yt_dlp/extractor/br.py

   1 from .common import InfoExtractor
   2 from ..utils import (
   3     ExtractorError,
   4     int_or_none,
   5     parse_duration,
   6     xpath_element,
   7     xpath_text,
   8 )
   9
  10
  11 class BRIE(InfoExtractor):
  12     _WORKING = False
  13     IE_DESC = 'Bayerischer Rundfunk'
  14     _VALID_URL = r'(?P<base_url>https?://(?:www\.)?br(?:-klassik)?\.de)/(?:[a-z0-9\-_]+/)+(?P<id>[a-z0-9\-_]+)\.html'
  15
  16     _TESTS = [
  17         {
  18             'url': 'http://www.br.de/mediathek/video/sendungen/abendschau/betriebliche-altersvorsorge-104.html',
  19             'md5': '83a0477cf0b8451027eb566d88b51106',
  20             'info_dict': {
  21                 'id': '48f656ef-287e-486f-be86-459122db22cc',
  22                 'ext': 'mp4',
  23                 'title': 'Die böse Überraschung',
  24                 'description': 'md5:ce9ac81b466ce775b8018f6801b48ac9',
  25                 'duration': 180,
  26                 'uploader': 'Reinhard Weber',
  27                 'upload_date': '20150422',
  28             },
  29             'skip': '404 not found',
  30         },
  31         {
  32             'url': 'http://www.br.de/nachrichten/oberbayern/inhalt/muenchner-polizeipraesident-schreiber-gestorben-100.html',
  33             'md5': 'af3a3a4aa43ff0ce6a89504c67f427ef',
  34             'info_dict': {
  35                 'id': 'a4b83e34-123d-4b81-9f4e-c0d3121a4e05',
  36                 'ext': 'flv',
  37                 'title': 'Manfred Schreiber ist tot',
  38                 'description': 'md5:b454d867f2a9fc524ebe88c3f5092d97',
  39                 'duration': 26,
  40             },
  41             'skip': '404 not found',
  42         },
  43         {
  44             'url': 'https://www.br-klassik.de/audio/peeping-tom-premierenkritik-dance-festival-muenchen-100.html',
  45             'md5': '8b5b27c0b090f3b35eac4ab3f7a73d3d',
  46             'info_dict': {
  47                 'id': '74c603c9-26d3-48bb-b85b-079aeed66e0b',
  48                 'ext': 'aac',
  49                 'title': 'Kurzweilig und sehr bewegend',
  50                 'description': 'md5:0351996e3283d64adeb38ede91fac54e',
  51                 'duration': 296,
  52             },
  53             'skip': '404 not found',
  54         },
  55         {
  56             'url': 'http://www.br.de/radio/bayern1/service/team/videos/team-video-erdelt100.html',
  57             'md5': 'dbab0aef2e047060ea7a21fc1ce1078a',
  58             'info_dict': {
  59                 'id': '6ba73750-d405-45d3-861d-1ce8c524e059',
  60                 'ext': 'mp4',
  61                 'title': 'Umweltbewusster Häuslebauer',
  62                 'description': 'md5:d52dae9792d00226348c1dbb13c9bae2',
  63                 'duration': 116,
  64             },
  65         },
  66         {
  67             'url': 'http://www.br.de/fernsehen/br-alpha/sendungen/kant-fuer-anfaenger/kritik-der-reinen-vernunft/kant-kritik-01-metaphysik100.html',
  68             'md5': '23bca295f1650d698f94fc570977dae3',
  69             'info_dict': {
  70                 'id': 'd982c9ce-8648-4753-b358-98abb8aec43d',
  71                 'ext': 'mp4',
  72                 'title': 'Folge 1 - Metaphysik',
  73                 'description': 'md5:bb659990e9e59905c3d41e369db1fbe3',
  74                 'duration': 893,
  75                 'uploader': 'Eva Maria Steimle',
  76                 'upload_date': '20170208',
  77             },
  78         },
  79     ]
  80
  81     def _real_extract(self, url):
  82         base_url, display_id = self._match_valid_url(url).groups()
  83         page = self._download_webpage(url, display_id)
  84         xml_url = self._search_regex(
  85             r"return BRavFramework\.register\(BRavFramework\('avPlayer_(?:[a-f0-9-]{36})'\)\.setup\({dataURL:'(/(?:[a-z0-9\-]+/)+[a-z0-9/~_.-]+)'}\)\);", page, 'XMLURL')
  86         xml = self._download_xml(base_url + xml_url, display_id)
  87
  88         medias = []
  89
  90         for xml_media in xml.findall('video') + xml.findall('audio'):
  91             media_id = xml_media.get('externalId')
  92             media = {
  93                 'id': media_id,
  94                 'title': xpath_text(xml_media, 'title', 'title', True),
  95                 'duration': parse_duration(xpath_text(xml_media, 'duration')),
  96                 'formats': self._extract_formats(xpath_element(
  97                     xml_media, 'assets'), media_id),
  98                 'thumbnails': self._extract_thumbnails(xpath_element(
  99                     xml_media, 'teaserImage/variants'), base_url),
 100                 'description': xpath_text(xml_media, 'desc'),
 101                 'webpage_url': xpath_text(xml_media, 'permalink'),
 102                 'uploader': xpath_text(xml_media, 'author'),
 103             }
 104             broadcast_date = xpath_text(xml_media, 'broadcastDate')
 105             if broadcast_date:
 106                 media['upload_date'] = ''.join(reversed(broadcast_date.split('.')))
 107             medias.append(media)
 108
 109         if len(medias) > 1:
 110             self.report_warning(
 111                 'found multiple medias; please '
 112                 'report this with the video URL to http://yt-dl.org/bug')
 113         if not medias:
 114             raise ExtractorError('No media entries found')
 115         return medias[0]
 116
 117     def _extract_formats(self, assets, media_id):
 118         formats = []
 119         for asset in assets.findall('asset'):
 120             format_url = xpath_text(asset, ['downloadUrl', 'url'])
 121             asset_type = asset.get('type')
 122             if asset_type.startswith('HDS'):
 123                 formats.extend(self._extract_f4m_formats(
 124                     format_url + '?hdcore=3.2.0', media_id, f4m_id='hds', fatal=False))
 125             elif asset_type.startswith('HLS'):
 126                 formats.extend(self._extract_m3u8_formats(
 127                     format_url, media_id, 'mp4', 'm3u8_native', m3u8_id='hds', fatal=False))
 128             else:
 129                 format_info = {
 130                     'ext': xpath_text(asset, 'mediaType'),
 131                     'width': int_or_none(xpath_text(asset, 'frameWidth')),
 132                     'height': int_or_none(xpath_text(asset, 'frameHeight')),
 133                     'tbr': int_or_none(xpath_text(asset, 'bitrateVideo')),
 134                     'abr': int_or_none(xpath_text(asset, 'bitrateAudio')),
 135                     'vcodec': xpath_text(asset, 'codecVideo'),
 136                     'acodec': xpath_text(asset, 'codecAudio'),
 137                     'container': xpath_text(asset, 'mediaType'),
 138                     'filesize': int_or_none(xpath_text(asset, 'size')),
 139                 }
 140                 format_url = self._proto_relative_url(format_url)
 141                 if format_url:
 142                     http_format_info = format_info.copy()
 143                     http_format_info.update({
 144                         'url': format_url,
 145                         'format_id': f'http-{asset_type}',
 146                     })
 147                     formats.append(http_format_info)
 148                 server_prefix = xpath_text(asset, 'serverPrefix')
 149                 if server_prefix:
 150                     rtmp_format_info = format_info.copy()
 151                     rtmp_format_info.update({
 152                         'url': server_prefix,
 153                         'play_path': xpath_text(asset, 'fileName'),
 154                         'format_id': f'rtmp-{asset_type}',
 155                     })
 156                     formats.append(rtmp_format_info)
 157         return formats
 158
 159     def _extract_thumbnails(self, variants, base_url):
 160         thumbnails = [{
 161             'url': base_url + xpath_text(variant, 'url'),
 162             'width': int_or_none(xpath_text(variant, 'width')),
 163             'height': int_or_none(xpath_text(variant, 'height')),
 164         } for variant in variants.findall('variant') if xpath_text(variant, 'url')]
 165         thumbnails.sort(key=lambda x: x['width'] * x['height'], reverse=True)
 166         return thumbnails