',
 275             'uploader_id': 'captainamerica',
 276             'uploader': 'Captain America',
 277         },
 278         'params': {
 279             'skip_download': True,  # requires ffmpeg
 280         },
 281     }]
 282
 283     def _real_extract(self, url):
 284         mobj = re.match(self._VALID_URL, url)
 285         user_id = mobj.group('user_id')
 286         twid = mobj.group('id')
 287
 288         webpage, urlh = self._download_webpage_handle(
 289             self._TEMPLATE_URL % (user_id, twid), twid)
 290
 291         if 'twitter.com/account/suspended' in urlh.geturl():
 292             raise ExtractorError('Account suspended by Twitter.', expected=True)
 293
 294         username = remove_end(self._og_search_title(webpage), ' on Twitter')
 295
 296         title = description = self._og_search_description(webpage).strip('').replace('\n', ' ').strip('“”')
 297
 298         # strip  'https -_t.co_BJYgOjSeGA' junk from filenames
 299         title = re.sub(r'\s+(https?://[^ ]+)', '', title)
 300
 301         info = {
 302             'uploader_id': user_id,
 303             'uploader': username,
 304             'webpage_url': url,
 305             'description': '%s on Twitter: "%s"' % (username, description),
 306             'title': username + ' - ' + title,
 307         }
 308
 309         mobj = re.search(r'''(?x)
 310             <video[^>]+class="animated-gif"(?P<more_info>[^>]+)>\s*
 311                 <source[^>]+video-src="(?P<url>[^"]+)"
 312         ''', webpage)
 313
 314         if mobj:
 315             more_info = mobj.group('more_info')
 316             height = int_or_none(self._search_regex(
 317                 r'data-height="(\d+)"', more_info, 'height', fatal=False))
 318             width = int_or_none(self._search_regex(
 319                 r'data-width="(\d+)"', more_info, 'width', fatal=False))
 320             thumbnail = self._search_regex(
 321                 r'poster="([^"]+)"', more_info, 'poster', fatal=False)
 322             info.update({
 323                 'id': twid,
 324                 'url': mobj.group('url'),
 325                 'height': height,
 326                 'width': width,
 327                 'thumbnail': thumbnail,
 328             })
 329             return info
 330
 331         if 'class="PlayableMedia' in webpage:
 332             info.update({
 333                 '_type': 'url_transparent',
 334                 'ie_key': 'TwitterCard',
 335                 'url': '%s//twitter.com/i/videos/tweet/%s' % (self.http_scheme(), twid),
 336             })
 337
 338             return info
 339
 340         raise ExtractorError('There\'s no video in this tweet.')
 341
 342
 343 class TwitterAmplifyIE(TwitterBaseIE):
 344     IE_NAME = 'twitter:amplify'
 345     _VALID_URL = 'https?://amp\.twimg\.com/v/(?P<id>[0-9a-f\-]{36})'
 346
 347     _TEST = {
 348         'url': 'https://amp.twimg.com/v/0ba0c3c7-0af3-4c0a-bed5-7efd1ffa2951',
 349         'md5': '7df102d0b9fd7066b86f3159f8e81bf6',
 350         'info_dict': {
 351             'id': '0ba0c3c7-0af3-4c0a-bed5-7efd1ffa2951',
 352             'ext': 'mp4',
 353             'title': 'Twitter Video',
 354             'thumbnail': 're:^https?://.*',
 355         },
 356     }
 357
 358     def _real_extract(self, url):
 359         video_id = self._match_id(url)
 360         webpage = self._download_webpage(url, video_id)
 361
 362         vmap_url = self._html_search_meta(
 363             'twitter:amplify:vmap', webpage, 'vmap url')
 364         video_url = self._get_vmap_video_url(vmap_url, video_id)
 365
 366         thumbnails = []
 367         thumbnail = self._html_search_meta(
 368             'twitter:image:src', webpage, 'thumbnail', fatal=False)
 369
 370         def _find_dimension(target):
 371             w = int_or_none(self._html_search_meta(
 372                 'twitter:%s:width' % target, webpage, fatal=False))
 373             h = int_or_none(self._html_search_meta(
 374                 'twitter:%s:height' % target, webpage, fatal=False))
 375             return w, h
 376
 377         if thumbnail:
 378             thumbnail_w, thumbnail_h = _find_dimension('image')
 379             thumbnails.append({
 380                 'url': thumbnail,
 381                 'width': thumbnail_w,
 382                 'height': thumbnail_h,
 383             })
 384
 385         video_w, video_h = _find_dimension('player')
 386         formats = [{
 387             'url': video_url,
 388             'width': video_w,
 389             'height': video_h,
 390         }]
 391
 392         return {
 393             'id': video_id,
 394             'title': 'Twitter Video',
 395             'formats': formats,
 396             'thumbnails': thumbnails,
 397         }