Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/drtv.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 from .common import InfoExtractor
   5 from ..utils import (
   6     ExtractorError,
   7     int_or_none,
   8     float_or_none,
   9     mimetype2ext,
  10     parse_iso8601,
  11     remove_end,
  12     update_url_query,
  13 )
  14
  15
  16 class DRTVIE(InfoExtractor):
  17     _VALID_URL = r'https?://(?:www\.)?dr\.dk/(?:tv/se|nyheder|radio/ondemand)/(?:[^/]+/)*(?P<id>[\da-z-]+)(?:[/#?]|$)'
  18     IE_NAME = 'drtv'
  19     _TESTS = [{
  20         'url': 'https://www.dr.dk/tv/se/boern/ultra/klassen-ultra/klassen-darlig-taber-10',
  21         'md5': '25e659cccc9a2ed956110a299fdf5983',
  22         'info_dict': {
  23             'id': 'klassen-darlig-taber-10',
  24             'ext': 'mp4',
  25             'title': 'Klassen - Dårlig taber (10)',
  26             'description': 'md5:815fe1b7fa656ed80580f31e8b3c79aa',
  27             'timestamp': 1471991907,
  28             'upload_date': '20160823',
  29             'duration': 606.84,
  30         },
  31         'params': {
  32             'skip_download': True,
  33         },
  34     }, {
  35         'url': 'https://www.dr.dk/nyheder/indland/live-christianias-rydning-af-pusher-street-er-i-gang',
  36         'md5': '2c37175c718155930f939ef59952474a',
  37         'info_dict': {
  38             'id': 'christiania-pusher-street-ryddes-drdkrjpo',
  39             'ext': 'mp4',
  40             'title': 'LIVE Christianias rydning af Pusher Street er i gang',
  41             'description': '- Det er det fedeste, der er sket i 20 år, fortæller christianit til DR Nyheder.',
  42             'timestamp': 1472800279,
  43             'upload_date': '20160902',
  44             'duration': 131.4,
  45         },
  46     }]
  47
  48     def _real_extract(self, url):
  49         video_id = self._match_id(url)
  50
  51         webpage = self._download_webpage(url, video_id)
  52
  53         if '>Programmet er ikke længere tilgængeligt' in webpage:
  54             raise ExtractorError(
  55                 'Video %s is not available' % video_id, expected=True)
  56
  57         video_id = self._search_regex(
  58             (r'data-(?:material-identifier|episode-slug)="([^"]+)"',
  59                 r'data-resource="[^>"]+mu/programcard/expanded/([^"]+)"'),
  60             webpage, 'video id')
  61
  62         programcard = self._download_json(
  63             'http://www.dr.dk/mu/programcard/expanded/%s' % video_id,
  64             video_id, 'Downloading video JSON')
  65         data = programcard['Data'][0]
  66
  67         title = remove_end(self._og_search_title(
  68             webpage, default=None), ' | TV | DR') or data['Title']
  69         description = self._og_search_description(
  70             webpage, default=None) or data.get('Description')
  71
  72         timestamp = parse_iso8601(data.get('CreatedTime'))
  73
  74         thumbnail = None
  75         duration = None
  76
  77         restricted_to_denmark = False
  78
  79         formats = []
  80         subtitles = {}
  81
  82         for asset in data['Assets']:
  83             kind = asset.get('Kind')
  84             if kind == 'Image':
  85                 thumbnail = asset.get('Uri')
  86             elif kind in ('VideoResource', 'AudioResource'):
  87                 duration = float_or_none(asset.get('DurationInMilliseconds'), 1000)
  88                 restricted_to_denmark = asset.get('RestrictedToDenmark')
  89                 spoken_subtitles = asset.get('Target') == 'SpokenSubtitles'
  90                 for link in asset.get('Links', []):
  91                     uri = link.get('Uri')
  92                     if not uri:
  93                         continue
  94                     target = link.get('Target')
  95                     format_id = target or ''
  96                     preference = None
  97                     if spoken_subtitles:
  98                         preference = -1
  99                         format_id += '-spoken-subtitles'
 100                     if target == 'HDS':
 101                         f4m_formats = self._extract_f4m_formats(
 102                             uri + '?hdcore=3.3.0&plugin=aasp-3.3.0.99.43',
 103                             video_id, preference, f4m_id=format_id)
 104                         if kind == 'AudioResource':
 105                             for f in f4m_formats:
 106                                 f['vcodec'] = 'none'
 107                         formats.extend(f4m_formats)
 108                     elif target == 'HLS':
 109                         formats.extend(self._extract_m3u8_formats(
 110                             uri, video_id, 'mp4', entry_protocol='m3u8_native',
 111                             preference=preference, m3u8_id=format_id))
 112                     else:
 113                         bitrate = link.get('Bitrate')
 114                         if bitrate:
 115                             format_id += '-%s' % bitrate
 116                         formats.append({
 117                             'url': uri,
 118                             'format_id': format_id,
 119                             'tbr': int_or_none(bitrate),
 120                             'ext': link.get('FileFormat'),
 121                             'vcodec': 'none' if kind == 'AudioResource' else None,
 122                         })
 123                 subtitles_list = asset.get('SubtitlesList')
 124                 if isinstance(subtitles_list, list):
 125                     LANGS = {
 126                         'Danish': 'da',
 127                     }
 128                     for subs in subtitles_list:
 129                         if not subs.get('Uri'):
 130                             continue
 131                         lang = subs.get('Language') or 'da'
 132                         subtitles.setdefault(LANGS.get(lang, lang), []).append({
 133                             'url': subs['Uri'],
 134                             'ext': mimetype2ext(subs.get('MimeType')) or 'vtt'
 135                         })
 136
 137         if not formats and restricted_to_denmark:
 138             self.raise_geo_restricted(
 139                 'Unfortunately, DR is not allowed to show this program outside Denmark.',
 140                 expected=True)
 141
 142         self._sort_formats(formats)
 143
 144         return {
 145             'id': video_id,
 146             'title': title,
 147             'description': description,
 148             'thumbnail': thumbnail,
 149             'timestamp': timestamp,
 150             'duration': duration,
 151             'formats': formats,
 152             'subtitles': subtitles,
 153         }
 154
 155
 156 class DRTVLiveIE(InfoExtractor):
 157     IE_NAME = 'drtv:live'
 158     _VALID_URL = r'https?://(?:www\.)?dr\.dk/(?:tv|TV)/live/(?P<id>[\da-z-]+)'
 159     _TEST = {
 160         'url': 'https://www.dr.dk/tv/live/dr1',
 161         'info_dict': {
 162             'id': 'dr1',
 163             'ext': 'mp4',
 164             'title': 're:^DR1 [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
 165         },
 166         'params': {
 167             # m3u8 download
 168             'skip_download': True,
 169         },
 170     }
 171
 172     def _real_extract(self, url):
 173         channel_id = self._match_id(url)
 174         channel_data = self._download_json(
 175             'https://www.dr.dk/mu-online/api/1.0/channel/' + channel_id,
 176             channel_id)
 177         title = self._live_title(channel_data['Title'])
 178
 179         formats = []
 180         for streaming_server in channel_data.get('StreamingServers', []):
 181             server = streaming_server.get('Server')
 182             if not server:
 183                 continue
 184             link_type = streaming_server.get('LinkType')
 185             for quality in streaming_server.get('Qualities', []):
 186                 for stream in quality.get('Streams', []):
 187                     stream_path = stream.get('Stream')
 188                     if not stream_path:
 189                         continue
 190                     stream_url = update_url_query(
 191                         '%s/%s' % (server, stream_path), {'b': ''})
 192                     if link_type == 'HLS':
 193                         formats.extend(self._extract_m3u8_formats(
 194                             stream_url, channel_id, 'mp4',
 195                             m3u8_id=link_type, fatal=False, live=True))
 196                     elif link_type == 'HDS':
 197                         formats.extend(self._extract_f4m_formats(update_url_query(
 198                             '%s/%s' % (server, stream_path), {'hdcore': '3.7.0'}),
 199                             channel_id, f4m_id=link_type, fatal=False))
 200         self._sort_formats(formats)
 201
 202         return {
 203             'id': channel_id,
 204             'title': title,
 205             'thumbnail': channel_data.get('PrimaryImageUri'),
 206             'formats': formats,
 207             'is_live': True,
 208         }