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$',
  94                 'upload_date': '20140122',
  95             },
  96             'params': {
  97                 'skip_download': True,  # requires ffmpeg
  98             },
  99         },
 100         {
 101             'url': 'http://www.pbs.org/wgbh/pages/frontline/united-states-of-secrets/',
 102             'info_dict': {
 103                 'id': 'united-states-of-secrets',
 104             },
 105             'playlist_count': 2,
 106         },
 107         {
 108             'url': 'http://www.pbs.org/wgbh/americanexperience/films/death/player/',
 109             'info_dict': {
 110                 'id': '2280706814',
 111                 'display_id': 'player',
 112                 'ext': 'mp4',
 113                 'title': 'American Experience - Death and the Civil War',
 114                 'description': 'American Experience, TV’s most-watched history series, brings to life the compelling stories from our past that inform our understanding of the world today.',
 115                 'duration': 6705,
 116                 'thumbnail': 're:^https?://.*\.jpg$',
 117             },
 118             'params': {
 119                 'skip_download': True,  # requires ffmpeg
 120             },
 121         },
 122         {
 123             'url': 'http://video.pbs.org/video/2365367186/',
 124             'info_dict': {
 125                 'id': '2365367186',
 126                 'display_id': '2365367186',
 127                 'ext': 'mp4',
 128                 'title': 'To Catch A Comet - Full Episode',
 129                 '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.',
 130                 'duration': 3342,
 131                 'thumbnail': 're:^https?://.*\.jpg$',
 132             },
 133             'params': {
 134                 'skip_download': True,  # requires ffmpeg
 135             },
 136         }
 137     ]
 138
 139     def _extract_webpage(self, url):
 140         mobj = re.match(self._VALID_URL, url)
 141
 142         presumptive_id = mobj.group('presumptive_id')
 143         display_id = presumptive_id
 144         if presumptive_id:
 145             webpage = self._download_webpage(url, display_id)
 146
 147             upload_date = unified_strdate(self._search_regex(
 148                 r'<input type="hidden" id="air_date_[0-9]+" value="([^"]+)"',
 149                 webpage, 'upload date', default=None))
 150
 151             # tabbed frontline videos
 152             tabbed_videos = re.findall(
 153                 r'<div[^>]+class="videotab[^"]*"[^>]+vid="(\d+)"', webpage)
 154             if tabbed_videos:
 155                 return tabbed_videos, presumptive_id, upload_date
 156
 157             MEDIA_ID_REGEXES = [
 158                 r"div\s*:\s*'videoembed'\s*,\s*mediaid\s*:\s*'(\d+)'",  # frontline video embed
 159                 r'class="coveplayerid">([^<]+)<',                       # coveplayer
 160                 r'<input type="hidden" id="pbs_video_id_[0-9]+" value="([0-9]+)"/>',  # jwplayer
 161             ]
 162
 163             media_id = self._search_regex(
 164                 MEDIA_ID_REGEXES, webpage, 'media ID', fatal=False, default=None)
 165             if media_id:
 166                 return media_id, presumptive_id, upload_date
 167
 168             url = self._search_regex(
 169                 r'<iframe\s+[^>]*\s+src=["\']([^\'"]+partnerplayer[^\'"]+)["\']',
 170                 webpage, 'player URL')
 171             mobj = re.match(self._VALID_URL, url)
 172
 173         player_id = mobj.group('player_id')
 174         if not display_id:
 175             display_id = player_id
 176         if player_id:
 177             player_page = self._download_webpage(
 178                 url, display_id, note='Downloading player page',
 179                 errnote='Could not download player page')
 180             video_id = self._search_regex(
 181                 r'<div\s+id="video_([0-9]+)"', player_page, 'video ID')
 182         else:
 183             video_id = mobj.group('id')
 184             display_id = video_id
 185
 186         return video_id, display_id, None
 187
 188     def _real_extract(self, url):
 189         video_id, display_id, upload_date = self._extract_webpage(url)
 190
 191         if isinstance(video_id, list):
 192             entries = [self.url_result(
 193                 'http://video.pbs.org/video/%s' % vid_id, 'PBS', vid_id)
 194                 for vid_id in video_id]
 195             return self.playlist_result(entries, display_id)
 196
 197         info = self._download_json(
 198             'http://video.pbs.org/videoInfo/%s?format=json&type=partner' % video_id,
 199             display_id)
 200
 201         formats = []
 202         for encoding_name in ('recommended_encoding', 'alternate_encoding'):
 203             redirect = info.get(encoding_name)
 204             if not redirect:
 205                 continue
 206             redirect_url = redirect.get('url')
 207             if not redirect_url:
 208                 continue
 209
 210             redirect_info = self._download_json(
 211                 redirect_url + '?format=json', display_id,
 212                 'Downloading %s video url info' % encoding_name)
 213
 214             if redirect_info['status'] == 'error':
 215                 if redirect_info['http_code'] == 403:
 216                     message = (
 217                         'The video is not available in your region due to '
 218                         'right restrictions')
 219                 else:
 220                     message = redirect_info['message']
 221                 raise ExtractorError(message, expected=True)
 222
 223             format_url = redirect_info.get('url')
 224             if not format_url:
 225                 continue
 226
 227             if determine_ext(format_url) == 'm3u8':
 228                 formats.extend(self._extract_m3u8_formats(
 229                     format_url, display_id, 'mp4', preference=1, m3u8_id='hls'))
 230             else:
 231                 formats.append({
 232                     'url': format_url,
 233                     'format_id': redirect.get('eeid'),
 234                 })
 235         self._sort_formats(formats)
 236
 237         rating_str = info.get('rating')
 238         if rating_str is not None:
 239             rating_str = rating_str.rpartition('-')[2]
 240         age_limit = US_RATINGS.get(rating_str)
 241
 242         subtitles = {}
 243         closed_captions_url = info.get('closed_captions_url')
 244         if closed_captions_url:
 245             subtitles['en'] = [{
 246                 'ext': 'ttml',
 247                 'url': closed_captions_url,
 248             }]
 249
 250         # info['title'] is often incomplete (e.g. 'Full Episode', 'Episode 5', etc)
 251         # Try turning it to 'program - title' naming scheme if possible
 252         alt_title = info.get('program', {}).get('title')
 253         if alt_title:
 254             info['title'] = alt_title + ' - ' + re.sub(r'^' + alt_title + '[\s\-:]+', '', info['title'])
 255
 256         return {
 257             'id': video_id,
 258             'display_id': display_id,
 259             'title': info['title'],
 260             'description': info['program'].get('description'),
 261             'thumbnail': info.get('image_url'),
 262             'duration': int_or_none(info.get('duration')),
 263             'age_limit': age_limit,
 264             'upload_date': upload_date,
 265             'formats': formats,
 266             'subtitles': subtitles,
 267         }