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     float_or_none,
   9     xpath_text,
  10     remove_end,
  11     int_or_none,
  12     ExtractorError,
  13     sanitized_Request,
  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/(?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': 'TwitterCard',
  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': 'TwitterCard',
  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-things-to-do-after-installing-ubuntu-11-10/',
  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     ]
  77
  78     def _real_extract(self, url):
  79         video_id = self._match_id(url)
  80
  81         # Different formats served for different User-Agents
  82         USER_AGENTS = [
  83             'Mozilla/5.0 (X11; Linux x86_64; rv:10.0) Gecko/20150101 Firefox/20.0 (Chrome)',  # mp4
  84             'Mozilla/5.0 (Windows NT 5.2; WOW64; rv:38.0) Gecko/20100101 Firefox/38.0',  # webm
  85         ]
  86
  87         config = None
  88         formats = []
  89         for user_agent in USER_AGENTS:
  90             request = sanitized_Request(url)
  91             request.add_header('User-Agent', user_agent)
  92             webpage = self._download_webpage(request, video_id)
  93
  94             iframe_url = self._html_search_regex(
  95                 r'<iframe[^>]+src="((?:https?:)?//(?:www.youtube.com/embed/[^"]+|(?:www\.)?vine\.co/v/\w+/card))"',
  96                 webpage, 'video iframe', default=None)
  97             if iframe_url:
  98                 return self.url_result(iframe_url)
  99
 100             config = self._parse_json(self._html_search_regex(
 101                 r'data-player-config="([^"]+)"', webpage, 'data player config'),
 102                 video_id)
 103             if 'playlist' not in config:
 104                 if 'vmapUrl' in config:
 105                     formats.append({
 106                         'url': self._get_vmap_video_url(config['vmapUrl'], video_id),
 107                     })
 108                     break   # same video regardless of UA
 109                 continue
 110
 111             video_url = config['playlist'][0]['source']
 112
 113             f = {
 114                 'url': video_url,
 115             }
 116
 117             m = re.search(r'/(?P<width>\d+)x(?P<height>\d+)/', video_url)
 118             if m:
 119                 f.update({
 120                     'width': int(m.group('width')),
 121                     'height': int(m.group('height')),
 122                 })
 123             formats.append(f)
 124         self._sort_formats(formats)
 125
 126         thumbnail = config.get('posterImageUrl')
 127         duration = float_or_none(config.get('duration'))
 128
 129         return {
 130             'id': video_id,
 131             'title': 'TwitterCard',
 132             'thumbnail': thumbnail,
 133             'duration': duration,
 134             'formats': formats,
 135         }
 136
 137
 138 class TwitterIE(InfoExtractor):
 139     IE_NAME = 'twitter'
 140     _VALID_URL = r'https?://(?:www\.|m\.|mobile\.)?twitter\.com/(?P<user_id>[^/]+)/status/(?P<id>\d+)'
 141     _TEMPLATE_URL = 'https://twitter.com/%s/status/%s'
 142
 143     _TESTS = [{
 144         'url': 'https://twitter.com/freethenipple/status/643211948184596480',
 145         # MD5 checksums are different in different places
 146         'info_dict': {
 147             'id': '643211948184596480',
 148             'ext': 'mp4',
 149             'title': 'FREE THE NIPPLE - FTN supporters on Hollywood Blvd today!',
 150             'thumbnail': 're:^https?://.*\.jpg',
 151             'duration': 12.922,
 152             'description': 'FREE THE NIPPLE on Twitter: "FTN supporters on Hollywood Blvd today! http://t.co/c7jHH749xJ"',
 153             'uploader': 'FREE THE NIPPLE',
 154             'uploader_id': 'freethenipple',
 155         },
 156     }, {
 157         'url': 'https://twitter.com/giphz/status/657991469417025536/photo/1',
 158         'md5': 'f36dcd5fb92bf7057f155e7d927eeb42',
 159         'info_dict': {
 160             'id': '657991469417025536',
 161             'ext': 'mp4',
 162             'title': 'Gifs - tu vai cai tu vai cai tu nao eh capaz disso tu vai cai',
 163             'description': 'Gifs on Twitter: "tu vai cai tu vai cai tu nao eh capaz disso tu vai cai https://t.co/tM46VHFlO5"',
 164             'thumbnail': 're:^https?://.*\.png',
 165             'uploader': 'Gifs',
 166             'uploader_id': 'giphz',
 167         },
 168         'expected_warnings': ['height', 'width'],
 169     }, {
 170         'url': 'https://twitter.com/starwars/status/665052190608723968',
 171         'md5': '39b7199856dee6cd4432e72c74bc69d4',
 172         'info_dict': {
 173             'id': '665052190608723968',
 174             'ext': 'mp4',
 175             'title': 'Star Wars - A new beginning is coming December 18. Watch the official 60 second #TV spot for #StarWars: #TheForceAwakens.',
 176             'description': 'Star Wars on Twitter: "A new beginning is coming December 18. Watch the official 60 second #TV spot for #StarWars: #TheForceAwakens."',
 177             'uploader_id': 'starwars',
 178             'uploader': 'Star Wars',
 179         },
 180     }]
 181
 182     def _real_extract(self, url):
 183         mobj = re.match(self._VALID_URL, url)
 184         user_id = mobj.group('user_id')
 185         twid = mobj.group('id')
 186
 187         webpage = self._download_webpage(self._TEMPLATE_URL % (user_id, twid), twid)
 188
 189         username = remove_end(self._og_search_title(webpage), ' on Twitter')
 190
 191         title = description = self._og_search_description(webpage).strip('').replace('\n', ' ').strip('“”')
 192
 193         # strip  'https -_t.co_BJYgOjSeGA' junk from filenames
 194         title = re.sub(r'\s+(https?://[^ ]+)', '', title)
 195
 196         info = {
 197             'uploader_id': user_id,
 198             'uploader': username,
 199             'webpage_url': url,
 200             'description': '%s on Twitter: "%s"' % (username, description),
 201             'title': username + ' - ' + title,
 202         }
 203
 204         card_id = self._search_regex(
 205             r'["\']/i/cards/tfw/v1/(\d+)', webpage, 'twitter card url', default=None)
 206         if card_id:
 207             card_url = 'https://twitter.com/i/cards/tfw/v1/' + card_id
 208             info.update({
 209                 '_type': 'url_transparent',
 210                 'ie_key': 'TwitterCard',
 211                 'url': card_url,
 212             })
 213             return info
 214
 215         mobj = re.search(r'''(?x)
 216             <video[^>]+class="animated-gif"(?P<more_info>[^>]+)>\s*
 217                 <source[^>]+video-src="(?P<url>[^"]+)"
 218         ''', webpage)
 219
 220         if mobj:
 221             more_info = mobj.group('more_info')
 222             height = int_or_none(self._search_regex(
 223                 r'data-height="(\d+)"', more_info, 'height', fatal=False))
 224             width = int_or_none(self._search_regex(
 225                 r'data-width="(\d+)"', more_info, 'width', fatal=False))
 226             thumbnail = self._search_regex(
 227                 r'poster="([^"]+)"', more_info, 'poster', fatal=False)
 228             info.update({
 229                 'id': twid,
 230                 'url': mobj.group('url'),
 231                 'height': height,
 232                 'width': width,
 233                 'thumbnail': thumbnail,
 234             })
 235             return info
 236
 237         raise ExtractorError('There\'s no video in this tweet.')
 238
 239
 240 class TwitterAmplifyIE(TwitterBaseIE):
 241     IE_NAME = 'twitter:amplify'
 242     _VALID_URL = 'https?://amp\.twimg\.com/v/(?P<id>[0-9a-f\-]{36})'
 243
 244     _TEST = {
 245         'url': 'https://amp.twimg.com/v/0ba0c3c7-0af3-4c0a-bed5-7efd1ffa2951',
 246         'md5': '7df102d0b9fd7066b86f3159f8e81bf6',
 247         'info_dict': {
 248             'id': '0ba0c3c7-0af3-4c0a-bed5-7efd1ffa2951',
 249             'ext': 'mp4',
 250             'title': 'Twitter Video',
 251             'thumbnail': 're:^https?://.*',
 252         },
 253     }
 254
 255     def _real_extract(self, url):
 256         video_id = self._match_id(url)
 257         webpage = self._download_webpage(url, video_id)
 258
 259         vmap_url = self._html_search_meta(
 260             'twitter:amplify:vmap', webpage, 'vmap url')
 261         video_url = self._get_vmap_video_url(vmap_url, video_id)
 262
 263         thumbnails = []
 264         thumbnail = self._html_search_meta(
 265             'twitter:image:src', webpage, 'thumbnail', fatal=False)
 266
 267         def _find_dimension(target):
 268             w = int_or_none(self._html_search_meta(
 269                 'twitter:%s:width' % target, webpage, fatal=False))
 270             h = int_or_none(self._html_search_meta(
 271                 'twitter:%s:height' % target, webpage, fatal=False))
 272             return w, h
 273
 274         if thumbnail:
 275             thumbnail_w, thumbnail_h = _find_dimension('image')
 276             thumbnails.append({
 277                 'url': thumbnail,
 278                 'width': thumbnail_w,
 279                 'height': thumbnail_h,
 280             })
 281
 282         video_w, video_h = _find_dimension('player')
 283         formats = [{
 284             'url': video_url,
 285             'width': video_w,
 286             'height': video_h,
 287         }]
 288
 289         return {
 290             'id': video_id,
 291             'title': 'Twitter Video',
 292             'formats': formats,
 293             'thumbnails': thumbnails,
 294         }