Raphaël G. Git Repositories - youtubedl/blob - 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     strip_jsonp,
  12     unified_strdate,
  13     US_RATINGS,
  14 )
  15
  16
  17 class PBSIE(InfoExtractor):
  18     _VALID_URL = r'''(?x)https?://
  19         (?:
  20            # Direct video URL
  21            video\.pbs\.org/(?:viralplayer|video)/(?P<id>[0-9]+)/? |
  22            # Article with embedded player (or direct video)
  23            (?:www\.)?pbs\.org/(?:[^/]+/){2,5}(?P<presumptive_id>[^/]+?)(?:\.html)?/?(?:$|[?\#]) |
  24            # Player
  25            video\.pbs\.org/(?:widget/)?partnerplayer/(?P<player_id>[^/]+)/
  26         )
  27     '''
  28
  29     _TESTS = [
  30         {
  31             'url': 'http://www.pbs.org/tpt/constitution-usa-peter-sagal/watch/a-more-perfect-union/',
  32             'md5': 'ce1888486f0908d555a8093cac9a7362',
  33             'info_dict': {
  34                 'id': '2365006249',
  35                 'ext': 'mp4',
  36                 'title': 'Constitution USA with Peter Sagal - A More Perfect Union',
  37                 'description': 'md5:ba0c207295339c8d6eced00b7c363c6a',
  38                 'duration': 3190,
  39             },
  40             'params': {
  41                 'skip_download': True,  # requires ffmpeg
  42             },
  43         },
  44         {
  45             'url': 'http://www.pbs.org/wgbh/pages/frontline/losing-iraq/',
  46             'md5': '143c98aa54a346738a3d78f54c925321',
  47             'info_dict': {
  48                 'id': '2365297690',
  49                 'ext': 'mp4',
  50                 'title': 'FRONTLINE - Losing Iraq',
  51                 'description': 'md5:f5bfbefadf421e8bb8647602011caf8e',
  52                 'duration': 5050,
  53             },
  54             'params': {
  55                 'skip_download': True,  # requires ffmpeg
  56             }
  57         },
  58         {
  59             'url': 'http://www.pbs.org/newshour/bb/education-jan-june12-cyberschools_02-23/',
  60             'md5': 'b19856d7f5351b17a5ab1dc6a64be633',
  61             'info_dict': {
  62                 'id': '2201174722',
  63                 'ext': 'mp4',
  64                 'title': 'PBS NewsHour - Cyber Schools Gain Popularity, but Quality Questions Persist',
  65                 'description': 'md5:5871c15cba347c1b3d28ac47a73c7c28',
  66                 'duration': 801,
  67             },
  68         },
  69         {
  70             'url': 'http://www.pbs.org/wnet/gperf/dudamel-conducts-verdi-requiem-hollywood-bowl-full-episode/3374/',
  71             'md5': 'c62859342be2a0358d6c9eb306595978',
  72             'info_dict': {
  73                 'id': '2365297708',
  74                 'ext': 'mp4',
  75                 'description': 'md5:68d87ef760660eb564455eb30ca464fe',
  76                 'title': 'Great Performances - Dudamel Conducts Verdi Requiem at the Hollywood Bowl - Full',
  77                 'duration': 6559,
  78                 'thumbnail': 're:^https?://.*\.jpg$',
  79             },
  80             'params': {
  81                 'skip_download': True,  # requires ffmpeg
  82             },
  83         },
  84         {
  85             'url': 'http://www.pbs.org/wgbh/nova/earth/killer-typhoon.html',
  86             'md5': '908f3e5473a693b266b84e25e1cf9703',
  87             'info_dict': {
  88                 'id': '2365160389',
  89                 'display_id': 'killer-typhoon',
  90                 'ext': 'mp4',
  91                 'description': 'md5:c741d14e979fc53228c575894094f157',
  92                 'title': 'NOVA - Killer Typhoon',
  93                 'duration': 3172,
  94                 'thumbnail': 're:^https?://.*\.jpg$',
  95                 'upload_date': '20140122',
  96                 'age_limit': 10,
  97             },
  98             'params': {
  99                 'skip_download': True,  # requires ffmpeg
 100             },
 101         },
 102         {
 103             'url': 'http://www.pbs.org/wgbh/pages/frontline/united-states-of-secrets/',
 104             'info_dict': {
 105                 'id': 'united-states-of-secrets',
 106             },
 107             'playlist_count': 2,
 108         },
 109         {
 110             'url': 'http://www.pbs.org/wgbh/americanexperience/films/death/player/',
 111             'info_dict': {
 112                 'id': '2276541483',
 113                 'display_id': 'player',
 114                 'ext': 'mp4',
 115                 'title': 'American Experience - Death and the Civil War, Chapter 1',
 116                 'description': 'American Experience, TV’s most-watched history series, brings to life the compelling stories from our past that inform our understanding of the world today.',
 117                 'duration': 682,
 118                 'thumbnail': 're:^https?://.*\.jpg$',
 119             },
 120             'params': {
 121                 'skip_download': True,  # requires ffmpeg
 122             },
 123         },
 124         {
 125             'url': 'http://video.pbs.org/video/2365367186/',
 126             'info_dict': {
 127                 'id': '2365367186',
 128                 'display_id': '2365367186',
 129                 'ext': 'mp4',
 130                 'title': 'To Catch A Comet - Full Episode',
 131                 '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.',
 132                 'duration': 3342,
 133                 'thumbnail': 're:^https?://.*\.jpg$',
 134             },
 135             'params': {
 136                 'skip_download': True,  # requires ffmpeg
 137             },
 138             'skip': 'Expired',
 139         },
 140         {
 141             # Video embedded in iframe containing angle brackets as attribute's value (e.g.
 142             # "<iframe style='position: absolute;<br />\ntop: 0; left: 0;' ...", see
 143             # https://github.com/rg3/youtube-dl/issues/7059)
 144             'url': 'http://www.pbs.org/food/features/a-chefs-life-season-3-episode-5-prickly-business/',
 145             'info_dict': {
 146                 'id': '2365546844',
 147                 'display_id': 'a-chefs-life-season-3-episode-5-prickly-business',
 148                 'ext': 'mp4',
 149                 'title': "A Chef's Life - Season 3, Ep. 5: Prickly Business",
 150                 'description': 'md5:61db2ddf27c9912f09c241014b118ed1',
 151                 'duration': 1480,
 152                 'thumbnail': 're:^https?://.*\.jpg$',
 153             },
 154             'params': {
 155                 'skip_download': True,  # requires ffmpeg
 156             },
 157         },
 158         {
 159             # Frontline video embedded via flp2012.js
 160             'url': 'http://www.pbs.org/wgbh/pages/frontline/the-atomic-artists',
 161             'info_dict': {
 162                 'id': '2070868960',
 163                 'display_id': 'the-atomic-artists',
 164                 'ext': 'mp4',
 165                 'title': 'FRONTLINE - The Atomic Artists',
 166                 'description': 'md5:f5bfbefadf421e8bb8647602011caf8e',
 167                 'duration': 723,
 168                 'thumbnail': 're:^https?://.*\.jpg$',
 169             },
 170             'params': {
 171                 'skip_download': True,  # requires ffmpeg
 172             },
 173         }
 174     ]
 175     _ERRORS = {
 176         101: 'We\'re sorry, but this video is not yet available.',
 177         403: 'We\'re sorry, but this video is not available in your region due to right restrictions.',
 178         404: 'We are experiencing technical difficulties that are preventing us from playing the video at this time. Please check back again soon.',
 179         410: 'This video has expired and is no longer available for online streaming.',
 180     }
 181
 182     def _extract_webpage(self, url):
 183         mobj = re.match(self._VALID_URL, url)
 184
 185         presumptive_id = mobj.group('presumptive_id')
 186         display_id = presumptive_id
 187         if presumptive_id:
 188             webpage = self._download_webpage(url, display_id)
 189
 190             upload_date = unified_strdate(self._search_regex(
 191                 r'<input type="hidden" id="air_date_[0-9]+" value="([^"]+)"',
 192                 webpage, 'upload date', default=None))
 193
 194             # tabbed frontline videos
 195             tabbed_videos = re.findall(
 196                 r'<div[^>]+class="videotab[^"]*"[^>]+vid="(\d+)"', webpage)
 197             if tabbed_videos:
 198                 return tabbed_videos, presumptive_id, upload_date
 199
 200             MEDIA_ID_REGEXES = [
 201                 r"div\s*:\s*'videoembed'\s*,\s*mediaid\s*:\s*'(\d+)'",  # frontline video embed
 202                 r'class="coveplayerid">([^<]+)<',                       # coveplayer
 203                 r'<input type="hidden" id="pbs_video_id_[0-9]+" value="([0-9]+)"/>',  # jwplayer
 204             ]
 205
 206             media_id = self._search_regex(
 207                 MEDIA_ID_REGEXES, webpage, 'media ID', fatal=False, default=None)
 208             if media_id:
 209                 return media_id, presumptive_id, upload_date
 210
 211             # Fronline video embedded via flp
 212             video_id = self._search_regex(
 213                 r'videoid\s*:\s*"([\d+a-z]{7,})"', webpage, 'videoid', default=None)
 214             if video_id:
 215                 # pkg_id calculation is reverse engineered from
 216                 # http://www.pbs.org/wgbh/pages/frontline/js/flp2012.js
 217                 prg_id = self._search_regex(
 218                     r'videoid\s*:\s*"([\d+a-z]{7,})"', webpage, 'videoid')[7:]
 219                 if 'q' in prg_id:
 220                     prg_id = prg_id.split('q')[1]
 221                 prg_id = int(prg_id, 16)
 222                 getdir = self._download_json(
 223                     'http://www.pbs.org/wgbh/pages/frontline/.json/getdir/getdir%d.json' % prg_id,
 224                     presumptive_id, 'Downloading getdir JSON',
 225                     transform_source=strip_jsonp)
 226                 return getdir['mid'], presumptive_id, upload_date
 227
 228             for iframe in re.findall(r'(?s)<iframe(.+?)></iframe>', webpage):
 229                 url = self._search_regex(
 230                     r'src=(["\'])(?P<url>.+?partnerplayer.+?)\1', iframe,
 231                     'player URL', default=None, group='url')
 232                 if url:
 233                     break
 234
 235             mobj = re.match(self._VALID_URL, url)
 236
 237         player_id = mobj.group('player_id')
 238         if not display_id:
 239             display_id = player_id
 240         if player_id:
 241             player_page = self._download_webpage(
 242                 url, display_id, note='Downloading player page',
 243                 errnote='Could not download player page')
 244             video_id = self._search_regex(
 245                 r'<div\s+id="video_([0-9]+)"', player_page, 'video ID')
 246         else:
 247             video_id = mobj.group('id')
 248             display_id = video_id
 249
 250         return video_id, display_id, None
 251
 252     def _real_extract(self, url):
 253         video_id, display_id, upload_date = self._extract_webpage(url)
 254
 255         if isinstance(video_id, list):
 256             entries = [self.url_result(
 257                 'http://video.pbs.org/video/%s' % vid_id, 'PBS', vid_id)
 258                 for vid_id in video_id]
 259             return self.playlist_result(entries, display_id)
 260
 261         info = self._download_json(
 262             'http://video.pbs.org/videoInfo/%s?format=json&type=partner' % video_id,
 263             display_id)
 264
 265         formats = []
 266         for encoding_name in ('recommended_encoding', 'alternate_encoding'):
 267             redirect = info.get(encoding_name)
 268             if not redirect:
 269                 continue
 270             redirect_url = redirect.get('url')
 271             if not redirect_url:
 272                 continue
 273
 274             redirect_info = self._download_json(
 275                 redirect_url + '?format=json', display_id,
 276                 'Downloading %s video url info' % encoding_name)
 277
 278             if redirect_info['status'] == 'error':
 279                 raise ExtractorError(
 280                     '%s said: %s' % (
 281                         self.IE_NAME,
 282                         self._ERRORS.get(redirect_info['http_code'], redirect_info['message'])),
 283                     expected=True)
 284
 285             format_url = redirect_info.get('url')
 286             if not format_url:
 287                 continue
 288
 289             if determine_ext(format_url) == 'm3u8':
 290                 formats.extend(self._extract_m3u8_formats(
 291                     format_url, display_id, 'mp4', preference=1, m3u8_id='hls'))
 292             else:
 293                 formats.append({
 294                     'url': format_url,
 295                     'format_id': redirect.get('eeid'),
 296                 })
 297         self._sort_formats(formats)
 298
 299         rating_str = info.get('rating')
 300         if rating_str is not None:
 301             rating_str = rating_str.rpartition('-')[2]
 302         age_limit = US_RATINGS.get(rating_str)
 303
 304         subtitles = {}
 305         closed_captions_url = info.get('closed_captions_url')
 306         if closed_captions_url:
 307             subtitles['en'] = [{
 308                 'ext': 'ttml',
 309                 'url': closed_captions_url,
 310             }]
 311
 312         # info['title'] is often incomplete (e.g. 'Full Episode', 'Episode 5', etc)
 313         # Try turning it to 'program - title' naming scheme if possible
 314         alt_title = info.get('program', {}).get('title')
 315         if alt_title:
 316             info['title'] = alt_title + ' - ' + re.sub(r'^' + alt_title + '[\s\-:]+', '', info['title'])
 317
 318         return {
 319             'id': video_id,
 320             'display_id': display_id,
 321             'title': info['title'],
 322             'description': info['program'].get('description'),
 323             'thumbnail': info.get('image_url'),
 324             'duration': int_or_none(info.get('duration')),
 325             'age_limit': age_limit,
 326             'upload_date': upload_date,
 327             'formats': formats,
 328             'subtitles': subtitles,
 329         }