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     _TESTS = [
  24         {
  25             'url': 'http://www.pbs.org/tpt/constitution-usa-peter-sagal/watch/a-more-perfect-union/',
  26             'md5': 'ce1888486f0908d555a8093cac9a7362',
  27             'info_dict': {
  28                 'id': '2365006249',
  29                 'ext': 'mp4',
  30                 'title': 'A More Perfect Union',
  31                 'description': 'md5:ba0c207295339c8d6eced00b7c363c6a',
  32                 'duration': 3190,
  33             },
  34         },
  35         {
  36             'url': 'http://www.pbs.org/wgbh/pages/frontline/losing-iraq/',
  37             'md5': '143c98aa54a346738a3d78f54c925321',
  38             'info_dict': {
  39                 'id': '2365297690',
  40                 'ext': 'mp4',
  41                 'title': 'Losing Iraq',
  42                 'description': 'md5:f5bfbefadf421e8bb8647602011caf8e',
  43                 'duration': 5050,
  44             },
  45         },
  46         {
  47             'url': 'http://www.pbs.org/newshour/bb/education-jan-june12-cyberschools_02-23/',
  48             'md5': 'b19856d7f5351b17a5ab1dc6a64be633',
  49             'info_dict': {
  50                 'id': '2201174722',
  51                 'ext': 'mp4',
  52                 'title': 'Cyber Schools Gain Popularity, but Quality Questions Persist',
  53                 'description': 'md5:5871c15cba347c1b3d28ac47a73c7c28',
  54                 'duration': 801,
  55             },
  56         },
  57         {
  58             'url': 'http://www.pbs.org/wnet/gperf/dudamel-conducts-verdi-requiem-hollywood-bowl-full-episode/3374/',
  59             'md5': 'c62859342be2a0358d6c9eb306595978',
  60             'info_dict': {
  61                 'id': '2365297708',
  62                 'ext': 'mp4',
  63                 'description': 'md5:68d87ef760660eb564455eb30ca464fe',
  64                 'title': 'Dudamel Conducts Verdi Requiem at the Hollywood Bowl - Full',
  65                 'duration': 6559,
  66                 'thumbnail': 're:^https?://.*\.jpg$',
  67             }
  68         }
  69     ]
  70
  71     def _extract_ids(self, url):
  72         mobj = re.match(self._VALID_URL, url)
  73
  74         presumptive_id = mobj.group('presumptive_id')
  75         display_id = presumptive_id
  76         if presumptive_id:
  77             webpage = self._download_webpage(url, display_id)
  78
  79             MEDIA_ID_REGEXES = [
  80                 r"div\s*:\s*'videoembed'\s*,\s*mediaid\s*:\s*'(\d+)'",  # frontline video embed
  81                 r'class="coveplayerid">([^<]+)<',                       # coveplayer
  82             ]
  83
  84             media_id = self._search_regex(
  85                 MEDIA_ID_REGEXES, webpage, 'media ID', fatal=False, default=None)
  86             if media_id:
  87                 return media_id, presumptive_id
  88
  89             url = self._search_regex(
  90                 r'<iframe\s+(?:class|id)=["\']partnerPlayer["\'].*?\s+src=["\'](.*?)["\']>',
  91                 webpage, 'player URL')
  92             mobj = re.match(self._VALID_URL, url)
  93
  94         player_id = mobj.group('player_id')
  95         if not display_id:
  96             display_id = player_id
  97         if player_id:
  98             player_page = self._download_webpage(
  99                 url, display_id, note='Downloading player page',
 100                 errnote='Could not download player page')
 101             video_id = self._search_regex(
 102                 r'<div\s+id="video_([0-9]+)"', player_page, 'video ID')
 103         else:
 104             video_id = mobj.group('id')
 105             display_id = video_id
 106
 107         return video_id, display_id
 108
 109     def _real_extract(self, url):
 110         video_id, display_id = self._extract_ids(url)
 111
 112         info_url = 'http://video.pbs.org/videoInfo/%s?format=json' % video_id
 113         info = self._download_json(info_url, display_id)
 114
 115         rating_str = info.get('rating')
 116         if rating_str is not None:
 117             rating_str = rating_str.rpartition('-')[2]
 118         age_limit = US_RATINGS.get(rating_str)
 119
 120         return {
 121             'id': video_id,
 122             'title': info['title'],
 123             'url': info['alternate_encoding']['url'],
 124             'ext': 'mp4',
 125             'description': info['program'].get('description'),
 126             'thumbnail': info.get('image_url'),
 127             'duration': info.get('duration'),
 128             'age_limit': age_limit,
 129         }