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 TwitterCardIE(InfoExtractor):
  18     IE_NAME = 'twitter:card'
  19     _VALID_URL = r'https?://(?:www\.)?twitter\.com/i/cards/tfw/v1/(?P<id>\d+)'
  20     _TESTS = [
  21         {
  22             'url': 'https://twitter.com/i/cards/tfw/v1/560070183650213889',
  23             'md5': '4fa26a35f9d1bf4b646590ba8e84be19',
  24             'info_dict': {
  25                 'id': '560070183650213889',
  26                 'ext': 'mp4',
  27                 'title': 'TwitterCard',
  28                 'thumbnail': 're:^https?://.*\.jpg$',
  29                 'duration': 30.033,
  30             }
  31         },
  32         {
  33             'url': 'https://twitter.com/i/cards/tfw/v1/623160978427936768',
  34             'md5': '7ee2a553b63d1bccba97fbed97d9e1c8',
  35             'info_dict': {
  36                 'id': '623160978427936768',
  37                 'ext': 'mp4',
  38                 'title': 'TwitterCard',
  39                 'thumbnail': 're:^https?://.*\.jpg',
  40                 'duration': 80.155,
  41             },
  42         },
  43         {
  44             'url': 'https://twitter.com/i/cards/tfw/v1/654001591733886977',
  45             'md5': 'b6f35e8b08a0bec6c8af77a2f4b3a814',
  46             'info_dict': {
  47                 'id': 'dq4Oj5quskI',
  48                 'ext': 'mp4',
  49                 'title': 'Ubuntu 11.10 Overview',
  50                 '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/',
  51                 'upload_date': '20111013',
  52                 'uploader': 'OMG! Ubuntu!',
  53                 'uploader_id': 'omgubuntu',
  54             },
  55             'add_ie': ['Youtube'],
  56         },
  57         {
  58             'url': 'https://twitter.com/i/cards/tfw/v1/665289828897005568',
  59             'md5': 'ab2745d0b0ce53319a534fccaa986439',
  60             'info_dict': {
  61                 'id': 'iBb2x00UVlv',
  62                 'ext': 'mp4',
  63                 'upload_date': '20151113',
  64                 'uploader_id': '1189339351084113920',
  65                 'uploader': '@ArsenalTerje',
  66                 'title': 'Vine by @ArsenalTerje',
  67             },
  68             'add_ie': ['Vine'],
  69         }
  70     ]
  71
  72     def _real_extract(self, url):
  73         video_id = self._match_id(url)
  74
  75         # Different formats served for different User-Agents
  76         USER_AGENTS = [
  77             'Mozilla/5.0 (X11; Linux x86_64; rv:10.0) Gecko/20150101 Firefox/20.0 (Chrome)',  # mp4
  78             'Mozilla/5.0 (Windows NT 5.2; WOW64; rv:38.0) Gecko/20100101 Firefox/38.0',  # webm
  79         ]
  80
  81         config = None
  82         formats = []
  83         for user_agent in USER_AGENTS:
  84             request = sanitized_Request(url)
  85             request.add_header('User-Agent', user_agent)
  86             webpage = self._download_webpage(request, video_id)
  87
  88             iframe_url = self._html_search_regex(
  89                 r'<iframe[^>]+src="((?:https?:)?//(?:www.youtube.com/embed/[^"]+|(?:www\.)?vine\.co/v/\w+/card))"',
  90                 webpage, 'video iframe', default=None)
  91             if iframe_url:
  92                 return self.url_result(iframe_url)
  93
  94             config = self._parse_json(self._html_search_regex(
  95                 r'data-player-config="([^"]+)"', webpage, 'data player config'),
  96                 video_id)
  97             if 'playlist' not in config:
  98                 if 'vmapUrl' in config:
  99                     vmap_data = self._download_xml(config['vmapUrl'], video_id)
 100                     video_url = xpath_text(vmap_data, './/MediaFile').strip()
 101                     formats.append({
 102                         'url': video_url,
 103                     })
 104                     break   # same video regardless of UA
 105                 continue
 106
 107             video_url = config['playlist'][0]['source']
 108
 109             f = {
 110                 'url': video_url,
 111             }
 112
 113             m = re.search(r'/(?P<width>\d+)x(?P<height>\d+)/', video_url)
 114             if m:
 115                 f.update({
 116                     'width': int(m.group('width')),
 117                     'height': int(m.group('height')),
 118                 })
 119             formats.append(f)
 120         self._sort_formats(formats)
 121
 122         thumbnail = config.get('posterImageUrl')
 123         duration = float_or_none(config.get('duration'))
 124
 125         return {
 126             'id': video_id,
 127             'title': 'TwitterCard',
 128             'thumbnail': thumbnail,
 129             'duration': duration,
 130             'formats': formats,
 131         }
 132
 133
 134 class TwitterIE(InfoExtractor):
 135     IE_NAME = 'twitter'
 136     _VALID_URL = r'https?://(?:www\.|m\.|mobile\.)?twitter\.com/(?P<user_id>[^/]+)/status/(?P<id>\d+)'
 137     _TEMPLATE_URL = 'https://twitter.com/%s/status/%s'
 138
 139     _TESTS = [{
 140         'url': 'https://twitter.com/freethenipple/status/643211948184596480',
 141         'md5': 'db6612ec5d03355953c3ca9250c97e5e',
 142         'info_dict': {
 143             'id': '643211948184596480',
 144             'ext': 'mp4',
 145             'title': 'FREE THE NIPPLE - FTN supporters on Hollywood Blvd today!',
 146             'thumbnail': 're:^https?://.*\.jpg',
 147             'duration': 12.922,
 148             'description': 'FREE THE NIPPLE on Twitter: "FTN supporters on Hollywood Blvd today! http://t.co/c7jHH749xJ"',
 149             'uploader': 'FREE THE NIPPLE',
 150             'uploader_id': 'freethenipple',
 151         },
 152     }, {
 153         'url': 'https://twitter.com/giphz/status/657991469417025536/photo/1',
 154         'md5': 'f36dcd5fb92bf7057f155e7d927eeb42',
 155         'info_dict': {
 156             'id': '657991469417025536',
 157             'ext': 'mp4',
 158             'title': 'Gifs - tu vai cai tu vai cai tu nao eh capaz disso tu vai cai',
 159             'description': 'Gifs on Twitter: "tu vai cai tu vai cai tu nao eh capaz disso tu vai cai https://t.co/tM46VHFlO5"',
 160             'thumbnail': 're:^https?://.*\.png',
 161             'uploader': 'Gifs',
 162             'uploader_id': 'giphz',
 163         },
 164     }, {
 165         'url': 'https://twitter.com/starwars/status/665052190608723968',
 166         'md5': '39b7199856dee6cd4432e72c74bc69d4',
 167         'info_dict': {
 168             'id': '665052190608723968',
 169             'ext': 'mp4',
 170             'title': 'Star Wars - A new beginning is coming December 18. Watch the official 60 second #TV spot for #StarWars: #TheForceAwakens.',
 171             'description': 'Star Wars on Twitter: "A new beginning is coming December 18. Watch the official 60 second #TV spot for #StarWars: #TheForceAwakens."',
 172             'uploader_id': 'starwars',
 173             'uploader': 'Star Wars',
 174         },
 175     }]
 176
 177     def _real_extract(self, url):
 178         mobj = re.match(self._VALID_URL, url)
 179         user_id = mobj.group('user_id')
 180         twid = mobj.group('id')
 181
 182         webpage = self._download_webpage(self._TEMPLATE_URL % (user_id, twid), twid)
 183
 184         username = remove_end(self._og_search_title(webpage), ' on Twitter')
 185
 186         title = description = self._og_search_description(webpage).strip('').replace('\n', ' ').strip('“”')
 187
 188         # strip  'https -_t.co_BJYgOjSeGA' junk from filenames
 189         title = re.sub(r'\s+(https?://[^ ]+)', '', title)
 190
 191         info = {
 192             'uploader_id': user_id,
 193             'uploader': username,
 194             'webpage_url': url,
 195             'description': '%s on Twitter: "%s"' % (username, description),
 196             'title': username + ' - ' + title,
 197         }
 198
 199         card_id = self._search_regex(
 200             r'["\']/i/cards/tfw/v1/(\d+)', webpage, 'twitter card url', default=None)
 201         if card_id:
 202             card_url = 'https://twitter.com/i/cards/tfw/v1/' + card_id
 203             info.update({
 204                 '_type': 'url_transparent',
 205                 'ie_key': 'TwitterCard',
 206                 'url': card_url,
 207             })
 208             return info
 209
 210         mobj = re.search(r'''(?x)
 211             <video[^>]+class="animated-gif"[^>]+
 212                 (?:data-height="(?P<height>\d+)")?[^>]+
 213                 (?:data-width="(?P<width>\d+)")?[^>]+
 214                 (?:poster="(?P<poster>[^"]+)")?[^>]*>\s*
 215                 <source[^>]+video-src="(?P<url>[^"]+)"
 216         ''', webpage)
 217
 218         if mobj:
 219             info.update({
 220                 'id': twid,
 221                 'url': mobj.group('url'),
 222                 'height': int_or_none(mobj.group('height')),
 223                 'width': int_or_none(mobj.group('width')),
 224                 'thumbnail': mobj.group('poster'),
 225             })
 226             return info
 227
 228         raise ExtractorError('There\'s not video in this tweet.')