_ 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     ExtractorError,
   8     determine_ext,
   9     int_or_none,
  10     unified_strdate,
  11     US_RATINGS,
  12 )
  13
  14
  15 class PBSIE(InfoExtractor):
  16     _VALID_URL = r'''(?x)https?://
  17         (?:
  18            # Direct video URL
  19            video\.pbs\.org/(?:viralplayer|video)/(?P<id>[0-9]+)/? |
  20            # Article with embedded player (or direct video)
  21            (?:www\.)?pbs\.org/(?:[^/]+/){2,5}(?P<presumptive_id>[^/]+?)(?:\.html)?/?(?:$|[?\#]) |
  22            # Player
  23            video\.pbs\.org/(?:widget/)?partnerplayer/(?P<player_id>[^/]+)/
  24         )
  25     '''
  26
  27     _TESTS = [
  28         {
  29             'url': 'http://www.pbs.org/tpt/constitution-usa-peter-sagal/watch/a-more-perfect-union/',
  30             'md5': 'ce1888486f0908d555a8093cac9a7362',
  31             'info_dict': {
  32                 'id': '2365006249',
  33                 'ext': 'mp4',
  34                 'title': 'A More Perfect Union',
  35                 'description': 'md5:ba0c207295339c8d6eced00b7c363c6a',
  36                 'duration': 3190,
  37             },
  38             'params': {
  39                 'skip_download': True,  # requires ffmpeg
  40             },
  41         },
  42         {
  43             'url': 'http://www.pbs.org/wgbh/pages/frontline/losing-iraq/',
  44             'md5': '143c98aa54a346738a3d78f54c925321',
  45             'info_dict': {
  46                 'id': '2365297690',
  47                 'ext': 'mp4',
  48                 'title': 'Losing Iraq',
  49                 'description': 'md5:f5bfbefadf421e8bb8647602011caf8e',
  50                 'duration': 5050,
  51             },
  52             'params': {
  53                 'skip_download': True,  # requires ffmpeg
  54             }
  55         },
  56         {
  57             'url': 'http://www.pbs.org/newshour/bb/education-jan-june12-cyberschools_02-23/',
  58             'md5': 'b19856d7f5351b17a5ab1dc6a64be633',
  59             'info_dict': {
  60                 'id': '2201174722',
  61                 'ext': 'mp4',
  62                 'title': 'Cyber Schools Gain Popularity, but Quality Questions Persist',
  63                 'description': 'md5:5871c15cba347c1b3d28ac47a73c7c28',
  64                 'duration': 801,
  65             },
  66         },
  67         {
  68             'url': 'http://www.pbs.org/wnet/gperf/dudamel-conducts-verdi-requiem-hollywood-bowl-full-episode/3374/',
  69             'md5': 'c62859342be2a0358d6c9eb306595978',
  70             'info_dict': {
  71                 'id': '2365297708',
  72                 'ext': 'mp4',
  73                 'description': 'md5:68d87ef760660eb564455eb30ca464fe',
  74                 'title': 'Dudamel Conducts Verdi Requiem at the Hollywood Bowl - Full',
  75                 'duration': 6559,
  76                 'thumbnail': 're:^https?://.*\.jpg$',
  77             },
  78             'params': {
  79                 'skip_download': True,  # requires ffmpeg
  80             },
  81         },
  82         {
  83             'url': 'http://www.pbs.org/wgbh/nova/earth/killer-typhoon.html',
  84             'md5': '908f3e5473a693b266b84e25e1cf9703',
  85             'info_dict': {
  86                 'id': '2365160389',
  87                 'display_id': 'killer-typhoon',
  88                 'ext': 'mp4',
  89                 'description': 'md5:c741d14e979fc53228c575894094f157',
  90                 'title': 'Killer Typhoon',
  91                 'duration': 3172,
  92                 'thumbnail': 're:^https?://.*\.jpg$',
  93                 'upload_date': '20140122',
  94             },
  95             'params': {
  96                 'skip_download': True,  # requires ffmpeg
  97             },
  98         },
  99         {
 100             'url': 'http://www.pbs.org/wgbh/pages/frontline/united-states-of-secrets/',
 101             'info_dict': {
 102                 'id': 'united-states-of-secrets',
 103             },
 104             'playlist_count': 2,
 105         },
 106         {
 107             'url': 'http://www.pbs.org/wgbh/americanexperience/films/death/player/',
 108             'info_dict': {
 109                 'id': '2280706814',
 110                 'display_id': 'player',
 111                 'ext': 'mp4',
 112                 'title': 'Death and the Civil War',
 113                 'description': 'American Experience, TV’s most-watched history series, brings to life the compelling stories from our past that inform our understanding of the world today.',
 114                 'duration': 6705,
 115                 'thumbnail': 're:^https?://.*\.jpg$',
 116             },
 117             'params': {
 118                 'skip_download': True,  # requires ffmpeg
 119             },
 120         }
 121     ]
 122
 123     def _extract_webpage(self, url):
 124         mobj = re.match(self._VALID_URL, url)
 125
 126         presumptive_id = mobj.group('presumptive_id')
 127         display_id = presumptive_id
 128         if presumptive_id:
 129             webpage = self._download_webpage(url, display_id)
 130
 131             upload_date = unified_strdate(self._search_regex(
 132                 r'<input type="hidden" id="air_date_[0-9]+" value="([^"]+)"',
 133                 webpage, 'upload date', default=None))
 134
 135             # tabbed frontline videos
 136             tabbed_videos = re.findall(
 137                 r'<div[^>]+class="videotab[^"]*"[^>]+vid="(\d+)"', webpage)
 138             if tabbed_videos:
 139                 return tabbed_videos, presumptive_id, upload_date
 140
 141             MEDIA_ID_REGEXES = [
 142                 r"div\s*:\s*'videoembed'\s*,\s*mediaid\s*:\s*'(\d+)'",  # frontline video embed
 143                 r'class="coveplayerid">([^<]+)<',                       # coveplayer
 144                 r'<input type="hidden" id="pbs_video_id_[0-9]+" value="([0-9]+)"/>',  # jwplayer
 145             ]
 146
 147             media_id = self._search_regex(
 148                 MEDIA_ID_REGEXES, webpage, 'media ID', fatal=False, default=None)
 149             if media_id:
 150                 return media_id, presumptive_id, upload_date
 151
 152             url = self._search_regex(
 153                 r'<iframe\s+[^>]*\s+src=["\']([^\'"]+partnerplayer[^\'"]+)["\']',
 154                 webpage, 'player URL')
 155             mobj = re.match(self._VALID_URL, url)
 156
 157         player_id = mobj.group('player_id')
 158         if not display_id:
 159             display_id = player_id
 160         if player_id:
 161             player_page = self._download_webpage(
 162                 url, display_id, note='Downloading player page',
 163                 errnote='Could not download player page')
 164             video_id = self._search_regex(
 165                 r'<div\s+id="video_([0-9]+)"', player_page, 'video ID')
 166         else:
 167             video_id = mobj.group('id')
 168             display_id = video_id
 169
 170         return video_id, display_id, None
 171
 172     def _real_extract(self, url):
 173         video_id, display_id, upload_date = self._extract_webpage(url)
 174
 175         if isinstance(video_id, list):
 176             entries = [self.url_result(
 177                 'http://video.pbs.org/video/%s' % vid_id, 'PBS', vid_id)
 178                 for vid_id in video_id]
 179             return self.playlist_result(entries, display_id)
 180
 181         info = self._download_json(
 182             'http://video.pbs.org/videoInfo/%s?format=json&type=partner' % video_id,
 183             display_id)
 184
 185         formats = []
 186         for encoding_name in ('recommended_encoding', 'alternate_encoding'):
 187             redirect = info.get(encoding_name)
 188             if not redirect:
 189                 continue
 190             redirect_url = redirect.get('url')
 191             if not redirect_url:
 192                 continue
 193
 194             redirect_info = self._download_json(
 195                 redirect_url + '?format=json', display_id,
 196                 'Downloading %s video url info' % encoding_name)
 197
 198             if redirect_info['status'] == 'error':
 199                 if redirect_info['http_code'] == 403:
 200                     message = (
 201                         'The video is not available in your region due to '
 202                         'right restrictions')
 203                 else:
 204                     message = redirect_info['message']
 205                 raise ExtractorError(message, expected=True)
 206
 207             format_url = redirect_info.get('url')
 208             if not format_url:
 209                 continue
 210
 211             if determine_ext(format_url) == 'm3u8':
 212                 formats.extend(self._extract_m3u8_formats(
 213                     format_url, display_id, 'mp4', preference=1, m3u8_id='hls'))
 214             else:
 215                 formats.append({
 216                     'url': format_url,
 217                     'format_id': redirect.get('eeid'),
 218                 })
 219         self._sort_formats(formats)
 220
 221         rating_str = info.get('rating')
 222         if rating_str is not None:
 223             rating_str = rating_str.rpartition('-')[2]
 224         age_limit = US_RATINGS.get(rating_str)
 225
 226         return {
 227             'id': video_id,
 228             'display_id': display_id,
 229             'title': info['title'],
 230             'description': info['program'].get('description'),
 231             'thumbnail': info.get('image_url'),
 232             'duration': int_or_none(info.get('duration')),
 233             'age_limit': age_limit,
 234             'upload_date': upload_date,
 235             'formats': formats,
 236         }