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$',
  94                 'upload_date': '20140122',
  95                 'age_limit': 10,
  96             },
  97             'params': {
  98                 'skip_download': True,  # requires ffmpeg
  99             },
 100         },
 101         {
 102             'url': 'http://www.pbs.org/wgbh/pages/frontline/united-states-of-secrets/',
 103             'info_dict': {
 104                 'id': 'united-states-of-secrets',
 105             },
 106             'playlist_count': 2,
 107         },
 108         {
 109             'url': 'http://www.pbs.org/wgbh/americanexperience/films/death/player/',
 110             'info_dict': {
 111                 'id': '2276541483',
 112                 'display_id': 'player',
 113                 'ext': 'mp4',
 114                 'title': 'American Experience - Death and the Civil War, Chapter 1',
 115                 'description': 'American Experience, TV’s most-watched history series, brings to life the compelling stories from our past that inform our understanding of the world today.',
 116                 'duration': 682,
 117                 'thumbnail': 're:^https?://.*\.jpg$',
 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$',
 133             },
 134             'params': {
 135                 'skip_download': True,  # requires ffmpeg
 136             },
 137             'skip': 'Expired',
 138         },
 139         {
 140             # Video embedded in iframe containing angle brackets as attribute's value (e.g.
 141             # "<iframe style='position: absolute;<br />\ntop: 0; left: 0;' ...", see
 142             # https://github.com/rg3/youtube-dl/issues/7059)
 143             'url': 'http://www.pbs.org/food/features/a-chefs-life-season-3-episode-5-prickly-business/',
 144             'info_dict': {
 145                 'id': '2365546844',
 146                 'display_id': 'a-chefs-life-season-3-episode-5-prickly-business',
 147                 'ext': 'mp4',
 148                 'title': "A Chef's Life - Season 3, Ep. 5: Prickly Business",
 149                 'description': 'md5:61db2ddf27c9912f09c241014b118ed1',
 150                 'duration': 1480,
 151                 'thumbnail': 're:^https?://.*\.jpg$',
 152             },
 153             'params': {
 154                 'skip_download': True,  # requires ffmpeg
 155             },
 156         }
 157     ]
 158     _ERRORS = {
 159         101: 'We\'re sorry, but this video is not yet available.',
 160         403: 'We\'re sorry, but this video is not available in your region due to right restrictions.',
 161         404: 'We are experiencing technical difficulties that are preventing us from playing the video at this time. Please check back again soon.',
 162         410: 'This video has expired and is no longer available for online streaming.',
 163     }
 164
 165     def _extract_webpage(self, url):
 166         mobj = re.match(self._VALID_URL, url)
 167
 168         presumptive_id = mobj.group('presumptive_id')
 169         display_id = presumptive_id
 170         if presumptive_id:
 171             webpage = self._download_webpage(url, display_id)
 172
 173             upload_date = unified_strdate(self._search_regex(
 174                 r'<input type="hidden" id="air_date_[0-9]+" value="([^"]+)"',
 175                 webpage, 'upload date', default=None))
 176
 177             # tabbed frontline videos
 178             tabbed_videos = re.findall(
 179                 r'<div[^>]+class="videotab[^"]*"[^>]+vid="(\d+)"', webpage)
 180             if tabbed_videos:
 181                 return tabbed_videos, presumptive_id, upload_date
 182
 183             MEDIA_ID_REGEXES = [
 184                 r"div\s*:\s*'videoembed'\s*,\s*mediaid\s*:\s*'(\d+)'",  # frontline video embed
 185                 r'class="coveplayerid">([^<]+)<',                       # coveplayer
 186                 r'<input type="hidden" id="pbs_video_id_[0-9]+" value="([0-9]+)"/>',  # jwplayer
 187             ]
 188
 189             media_id = self._search_regex(
 190                 MEDIA_ID_REGEXES, webpage, 'media ID', fatal=False, default=None)
 191             if media_id:
 192                 return media_id, presumptive_id, upload_date
 193
 194             url = self._search_regex(
 195                 r'(?s)<iframe[^>]+?(?:[a-z-]+?=["\'].*?["\'][^>]+?)*?\bsrc=["\']([^\'"]+partnerplayer[^\'"]+)["\']',
 196                 webpage, 'player URL')
 197             mobj = re.match(self._VALID_URL, url)
 198
 199         player_id = mobj.group('player_id')
 200         if not display_id:
 201             display_id = player_id
 202         if player_id:
 203             player_page = self._download_webpage(
 204                 url, display_id, note='Downloading player page',
 205                 errnote='Could not download player page')
 206             video_id = self._search_regex(
 207                 r'<div\s+id="video_([0-9]+)"', player_page, 'video ID')
 208         else:
 209             video_id = mobj.group('id')
 210             display_id = video_id
 211
 212         return video_id, display_id, None
 213
 214     def _real_extract(self, url):
 215         video_id, display_id, upload_date = self._extract_webpage(url)
 216
 217         if isinstance(video_id, list):
 218             entries = [self.url_result(
 219                 'http://video.pbs.org/video/%s' % vid_id, 'PBS', vid_id)
 220                 for vid_id in video_id]
 221             return self.playlist_result(entries, display_id)
 222
 223         info = self._download_json(
 224             'http://video.pbs.org/videoInfo/%s?format=json&type=partner' % video_id,
 225             display_id)
 226
 227         formats = []
 228         for encoding_name in ('recommended_encoding', 'alternate_encoding'):
 229             redirect = info.get(encoding_name)
 230             if not redirect:
 231                 continue
 232             redirect_url = redirect.get('url')
 233             if not redirect_url:
 234                 continue
 235
 236             redirect_info = self._download_json(
 237                 redirect_url + '?format=json', display_id,
 238                 'Downloading %s video url info' % encoding_name)
 239
 240             if redirect_info['status'] == 'error':
 241                 raise ExtractorError(
 242                     '%s said: %s' % (
 243                         self.IE_NAME,
 244                         self._ERRORS.get(redirect_info['http_code'], redirect_info['message'])),
 245                     expected=True)
 246
 247             format_url = redirect_info.get('url')
 248             if not format_url:
 249                 continue
 250
 251             if determine_ext(format_url) == 'm3u8':
 252                 formats.extend(self._extract_m3u8_formats(
 253                     format_url, display_id, 'mp4', preference=1, m3u8_id='hls'))
 254             else:
 255                 formats.append({
 256                     'url': format_url,
 257                     'format_id': redirect.get('eeid'),
 258                 })
 259         self._sort_formats(formats)
 260
 261         rating_str = info.get('rating')
 262         if rating_str is not None:
 263             rating_str = rating_str.rpartition('-')[2]
 264         age_limit = US_RATINGS.get(rating_str)
 265
 266         subtitles = {}
 267         closed_captions_url = info.get('closed_captions_url')
 268         if closed_captions_url:
 269             subtitles['en'] = [{
 270                 'ext': 'ttml',
 271                 'url': closed_captions_url,
 272             }]
 273
 274         # info['title'] is often incomplete (e.g. 'Full Episode', 'Episode 5', etc)
 275         # Try turning it to 'program - title' naming scheme if possible
 276         alt_title = info.get('program', {}).get('title')
 277         if alt_title:
 278             info['title'] = alt_title + ' - ' + re.sub(r'^' + alt_title + '[\s\-:]+', '', info['title'])
 279
 280         return {
 281             'id': video_id,
 282             'display_id': display_id,
 283             'title': info['title'],
 284             'description': info['program'].get('description'),
 285             'thumbnail': info.get('image_url'),
 286             'duration': int_or_none(info.get('duration')),
 287             'age_limit': age_limit,
 288             'upload_date': upload_date,
 289             'formats': formats,
 290             'subtitles': subtitles,
 291         }