_ Git - youtube-dl/blob - youtube_dl/extractor/comcarcoff.py

   1 # encoding: utf-8
   2 import json
   3
   4 from .common import InfoExtractor
   5 from ..utils import parse_iso8601
   6
   7
   8 class ComCarCoffIE(InfoExtractor):
   9     _VALID_URL = r'http://(?:www\.)?comediansincarsgettingcoffee\.com/(?P<id>[a-z0-9\-]*)'
  10     _TESTS = [{
  11         'url': 'http://comediansincarsgettingcoffee.com/miranda-sings-happy-thanksgiving-miranda/',
  12         'info_dict': {
  13             'id': 'miranda-sings-happy-thanksgiving-miranda',
  14             'ext': 'mp4',
  15             'upload_date': '20141127',
  16             'timestamp': 1417107600,
  17             'title': 'Happy Thanksgiving Miranda',
  18             'description': 'Jerry Seinfeld and his special guest Miranda Sings cruise around town in search of coffee, complaining and apologizing along the way.',
  19             'thumbnail': 'http://ccc.crackle.com/images/s5e4_thumb.jpg',
  20         },
  21         'params': {
  22             'skip_download': 'requires ffmpeg',
  23         }
  24     }]
  25
  26     def _real_extract(self, url):
  27         display_id = self._match_id(url)
  28         if not display_id:
  29             display_id = 'comediansincarsgettingcoffee.com'
  30         webpage = self._download_webpage(url, display_id)
  31
  32         full_data = json.loads(self._search_regex(
  33             r'<script type="application/json" id="videoData">(?P<json>.+?)</script>',
  34             webpage, 'full data json'))
  35
  36         video_id = full_data['activeVideo']['video']
  37         video_data = full_data['videos'][video_id]
  38         thumbnails = [{
  39             'url': video_data['images']['thumb'],
  40         }, {
  41             'url': video_data['images']['poster'],
  42         }]
  43         formats = self._extract_m3u8_formats(
  44             video_data['mediaUrl'], video_id, ext='mp4')
  45
  46         return {
  47             'id': video_id,
  48             'display_id': display_id,
  49             'title': video_data['title'],
  50             'description': video_data.get('description'),
  51             'timestamp': parse_iso8601(video_data.get('pubDate')),
  52             'thumbnails': thumbnails,
  53             'formats': formats,
  54         }