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

   1 from __future__ import unicode_literals
   2
   3 import re
   4
   5 from .common import InfoExtractor
   6 from ..utils import (
   7     US_RATINGS,
   8 )
   9
  10
  11 class PBSIE(InfoExtractor):
  12     _VALID_URL = r'''(?x)https?://
  13         (?:
  14             # Direct video URL
  15             video\.pbs\.org/(?:viralplayer|video)/(?P<id>[0-9]+)/? |
  16             # Article with embedded player
  17            (?:www\.)?pbs\.org/(?:[^/]+/){2,5}(?P<presumptive_id>[^/]+)/?(?:$|[?\#]) |
  18            # Player
  19            video\.pbs\.org/(?:widget/)?partnerplayer/(?P<player_id>[^/]+)/
  20         )
  21     '''
  22
  23     _TEST = {
  24         'url': 'http://www.pbs.org/tpt/constitution-usa-peter-sagal/watch/a-more-perfect-union/',
  25         'md5': 'ce1888486f0908d555a8093cac9a7362',
  26         'info_dict': {
  27             'id': '2365006249',
  28             'ext': 'mp4',
  29             'title': 'A More Perfect Union',
  30             'description': 'md5:ba0c207295339c8d6eced00b7c363c6a',
  31             'duration': 3190,
  32         },
  33     }
  34
  35     def _extract_ids(self, url):
  36         mobj = re.match(self._VALID_URL, url)
  37
  38         presumptive_id = mobj.group('presumptive_id')
  39         display_id = presumptive_id
  40         if presumptive_id:
  41             webpage = self._download_webpage(url, display_id)
  42
  43             # frontline video embed
  44             media_id = self._search_regex(
  45                 r"div\s*:\s*'videoembed'\s*,\s*mediaid\s*:\s*'(\d+)'",
  46                 webpage, 'frontline video ID', fatal=False, default=None)
  47             if media_id:
  48                 return media_id, presumptive_id
  49
  50             url = self._search_regex(
  51                 r'<iframe\s+id=["\']partnerPlayer["\'].*?\s+src=["\'](.*?)["\']>',
  52                 webpage, 'player URL')
  53             mobj = re.match(self._VALID_URL, url)
  54
  55         player_id = mobj.group('player_id')
  56         if not display_id:
  57             display_id = player_id
  58         if player_id:
  59             player_page = self._download_webpage(
  60                 url, display_id, note='Downloading player page',
  61                 errnote='Could not download player page')
  62             video_id = self._search_regex(
  63                 r'<div\s+id="video_([0-9]+)"', player_page, 'video ID')
  64         else:
  65             video_id = mobj.group('id')
  66             display_id = video_id
  67
  68         return video_id, display_id
  69
  70     def _real_extract(self, url):
  71         video_id, display_id = self._extract_ids(url)
  72
  73         info_url = 'http://video.pbs.org/videoInfo/%s?format=json' % video_id
  74         info = self._download_json(info_url, display_id)
  75
  76         rating_str = info.get('rating')
  77         if rating_str is not None:
  78             rating_str = rating_str.rpartition('-')[2]
  79         age_limit = US_RATINGS.get(rating_str)
  80
  81         return {
  82             'id': video_id,
  83             'title': info['title'],
  84             'url': info['alternate_encoding']['url'],
  85             'ext': 'mp4',
  86             'description': info['program'].get('description'),
  87             'thumbnail': info.get('image_url'),
  88             'duration': info.get('duration'),
  89             'age_limit': age_limit,
  90         }