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