Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/br.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     parse_duration,
   9 )
  10
  11
  12 class BRIE(InfoExtractor):
  13     IE_DESC = 'Bayerischer Rundfunk Mediathek'
  14     _VALID_URL = r'https?://(?:www\.)?br\.de/(?:[a-z0-9\-_]+/)+(?P<id>[a-z0-9\-_]+)\.html'
  15     _BASE_URL = 'http://www.br.de'
  16
  17     _TESTS = [
  18         {
  19             'url': 'http://www.br.de/mediathek/video/sendungen/abendschau/betriebliche-altersvorsorge-104.html',
  20             'md5': '83a0477cf0b8451027eb566d88b51106',
  21             'info_dict': {
  22                 'id': '48f656ef-287e-486f-be86-459122db22cc',
  23                 'ext': 'mp4',
  24                 'title': 'Die böse Überraschung',
  25                 'description': 'Betriebliche Altersvorsorge: Die böse Überraschung',
  26                 'duration': 180,
  27                 'uploader': 'Reinhard Weber',
  28                 'upload_date': '20150422',
  29             }
  30         },
  31         {
  32             'url': 'http://www.br.de/nachrichten/oberbayern/inhalt/muenchner-polizeipraesident-schreiber-gestorben-100.html',
  33             'md5': 'a44396d73ab6a68a69a568fae10705bb',
  34             'info_dict': {
  35                 'id': 'a4b83e34-123d-4b81-9f4e-c0d3121a4e05',
  36                 'ext': 'mp4',
  37                 'title': 'Manfred Schreiber ist tot',
  38                 'description': 'Abendschau kompakt: Manfred Schreiber ist tot',
  39                 'duration': 26,
  40             }
  41         },
  42         {
  43             'url': 'http://www.br.de/radio/br-klassik/sendungen/allegro/premiere-urauffuehrung-the-land-2015-dance-festival-muenchen-100.html',
  44             'md5': '8b5b27c0b090f3b35eac4ab3f7a73d3d',
  45             'info_dict': {
  46                 'id': '74c603c9-26d3-48bb-b85b-079aeed66e0b',
  47                 'ext': 'aac',
  48                 'title': 'Kurzweilig und sehr bewegend',
  49                 'description': '"The Land" von Peeping Tom: Kurzweilig und sehr bewegend',
  50                 'duration': 296,
  51             }
  52         },
  53         {
  54             'url': 'http://www.br.de/radio/bayern1/service/team/videos/team-video-erdelt100.html',
  55             'md5': 'dbab0aef2e047060ea7a21fc1ce1078a',
  56             'info_dict': {
  57                 'id': '6ba73750-d405-45d3-861d-1ce8c524e059',
  58                 'ext': 'mp4',
  59                 'title': 'Umweltbewusster Häuslebauer',
  60                 'description': 'Uwe Erdelt: Umweltbewusster Häuslebauer',
  61                 'duration': 116,
  62             }
  63         },
  64         {
  65             'url': 'http://www.br.de/fernsehen/br-alpha/sendungen/kant-fuer-anfaenger/kritik-der-reinen-vernunft/kant-kritik-01-metaphysik100.html',
  66             'md5': '23bca295f1650d698f94fc570977dae3',
  67             'info_dict': {
  68                 'id': 'd982c9ce-8648-4753-b358-98abb8aec43d',
  69                 'ext': 'mp4',
  70                 'title': 'Folge 1 - Metaphysik',
  71                 'description': 'Kant für Anfänger: Folge 1 - Metaphysik',
  72                 'duration': 893,
  73                 'uploader': 'Eva Maria Steimle',
  74                 'upload_date': '20140117',
  75             }
  76         },
  77     ]
  78
  79     def _real_extract(self, url):
  80         display_id = self._match_id(url)
  81         page = self._download_webpage(url, display_id)
  82         xml_url = self._search_regex(
  83             r"return BRavFramework\.register\(BRavFramework\('avPlayer_(?:[a-f0-9-]{36})'\)\.setup\({dataURL:'(/(?:[a-z0-9\-]+/)+[a-z0-9/~_.-]+)'}\)\);", page, 'XMLURL')
  84         xml = self._download_xml(self._BASE_URL + xml_url, None)
  85
  86         medias = []
  87
  88         for xml_media in xml.findall('video') + xml.findall('audio'):
  89             media = {
  90                 'id': xml_media.get('externalId'),
  91                 'title': xml_media.find('title').text,
  92                 'duration': parse_duration(xml_media.find('duration').text),
  93                 'formats': self._extract_formats(xml_media.find('assets')),
  94                 'thumbnails': self._extract_thumbnails(xml_media.find('teaserImage/variants')),
  95                 'description': ' '.join(xml_media.find('shareTitle').text.splitlines()),
  96                 'webpage_url': xml_media.find('permalink').text
  97             }
  98             if xml_media.find('author').text:
  99                 media['uploader'] = xml_media.find('author').text
 100             if xml_media.find('broadcastDate').text:
 101                 media['upload_date'] = ''.join(reversed(xml_media.find('broadcastDate').text.split('.')))
 102             medias.append(media)
 103
 104         if len(medias) > 1:
 105             self._downloader.report_warning(
 106                 'found multiple medias; please '
 107                 'report this with the video URL to http://yt-dl.org/bug')
 108         if not medias:
 109             raise ExtractorError('No media entries found')
 110         return medias[0]
 111
 112     def _extract_formats(self, assets):
 113
 114         def text_or_none(asset, tag):
 115             elem = asset.find(tag)
 116             return None if elem is None else elem.text
 117
 118         formats = [{
 119             'url': text_or_none(asset, 'downloadUrl'),
 120             'ext': text_or_none(asset, 'mediaType'),
 121             'format_id': asset.get('type'),
 122             'width': int_or_none(text_or_none(asset, 'frameWidth')),
 123             'height': int_or_none(text_or_none(asset, 'frameHeight')),
 124             'tbr': int_or_none(text_or_none(asset, 'bitrateVideo')),
 125             'abr': int_or_none(text_or_none(asset, 'bitrateAudio')),
 126             'vcodec': text_or_none(asset, 'codecVideo'),
 127             'acodec': text_or_none(asset, 'codecAudio'),
 128             'container': text_or_none(asset, 'mediaType'),
 129             'filesize': int_or_none(text_or_none(asset, 'size')),
 130         } for asset in assets.findall('asset')
 131             if asset.find('downloadUrl') is not None]
 132
 133         self._sort_formats(formats)
 134         return formats
 135
 136     def _extract_thumbnails(self, variants):
 137         thumbnails = [{
 138             'url': self._BASE_URL + variant.find('url').text,
 139             'width': int_or_none(variant.find('width').text),
 140             'height': int_or_none(variant.find('height').text),
 141         } for variant in variants.findall('variant')]
 142         thumbnails.sort(key=lambda x: x['width'] * x['height'], reverse=True)
 143         return thumbnails