youtube_dl/extractor/pbs.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..utils import (
   8     ExtractorError,
   9     determine_ext,
  10     int_or_none,
  11     unified_strdate,
  12     US_RATINGS,
  13 )
  14
  15
  16 class PBSIE(InfoExtractor):
  17     _VALID_URL = r'''(?x)https?://
  18         (?:
  19            # Direct video URL
  20            video\.pbs\.org/(?:viralplayer|video)/(?P<id>[0-9]+)/? |
  21            # Article with embedded player (or direct video)
  22            (?:www\.)?pbs\.org/(?:[^/]+/){2,5}(?P<presumptive_id>[^/]+?)(?:\.html)?/?(?:$|[?\#]) |
  23            # Player
  24            video\.pbs\.org/(?:widget/)?partnerplayer/(?P<player_id>[^/]+)/
  25         )
  26     '''
  27
  28     _TESTS = [
  29         {
  30             'url': 'http://www.pbs.org/tpt/constitution-usa-peter-sagal/watch/a-more-perfect-union/',
  31             'md5': 'ce1888486f0908d555a8093cac9a7362',
  32             'info_dict': {
  33                 'id': '2365006249',
  34                 'ext': 'mp4',
  35                 'title': 'Constitution USA with Peter Sagal - A More Perfect Union',
  36                 'description': 'md5:ba0c207295339c8d6eced00b7c363c6a',
  37                 'duration': 3190,
  38             },
  39             'params': {
  40                 'skip_download': True,  # requires ffmpeg
  41             },
  42         },
  43         {
  44             'url': 'http://www.pbs.org/wgbh/pages/frontline/losing-iraq/',
  45             'md5': '143c98aa54a346738a3d78f54c925321',
  46             'info_dict': {
  47                 'id': '2365297690',
  48                 'ext': 'mp4',
  49                 'title': 'FRONTLINE - Losing Iraq',
  50                 'description': 'md5:f5bfbefadf421e8bb8647602011caf8e',
  51                 'duration': 5050,
  52             },
  53             'params': {
  54                 'skip_download': True,  # requires ffmpeg
  55             }
  56         },
  57         {
  58             'url': 'http://www.pbs.org/newshour/bb/education-jan-june12-cyberschools_02-23/',
  59             'md5': 'b19856d7f5351b17a5ab1dc6a64be633',
  60             'info_dict': {
  61                 'id': '2201174722',
  62                 'ext': 'mp4',
  63                 'title': 'PBS NewsHour - Cyber Schools Gain Popularity, but Quality Questions Persist',
  64                 'description': 'md5:5871c15cba347c1b3d28ac47a73c7c28',
  65                 'duration': 801,
  66             },
  67         },
  68         {
  69             'url': 'http://www.pbs.org/wnet/gperf/dudamel-conducts-verdi-requiem-hollywood-bowl-full-episode/3374/',
  70             'md5': 'c62859342be2a0358d6c9eb306595978',
  71             'info_dict': {
  72                 'id': '2365297708',
  73                 'ext': 'mp4',
  74                 'description': 'md5:68d87ef760660eb564455eb30ca464fe',
  75                 'title': 'Great Performances - Dudamel Conducts Verdi Requiem at the Hollywood Bowl - Full',
  76                 'duration': 6559,
  77                 'thumbnail': 're:^https?://.*\.jpg$',
  78             },
  79             'params': {
  80                 'skip_download': True,  # requires ffmpeg
  81             },
  82         },
  83         {
  84             'url': 'http://www.pbs.org/wgbh/nova/earth/killer-typhoon.html',
  85             'md5': '908f3e5473a693b266b84e25e1cf9703',
  86             'info_dict': {
  87                 'id': '2365160389',
  88                 'display_id': 'killer-typhoon',
  89                 'ext': 'mp4',
  90                 'description': 'md5:c741d14e979fc53228c575894094f157',
  91                 'title': 'NOVA - Killer Typhoon',
  92                 'duration': 3172,
  93                 'thumbnail': 're:^https?://.*\.jpg$',
  94                 'upload_date': '20140122',
  95                 'age_limit': 10,
  96             },
  97             'params': {
  98                 'skip_download': True,  # requires ffmpeg
  99             },
 100         },
 101         {
 102             'url': 'http://www.pbs.org/wgbh/pages/frontline/united-states-of-secrets/',
 103             'info_dict': {
 104                 'id': 'united-states-of-secrets',
 105             },
 106             'playlist_count': 2,
 107         },
 108         {
 109             'url': 'http://www.pbs.org/wgbh/americanexperience/films/death/player/',
 110             'info_dict': {
 111                 'id': '2280706814',
 112                 'display_id': 'player',
 113                 'ext': 'mp4',
 114                 'title': 'American Experience - Death and the Civil War',
 115                 'description': 'American Experience, TV’s most-watched history series, brings to life the compelling stories from our past that inform our understanding of the world today.',
 116                 'duration': 6705,
 117                 'thumbnail': 're:^https?://.*\.jpg$',
 118             },
 119             'params': {
 120                 'skip_download': True,  # requires ffmpeg
 121             },
 122         },
 123         {
 124             'url': 'http://video.pbs.org/video/2365367186/',
 125             'info_dict': {
 126                 'id': '2365367186',
 127                 'display_id': '2365367186',
 128                 'ext': 'mp4',
 129                 'title': 'To Catch A Comet - Full Episode',
 130                 'description': 'On November 12, 2014, billions of kilometers from Earth, spacecraft orbiter Rosetta and lander Philae did what no other had dared to attempt \u2014 land on the volatile surface of a comet as it zooms around the sun at 67,000 km/hr. The European Space Agency hopes this mission can help peer into our past and unlock secrets of our origins.',
 131                 'duration': 3342,
 132                 'thumbnail': 're:^https?://.*\.jpg$',
 133             },
 134             'params': {
 135                 'skip_download': True,  # requires ffmpeg
 136             },
 137         },
 138         {
 139             # Video embedded in iframe containing angle brackets as attribute's value (e.g.
 140             # "<iframe style='position: absolute;<br />\ntop: 0; left: 0;' ...", see
 141             # https://github.com/rg3/youtube-dl/issues/7059)
 142             'url': 'http://www.pbs.org/food/features/a-chefs-life-season-3-episode-5-prickly-business/',
 143             'info_dict': {
 144                 'id': '2365546844',
 145                 'display_id': 'a-chefs-life-season-3-episode-5-prickly-business',
 146                 'ext': 'mp4',
 147                 'title': "A Chef's Life - Season 3, Ep. 5: Prickly Business",
 148                 'description': 'md5:61db2ddf27c9912f09c241014b118ed1',
 149                 'duration': 1480,
 150                 'thumbnail': 're:^https?://.*\.jpg$',
 151             },
 152             'params': {
 153                 'skip_download': True,  # requires ffmpeg
 154             },
 155         }
 156     ]
 157
 158     def _extract_webpage(self, url):
 159         mobj = re.match(self._VALID_URL, url)
 160
 161         presumptive_id = mobj.group('presumptive_id')
 162         display_id = presumptive_id
 163         if presumptive_id:
 164             webpage = self._download_webpage(url, display_id)
 165
 166             upload_date = unified_strdate(self._search_regex(
 167                 r'<input type="hidden" id="air_date_[0-9]+" value="([^"]+)"',
 168                 webpage, 'upload date', default=None))
 169
 170             # tabbed frontline videos
 171             tabbed_videos = re.findall(
 172                 r'<div[^>]+class="videotab[^"]*"[^>]+vid="(\d+)"', webpage)
 173             if tabbed_videos:
 174                 return tabbed_videos, presumptive_id, upload_date
 175
 176             MEDIA_ID_REGEXES = [
 177                 r"div\s*:\s*'videoembed'\s*,\s*mediaid\s*:\s*'(\d+)'",  # frontline video embed
 178                 r'class="coveplayerid">([^<]+)<',                       # coveplayer
 179                 r'<input type="hidden" id="pbs_video_id_[0-9]+" value="([0-9]+)"/>',  # jwplayer
 180             ]
 181
 182             media_id = self._search_regex(
 183                 MEDIA_ID_REGEXES, webpage, 'media ID', fatal=False, default=None)
 184             if media_id:
 185                 return media_id, presumptive_id, upload_date
 186
 187             url = self._search_regex(
 188                 r'(?s)<iframe[^>]+?(?:[a-z-]+?=["\'].*?["\'][^>]+?)*?\bsrc=["\']([^\'"]+partnerplayer[^\'"]+)["\']',
 189                 webpage, 'player URL')
 190             mobj = re.match(self._VALID_URL, url)
 191
 192         player_id = mobj.group('player_id')
 193         if not display_id:
 194             display_id = player_id
 195         if player_id:
 196             player_page = self._download_webpage(
 197                 url, display_id, note='Downloading player page',
 198                 errnote='Could not download player page')
 199             video_id = self._search_regex(
 200                 r'<div\s+id="video_([0-9]+)"', player_page, 'video ID')
 201         else:
 202             video_id = mobj.group('id')
 203             display_id = video_id
 204
 205         return video_id, display_id, None
 206
 207     def _real_extract(self, url):
 208         video_id, display_id, upload_date = self._extract_webpage(url)
 209
 210         if isinstance(video_id, list):
 211             entries = [self.url_result(
 212                 'http://video.pbs.org/video/%s' % vid_id, 'PBS', vid_id)
 213                 for vid_id in video_id]
 214             return self.playlist_result(entries, display_id)
 215
 216         info = self._download_json(
 217             'http://video.pbs.org/videoInfo/%s?format=json&type=partner' % video_id,
 218             display_id)
 219
 220         formats = []
 221         for encoding_name in ('recommended_encoding', 'alternate_encoding'):
 222             redirect = info.get(encoding_name)
 223             if not redirect:
 224                 continue
 225             redirect_url = redirect.get('url')
 226             if not redirect_url:
 227                 continue
 228
 229             redirect_info = self._download_json(
 230                 redirect_url + '?format=json', display_id,
 231                 'Downloading %s video url info' % encoding_name)
 232
 233             if redirect_info['status'] == 'error':
 234                 if redirect_info['http_code'] == 403:
 235                     message = (
 236                         'The video is not available in your region due to '
 237                         'right restrictions')
 238                 else:
 239                     message = redirect_info['message']
 240                 raise ExtractorError(message, expected=True)
 241
 242             format_url = redirect_info.get('url')
 243             if not format_url:
 244                 continue
 245
 246             if determine_ext(format_url) == 'm3u8':
 247                 formats.extend(self._extract_m3u8_formats(
 248                     format_url, display_id, 'mp4', preference=1, m3u8_id='hls'))
 249             else:
 250                 formats.append({
 251                     'url': format_url,
 252                     'format_id': redirect.get('eeid'),
 253                 })
 254         self._sort_formats(formats)
 255
 256         rating_str = info.get('rating')
 257         if rating_str is not None:
 258             rating_str = rating_str.rpartition('-')[2]
 259         age_limit = US_RATINGS.get(rating_str)
 260
 261         subtitles = {}
 262         closed_captions_url = info.get('closed_captions_url')
 263         if closed_captions_url:
 264             subtitles['en'] = [{
 265                 'ext': 'ttml',
 266                 'url': closed_captions_url,
 267             }]
 268
 269         # info['title'] is often incomplete (e.g. 'Full Episode', 'Episode 5', etc)
 270         # Try turning it to 'program - title' naming scheme if possible
 271         alt_title = info.get('program', {}).get('title')
 272         if alt_title:
 273             info['title'] = alt_title + ' - ' + re.sub(r'^' + alt_title + '[\s\-:]+', '', info['title'])
 274
 275         return {
 276             'id': video_id,
 277             'display_id': display_id,
 278             'title': info['title'],
 279             'description': info['program'].get('description'),
 280             'thumbnail': info.get('image_url'),
 281             'duration': int_or_none(info.get('duration')),
 282             'age_limit': age_limit,
 283             'upload_date': upload_date,
 284             'formats': formats,
 285             'subtitles': subtitles,
 286         }