youtube_dl/extractor/ted.py

   1 from __future__ import unicode_literals
   2
   3 import json
   4 import re
   5
   6 from .subtitles import SubtitlesInfoExtractor
   7
   8 from ..utils import (
   9     compat_str,
  10 )
  11
  12
  13 class TEDIE(SubtitlesInfoExtractor):
  14     _VALID_URL = r'''(?x)
  15         (?P<proto>https?://)
  16         (?P<type>www|embed)(?P<urlmain>\.ted\.com/
  17         (
  18             (?P<type_playlist>playlists(?:/\d+)?) # We have a playlist
  19             |
  20             ((?P<type_talk>talks)) # We have a simple talk
  21         )
  22         (/lang/(.*?))? # The url may contain the language
  23         /(?P<name>\w+) # Here goes the name and then ".html"
  24         .*)$
  25         '''
  26     _TEST = {
  27         'url': 'http://www.ted.com/talks/dan_dennett_on_our_consciousness.html',
  28         'md5': '4ea1dada91e4174b53dac2bb8ace429d',
  29         'info_dict': {
  30             'id': '102',
  31             'ext': 'mp4',
  32             'title': 'The illusion of consciousness',
  33             'description': ('Philosopher Dan Dennett makes a compelling '
  34                 'argument that not only don\'t we understand our own '
  35                 'consciousness, but that half the time our brains are '
  36                 'actively fooling us.'),
  37             'uploader': 'Dan Dennett',
  38         }
  39     }
  40
  41     _FORMATS_PREFERENCE = {
  42         'low': 1,
  43         'medium': 2,
  44         'high': 3,
  45     }
  46
  47     def _extract_info(self, webpage):
  48         info_json = self._search_regex(r'q\("\w+.init",({.+})\)</script>',
  49             webpage, 'info json')
  50         return json.loads(info_json)
  51
  52     def _real_extract(self, url):
  53         m = re.match(self._VALID_URL, url, re.VERBOSE)
  54         if m.group('type') == 'embed':
  55             desktop_url = m.group('proto') + 'www' + m.group('urlmain')
  56             return self.url_result(desktop_url, 'TED')
  57         name = m.group('name')
  58         if m.group('type_talk'):
  59             return self._talk_info(url, name)
  60         else:
  61             return self._playlist_videos_info(url, name)
  62
  63     def _playlist_videos_info(self, url, name):
  64         '''Returns the videos of the playlist'''
  65
  66         webpage = self._download_webpage(url, name,
  67             'Downloading playlist webpage')
  68         info = self._extract_info(webpage)
  69         playlist_info = info['playlist']
  70
  71         playlist_entries = [
  72             self.url_result(u'http://www.ted.com/talks/' + talk['slug'], self.ie_key())
  73             for talk in info['talks']
  74         ]
  75         return self.playlist_result(
  76             playlist_entries,
  77             playlist_id=compat_str(playlist_info['id']),
  78             playlist_title=playlist_info['title'])
  79
  80     def _talk_info(self, url, video_name):
  81         webpage = self._download_webpage(url, video_name)
  82         self.report_extraction(video_name)
  83
  84         talk_info = self._extract_info(webpage)['talks'][0]
  85
  86         formats = [{
  87             'ext': 'mp4',
  88             'url': format_url,
  89             'format_id': format_id,
  90             'format': format_id,
  91             'preference': self._FORMATS_PREFERENCE.get(format_id, -1),
  92         } for (format_id, format_url) in talk_info['nativeDownloads'].items()]
  93         self._sort_formats(formats)
  94
  95         video_id = compat_str(talk_info['id'])
  96         # subtitles
  97         video_subtitles = self.extract_subtitles(video_id, talk_info)
  98         if self._downloader.params.get('listsubtitles', False):
  99             self._list_available_subtitles(video_id, talk_info)
 100             return
 101
 102         thumbnail = talk_info['thumb']
 103         if not thumbnail.startswith('http'):
 104             thumbnail = 'http://' + thumbnail
 105         return {
 106             'id': video_id,
 107             'title': talk_info['title'],
 108             'uploader': talk_info['speaker'],
 109             'thumbnail': thumbnail,
 110             'description': self._og_search_description(webpage),
 111             'subtitles': video_subtitles,
 112             'formats': formats,
 113         }
 114
 115     def _get_available_subtitles(self, video_id, talk_info):
 116         languages = [lang['languageCode'] for lang in talk_info.get('languages', [])]
 117         if languages:
 118             sub_lang_list = {}
 119             for l in languages:
 120                 url = 'http://www.ted.com/talks/subtitles/id/%s/lang/%s/format/srt' % (video_id, l)
 121                 sub_lang_list[l] = url
 122             return sub_lang_list
 123         else:
 124             self._downloader.report_warning(u'video doesn\'t have subtitles')
 125             return {}