youtube_dl/extractor/comcarcoff.py

   1 # encoding: utf-8
   2 from __future__ import unicode_literals
   3
   4 from .common import InfoExtractor
   5 from ..utils import (
   6     int_or_none,
   7     parse_duration,
   8     parse_iso8601,
   9 )
  10
  11
  12 class ComCarCoffIE(InfoExtractor):
  13     _VALID_URL = r'http://(?:www\.)?comediansincarsgettingcoffee\.com/(?P<id>[a-z0-9\-]*)'
  14     _TESTS = [{
  15         'url': 'http://comediansincarsgettingcoffee.com/miranda-sings-happy-thanksgiving-miranda/',
  16         'info_dict': {
  17             'id': 'miranda-sings-happy-thanksgiving-miranda',
  18             'ext': 'mp4',
  19             'upload_date': '20141127',
  20             'timestamp': 1417107600,
  21             'duration': 1232,
  22             'title': 'Happy Thanksgiving Miranda',
  23             'description': 'Jerry Seinfeld and his special guest Miranda Sings cruise around town in search of coffee, complaining and apologizing along the way.',
  24             'thumbnail': 'http://ccc.crackle.com/images/s5e4_thumb.jpg',
  25         },
  26         'params': {
  27             'skip_download': 'requires ffmpeg',
  28         }
  29     }]
  30
  31     def _real_extract(self, url):
  32         display_id = self._match_id(url)
  33         if not display_id:
  34             display_id = 'comediansincarsgettingcoffee.com'
  35         webpage = self._download_webpage(url, display_id)
  36
  37         full_data = self._parse_json(
  38             self._search_regex(
  39                 r'window\.app\s*=\s*({.+?});\n', webpage, 'full data json'),
  40             display_id)['videoData']
  41
  42         video_id = full_data['activeVideo']['video']
  43         video_data = full_data.get('videos', {}).get(video_id) or full_data['singleshots'][video_id]
  44         thumbnails = [{
  45             'url': video_data['images']['thumb'],
  46         }, {
  47             'url': video_data['images']['poster'],
  48         }]
  49         formats = self._extract_m3u8_formats(
  50             video_data['mediaUrl'], video_id, ext='mp4')
  51
  52         timestamp = int_or_none(video_data.get('pubDateTime')) or parse_iso8601(
  53             video_data.get('pubDate'))
  54         duration = int_or_none(video_data.get('durationSeconds')) or parse_duration(
  55             video_data.get('duration'))
  56
  57         return {
  58             'id': video_id,
  59             'display_id': display_id,
  60             'title': video_data['title'],
  61             'description': video_data.get('description'),
  62             'timestamp': timestamp,
  63             'duration': duration,
  64             'thumbnails': thumbnails,
  65             'formats': formats,
  66             'webpage_url': 'http://comediansincarsgettingcoffee.com/%s' % (video_data.get('urlSlug', video_data.get('slug'))),
  67         }