Raphaël G. Git Repositories - youtubedl/blob - 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     qualities,
  13 )
  14
  15
  16 class Revision3IE(InfoExtractor):
  17     _VALID_URL = r'https?://(?:www\.)?(?P<domain>(?:revision3|testtube|animalist)\.com)/(?P<id>[^/]+(?:/[^/?#]+)?)'
  18     _TESTS = [{
  19         'url': 'http://www.revision3.com/technobuffalo/5-google-predictions-for-2016',
  20         'md5': 'd94a72d85d0a829766de4deb8daaf7df',
  21         'info_dict': {
  22             'id': '73034',
  23             'display_id': 'technobuffalo/5-google-predictions-for-2016',
  24             'ext': 'webm',
  25             'title': '5 Google Predictions for 2016',
  26             'description': 'Google had a great 2015, but it\'s already time to look ahead. Here are our five predictions for 2016.',
  27             'upload_date': '20151228',
  28             'timestamp': 1451325600,
  29             'duration': 187,
  30             'uploader': 'TechnoBuffalo',
  31             'uploader_id': 'technobuffalo',
  32         }
  33     }, {
  34         'url': 'http://testtube.com/brainstuff',
  35         'info_dict': {
  36             'id': '251',
  37             'title': 'BrainStuff',
  38             '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.',
  39         },
  40         'playlist_mincount': 93,
  41     }, {
  42         'url': 'https://testtube.com/dnews/5-weird-ways-plants-can-eat-animals?utm_source=FB&utm_medium=DNews&utm_campaign=DNewsSocial',
  43         'info_dict': {
  44             'id': '60163',
  45             'display_id': 'dnews/5-weird-ways-plants-can-eat-animals',
  46             'duration': 275,
  47             'ext': 'webm',
  48             'title': '5 Weird Ways Plants Can Eat Animals',
  49             'description': 'Why have some plants evolved to eat meat?',
  50             'upload_date': '20150120',
  51             'timestamp': 1421763300,
  52             'uploader': 'DNews',
  53             'uploader_id': 'dnews',
  54         },
  55     }]
  56     _PAGE_DATA_TEMPLATE = 'http://www.%s/apiProxy/ddn/%s?domain=%s'
  57     _API_KEY = 'ba9c741bce1b9d8e3defcc22193f3651b8867e62'
  58
  59     def _real_extract(self, url):
  60         domain, display_id = re.match(self._VALID_URL, url).groups()
  61         page_info = self._download_json(
  62             self._PAGE_DATA_TEMPLATE % (domain, display_id, domain), display_id)
  63
  64         if page_info['data']['type'] == 'episode':
  65             episode_data = page_info['data']
  66             video_id = compat_str(episode_data['video']['data']['id'])
  67             video_data = self._download_json(
  68                 'http://revision3.com/api/getPlaylist.json?api_key=%s&codecs=h264,vp8,theora&video_id=%s' % (self._API_KEY, video_id),
  69                 video_id)['items'][0]
  70
  71             formats = []
  72             for vcodec, media in video_data['media'].items():
  73                 for quality_id, quality in media.items():
  74                     if quality_id == 'hls':
  75                         formats.extend(self._extract_m3u8_formats(
  76                             quality['url'], video_id, 'mp4',
  77                             'm3u8_native', m3u8_id='hls', fatal=False))
  78                     else:
  79                         formats.append({
  80                             'url': quality['url'],
  81                             'format_id': '%s-%s' % (vcodec, quality_id),
  82                             'tbr': int_or_none(quality.get('bitrate')),
  83                             'vcodec': vcodec,
  84                         })
  85             self._sort_formats(formats)
  86
  87             preference = qualities(['mini', 'small', 'medium', 'large'])
  88             thumbnails = [{
  89                 'url': image_url,
  90                 'id': image_id,
  91                 'preference': preference(image_id)
  92             } for image_id, image_url in video_data.get('images', {}).items()]
  93
  94             return {
  95                 'id': video_id,
  96                 'display_id': display_id,
  97                 'title': unescapeHTML(video_data['title']),
  98                 'description': unescapeHTML(video_data.get('summary')),
  99                 'timestamp': parse_iso8601(episode_data.get('publishTime'), ' '),
 100                 'author': episode_data.get('author'),
 101                 'uploader': video_data.get('show', {}).get('name'),
 102                 'uploader_id': video_data.get('show', {}).get('slug'),
 103                 'duration': int_or_none(video_data.get('duration')),
 104                 'thumbnails': thumbnails,
 105                 'formats': formats,
 106             }
 107         else:
 108             show_data = page_info['show']['data']
 109             episodes_data = page_info['episodes']['data']
 110             num_episodes = page_info['meta']['totalEpisodes']
 111             processed_episodes = 0
 112             entries = []
 113             page_num = 1
 114             while True:
 115                 entries.extend([self.url_result(
 116                     'http://%s/%s/%s' % (domain, display_id, episode['slug'])) for episode in episodes_data])
 117                 processed_episodes += len(episodes_data)
 118                 if processed_episodes == num_episodes:
 119                     break
 120                 page_num += 1
 121                 episodes_data = self._download_json(self._PAGE_DATA_TEMPLATE % (
 122                     domain, display_id + '/' + compat_str(page_num), domain),
 123                     display_id)['episodes']['data']
 124
 125             return self.playlist_result(
 126                 entries, compat_str(show_data['id']),
 127                 show_data.get('name'), show_data.get('summary'))