_ Git - youtube-dl/blob - youtube_dl/extractor/pbs.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4 import json
   5
   6 from .common import InfoExtractor
   7
   8
   9 class PBSIE(InfoExtractor):
  10     _VALID_URL = r'''(?x)https?://
  11         (?:
  12             # Direct video URL
  13             video\.pbs\.org/video/(?P<id>[0-9]+)/? |
  14             # Article with embedded player
  15            (?:www\.)?pbs\.org/(?:[^/]+/){2,5}(?P<presumptive_id>[^/]+)/?(?:$|[?\#]) |
  16            # Player
  17            video\.pbs\.org/partnerplayer/(?P<player_id>[^/]+)/
  18         )
  19     '''
  20
  21     _TEST = {
  22         'url': 'http://www.pbs.org/tpt/constitution-usa-peter-sagal/watch/a-more-perfect-union/',
  23         'md5': 'ce1888486f0908d555a8093cac9a7362',
  24         'info_dict': {
  25             'id': '2365006249',
  26             'ext': 'mp4',
  27             'title': 'A More Perfect Union',
  28             'description': 'md5:ba0c207295339c8d6eced00b7c363c6a',
  29             'duration': 3190,
  30         },
  31     }
  32
  33     def _real_extract(self, url):
  34         mobj = re.match(self._VALID_URL, url)
  35
  36         presumptive_id = mobj.group('presumptive_id')
  37         display_id = presumptive_id
  38         if presumptive_id:
  39             webpage = self._download_webpage(url, display_id)
  40             url = self._search_regex(
  41                 r'<iframe\s+id=["\']partnerPlayer["\'].*?\s+src=["\'](.*?)["\']>',
  42                 webpage, 'player URL')
  43             mobj = re.match(self._VALID_URL, url)
  44
  45         player_id = mobj.group('player_id')
  46         if not display_id:
  47             display_id = player_id
  48         if player_id:
  49             player_page = self._download_webpage(
  50                 url, display_id, note='Downloading player page',
  51                 errnote='Could not download player page')
  52             video_id = self._search_regex(
  53                 r'<div\s+id="video_([0-9]+)"', player_page, 'video ID')
  54         else:
  55             video_id = mobj.group('id')
  56             display_id = video_id
  57
  58         info_url = 'http://video.pbs.org/videoInfo/%s?format=json' % video_id
  59         info = self._download_json(info_url, display_id)
  60
  61         return {
  62             'id': video_id,
  63             'title': info['title'],
  64             'url': info['alternate_encoding']['url'],
  65             'ext': 'mp4',
  66             'description': info['program'].get('description'),
  67             'thumbnail': info.get('image_url'),
  68             'duration': info.get('duration'),
  69         }