_ Git - youtube-dl/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 ..compat import compat_urllib_request
   8 from ..utils import (
   9     float_or_none,
  10     unescapeHTML,
  11     xpath_text,
  12     remove_end,
  13 )
  14
  15
  16 class TwitterCardIE(InfoExtractor):
  17     IE_NAME = 'twitter:card'
  18     _VALID_URL = r'https?://(?:www\.)?twitter\.com/i/cards/tfw/v1/(?P<id>\d+)'
  19     _TESTS = [
  20         {
  21             'url': 'https://twitter.com/i/cards/tfw/v1/560070183650213889',
  22             'md5': '7d2f6b4d2eb841a7ccc893d479bfceb4',
  23             'info_dict': {
  24                 'id': '560070183650213889',
  25                 'ext': 'mp4',
  26                 'title': 'TwitterCard',
  27                 'thumbnail': 're:^https?://.*\.jpg$',
  28                 'duration': 30.033,
  29             }
  30         },
  31         {
  32             'url': 'https://twitter.com/i/cards/tfw/v1/623160978427936768',
  33             'md5': '7ee2a553b63d1bccba97fbed97d9e1c8',
  34             'info_dict': {
  35                 'id': '623160978427936768',
  36                 'ext': 'mp4',
  37                 'title': 'TwitterCard',
  38                 'thumbnail': 're:^https?://.*\.jpg',
  39                 'duration': 80.155,
  40             },
  41         }
  42     ]
  43
  44     def _real_extract(self, url):
  45         video_id = self._match_id(url)
  46
  47         # Different formats served for different User-Agents
  48         USER_AGENTS = [
  49             'Mozilla/5.0 (X11; Linux x86_64; rv:10.0) Gecko/20150101 Firefox/20.0 (Chrome)',  # mp4
  50             'Mozilla/5.0 (Windows NT 5.2; WOW64; rv:38.0) Gecko/20100101 Firefox/38.0',  # webm
  51         ]
  52
  53         config = None
  54         formats = []
  55         for user_agent in USER_AGENTS:
  56             request = compat_urllib_request.Request(url)
  57             request.add_header('User-Agent', user_agent)
  58             webpage = self._download_webpage(request, video_id)
  59
  60             config = self._parse_json(
  61                 unescapeHTML(self._search_regex(
  62                     r'data-player-config="([^"]+)"', webpage, 'data player config')),
  63                 video_id)
  64             if 'playlist' not in config:
  65                 if 'vmapUrl' in config:
  66                     vmap_data = self._download_xml(config['vmapUrl'], video_id)
  67                     video_url = xpath_text(vmap_data, './/MediaFile').strip()
  68                     formats.append({
  69                         'url': video_url,
  70                     })
  71                     break   # same video regardless of UA
  72                 continue
  73
  74             video_url = config['playlist'][0]['source']
  75
  76             f = {
  77                 'url': video_url,
  78             }
  79
  80             m = re.search(r'/(?P<width>\d+)x(?P<height>\d+)/', video_url)
  81             if m:
  82                 f.update({
  83                     'width': int(m.group('width')),
  84                     'height': int(m.group('height')),
  85                 })
  86             formats.append(f)
  87         self._sort_formats(formats)
  88
  89         thumbnail = config.get('posterImageUrl')
  90         duration = float_or_none(config.get('duration'))
  91
  92         return {
  93             'id': video_id,
  94             'title': 'TwitterCard',
  95             'thumbnail': thumbnail,
  96             'duration': duration,
  97             'formats': formats,
  98         }
  99
 100
 101 class TwitterIE(InfoExtractor):
 102     IE_NAME = 'twitter'
 103     _VALID_URL = r'https?://(?:www\.|m\.|mobile\.)?twitter\.com/(?P<user_id>[^/]+)/status/(?P<id>\d+)'
 104     _TEMPLATE_URL = 'https://twitter.com/%s/status/%s'
 105
 106     _TEST = {
 107         'url': 'https://twitter.com/freethenipple/status/643211948184596480',
 108         'md5': '31cd83a116fc41f99ae3d909d4caf6a0',
 109         'info_dict': {
 110             'id': '643211948184596480',
 111             'ext': 'mp4',
 112             'title': 'FREE THE NIPPLE - FTN supporters on Hollywood Blvd today!',
 113             'thumbnail': 're:^https?://.*\.jpg',
 114             'duration': 12.922,
 115             'description': 'FREE THE NIPPLE on Twitter: "FTN supporters on Hollywood Blvd today! http://t.co/c7jHH749xJ"',
 116             'uploader': 'FREE THE NIPPLE',
 117             'uploader_id': 'freethenipple',
 118         },
 119     }
 120
 121     def _real_extract(self, url):
 122         mobj = re.match(self._VALID_URL, url)
 123         user_id = mobj.group('user_id')
 124         twid = mobj.group('id')
 125
 126         webpage = self._download_webpage(self._TEMPLATE_URL % (user_id, twid), twid)
 127
 128         username = remove_end(self._og_search_title(webpage), ' on Twitter')
 129
 130         title = self._og_search_description(webpage).strip('').replace('\n', ' ')
 131
 132         # strip  'https -_t.co_BJYgOjSeGA' junk from filenames
 133         mobj = re.match(r'“(.*)\s+(http://[^ ]+)”', title)
 134         title, short_url = mobj.groups()
 135
 136         card_id = self._search_regex(
 137             r'["\']/i/cards/tfw/v1/(\d+)', webpage, 'twitter card url')
 138         card_url = 'https://twitter.com/i/cards/tfw/v1/' + card_id
 139
 140         return {
 141             '_type': 'url_transparent',
 142             'ie_key': 'TwitterCard',
 143             'uploader_id': user_id,
 144             'uploader': username,
 145             'url': card_url,
 146             'webpage_url': url,
 147             'description': '%s on Twitter: "%s %s"' % (username, title, short_url),
 148             'title': username + ' - ' + title,
 149         }