youtube_dl/extractor/webofstories.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..utils import int_or_none
   8
   9
  10 class WebOfStoriesIE(InfoExtractor):
  11     _VALID_URL = r'https?://(?:www\.)?webofstories\.com/play/(?:[^/]+/)?(?P<id>[0-9]+)'
  12     _VIDEO_DOMAIN = 'http://eu-mobile.webofstories.com/'
  13     _GREAT_LIFE_STREAMER = 'rtmp://eu-cdn1.webofstories.com/cfx/st/'
  14     _USER_STREAMER = 'rtmp://eu-users.webofstories.com/cfx/st/'
  15     _TESTS = [{
  16         'url': 'http://www.webofstories.com/play/hans.bethe/71',
  17         'md5': '373e4dd915f60cfe3116322642ddf364',
  18         'info_dict': {
  19             'id': '4536',
  20             'ext': 'mp4',
  21             'title': 'The temperature of the sun',
  22             'thumbnail': 're:^https?://.*\.jpg$',
  23             'description': 'Hans Bethe talks about calculating the temperature of the sun',
  24             'duration': 238,
  25         }
  26     }, {
  27         'url': 'http://www.webofstories.com/play/55908',
  28         'md5': '2985a698e1fe3211022422c4b5ed962c',
  29         'info_dict': {
  30             'id': '55908',
  31             'ext': 'mp4',
  32             'title': 'The story of Gemmata obscuriglobus',
  33             'thumbnail': 're:^https?://.*\.jpg$',
  34             'description': 'Planctomycete talks about The story of Gemmata obscuriglobus',
  35             'duration': 169,
  36         },
  37         'skip': 'notfound',
  38     }, {
  39         # malformed og:title meta
  40         'url': 'http://www.webofstories.com/play/54215?o=MS',
  41         'info_dict': {
  42             'id': '54215',
  43             'ext': 'mp4',
  44             'title': '"A Leg to Stand On"',
  45             'thumbnail': 're:^https?://.*\.jpg$',
  46             'description': 'Oliver Sacks talks about the death and resurrection of a limb',
  47             'duration': 97,
  48         },
  49         'params': {
  50             'skip_download': True,
  51         },
  52     }]
  53
  54     def _real_extract(self, url):
  55         video_id = self._match_id(url)
  56
  57         webpage = self._download_webpage(url, video_id)
  58         # Sometimes og:title meta is malformed
  59         title = self._og_search_title(webpage, default=None) or self._html_search_regex(
  60             r'(?s)<strong>Title:\s*</strong>(.+?)<', webpage, 'title')
  61         description = self._html_search_meta('description', webpage)
  62         thumbnail = self._og_search_thumbnail(webpage)
  63
  64         embed_params = [s.strip(" \r\n\t'") for s in self._search_regex(
  65             r'(?s)\$\("#embedCode"\).html\(getEmbedCode\((.*?)\)',
  66             webpage, 'embed params').split(',')]
  67
  68         (
  69             _, speaker_id, story_id, story_duration,
  70             speaker_type, great_life, _thumbnail, _has_subtitles,
  71             story_filename, _story_order) = embed_params
  72
  73         is_great_life_series = great_life == 'true'
  74         duration = int_or_none(story_duration)
  75
  76         # URL building, see: http://www.webofstories.com/scripts/player.js
  77         ms_prefix = ''
  78         if speaker_type.lower() == 'ms':
  79             ms_prefix = 'mini_sites/'
  80
  81         if is_great_life_series:
  82             mp4_url = '{0:}lives/{1:}/{2:}.mp4'.format(
  83                 self._VIDEO_DOMAIN, speaker_id, story_filename)
  84             rtmp_ext = 'flv'
  85             streamer = self._GREAT_LIFE_STREAMER
  86             play_path = 'stories/{0:}/{1:}'.format(
  87                 speaker_id, story_filename)
  88         else:
  89             mp4_url = '{0:}{1:}{2:}/{3:}.mp4'.format(
  90                 self._VIDEO_DOMAIN, ms_prefix, speaker_id, story_filename)
  91             rtmp_ext = 'mp4'
  92             streamer = self._USER_STREAMER
  93             play_path = 'mp4:{0:}{1:}/{2}.mp4'.format(
  94                 ms_prefix, speaker_id, story_filename)
  95
  96         formats = [{
  97             'format_id': 'mp4_sd',
  98             'url': mp4_url,
  99         }, {
 100             'format_id': 'rtmp_sd',
 101             'page_url': url,
 102             'url': streamer,
 103             'ext': rtmp_ext,
 104             'play_path': play_path,
 105         }]
 106
 107         self._sort_formats(formats)
 108
 109         return {
 110             'id': story_id,
 111             'title': title,
 112             'formats': formats,
 113             'thumbnail': thumbnail,
 114             'description': description,
 115             'duration': duration,
 116         }
 117
 118
 119 class WebOfStoriesPlaylistIE(InfoExtractor):
 120     _VALID_URL = r'https?://(?:www\.)?webofstories\.com/playAll/(?P<id>[^/]+)'
 121     _TEST = {
 122         'url': 'http://www.webofstories.com/playAll/donald.knuth',
 123         'info_dict': {
 124             'id': 'donald.knuth',
 125             'title': 'Donald Knuth (Scientist)',
 126         },
 127         'playlist_mincount': 97,
 128     }
 129
 130     def _real_extract(self, url):
 131         playlist_id = self._match_id(url)
 132
 133         webpage = self._download_webpage(url, playlist_id)
 134
 135         entries = [
 136             self.url_result('http://www.webofstories.com/play/%s' % video_number, 'WebOfStories')
 137             for video_number in set(re.findall('href="/playAll/%s\?sId=(\d+)"' % playlist_id, webpage))
 138         ]
 139
 140         title = self._search_regex(
 141             r'<div id="speakerName">\s*<span>([^<]+)</span>',
 142             webpage, 'speaker', default=None)
 143         if title:
 144             field = self._search_regex(
 145                 r'<span id="primaryField">([^<]+)</span>',
 146                 webpage, 'field', default=None)
 147             if field:
 148                 title += ' (%s)' % field
 149
 150         if not title:
 151             title = self._search_regex(
 152                 r'<title>Play\s+all\s+stories\s*-\s*([^<]+)\s*-\s*Web\s+of\s+Stories</title>',
 153                 webpage, 'title')
 154
 155         return self.playlist_result(entries, playlist_id, title)