youtube_dl/extractor/revision3.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_str
   8 from ..utils import (
   9     int_or_none,
  10     parse_iso8601,
  11     unescapeHTML,
  12 )
  13
  14
  15 class Revision3IE(InfoExtractor):
  16     _VALID_URL = r'https?://(?:www\.)?(?P<domain>(?:revision3|testtube|animalist)\.com)/(?P<id>[^/]+(?:/[^/?#]+)?)'
  17     _TESTS = [{
  18         'url': 'http://www.revision3.com/technobuffalo/5-google-predictions-for-2016',
  19         'md5': 'd94a72d85d0a829766de4deb8daaf7df',
  20         'info_dict': {
  21             'id': '73034',
  22             'ext': 'webm',
  23             'title': '5 Google Predictions for 2016',
  24             'description': 'Google had a great 2015, but it\'s already time to look ahead. Here are our five predictions for 2016.',
  25             'upload_date': '20151228',
  26             'timestamp': 1451325600,
  27             'duration': 187,
  28         }
  29     }, {
  30         'url': 'http://testtube.com/brainstuff',
  31         'info_dict': {
  32             'id': '251',
  33             'title': 'BrainStuff',
  34             'description': 'Whether the topic is popcorn or particle physics, you can count on the HowStuffWorks team to explore-and explain-the everyday science in the world around us on BrainStuff.',
  35         },
  36         'playlist_mincount': 93,
  37     }]
  38     _PAGE_DATA_TEMPLATE = 'http://www.%s/apiProxy/ddn/%s?domain=%s'
  39     _API_KEY = 'ba9c741bce1b9d8e3defcc22193f3651b8867e62'
  40
  41     def _real_extract(self, url):
  42         domain, display_id = re.match(self._VALID_URL, url).groups()
  43         page_info = self._download_json(
  44             self._PAGE_DATA_TEMPLATE % (domain, display_id, domain), display_id)
  45
  46         if page_info['data']['type'] == 'episode':
  47             episode_data = page_info['data']
  48             video_id = compat_str(episode_data['video']['data']['id'])
  49             video_data = self._download_json(
  50                 'http://revision3.com/api/getPlaylist.json?api_key=%s&codecs=h264,vp8,theora&video_id=%s' % (self._API_KEY, video_id),
  51                 video_id)['items'][0]
  52
  53             formats = []
  54             for media_type, media in video_data['media'].items():
  55                 for quality_id, quality in media.items():
  56                     if quality_id == 'hls':
  57                         formats.extend(self._extract_m3u8_formats(
  58                             quality['url'], video_id, 'mp4',
  59                             'm3u8_native', m3u8_id='hls', fatal=False))
  60                     else:
  61                         formats.append({
  62                             'url': quality['url'],
  63                             'format_id': '%s-%s' % (media_type, quality_id),
  64                             'tbr': int_or_none(quality.get('bitrate')),
  65                         })
  66             self._sort_formats(formats)
  67
  68             thumbnails = [{
  69                 'url': image_url,
  70                 'id': image_id,
  71             } for image_id, image_url in video_data.get('images', {}).items()]
  72
  73             return {
  74                 'id': video_id,
  75                 'title': unescapeHTML(video_data['title']),
  76                 'description': unescapeHTML(video_data.get('summary')),
  77                 'timestamp': parse_iso8601(episode_data.get('publishTime'), ' '),
  78                 'author': episode_data.get('author'),
  79                 'duration': int_or_none(video_data.get('duration')),
  80                 'thumbnails': thumbnails,
  81                 'formats': formats,
  82             }
  83         else:
  84             show_data = page_info['show']['data']
  85             episodes_data = page_info['episodes']['data']
  86             num_episodes = page_info['meta']['totalEpisodes']
  87             processed_episodes = 0
  88             entries = []
  89             page_num = 1
  90             while True:
  91                 entries.extend([self.url_result(
  92                     url + '/%s' % episode['slug']) for episode in episodes_data])
  93                 processed_episodes += len(episodes_data)
  94                 if processed_episodes == num_episodes:
  95                     break
  96                 page_num += 1
  97                 episodes_data = self._download_json(self._PAGE_DATA_TEMPLATE % (
  98                     domain, display_id + '/' + compat_str(page_num), domain),
  99                     display_id)['episodes']['data']
 100
 101             return self.playlist_result(
 102                 entries, compat_str(show_data['id']),
 103                 show_data.get('name'), show_data.get('summary'))