Raphaël G. Git Repositories - youtubedl/blob - 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 )
  12
  13
  14 class BRIE(InfoExtractor):
  15     IE_DESC = 'Bayerischer Rundfunk Mediathek'
  16     _VALID_URL = r'https?://(?:www\.)?br\.de/(?:[a-z0-9\-_]+/)+(?P<id>[a-z0-9\-_]+)\.html'
  17     _BASE_URL = 'http://www.br.de'
  18
  19     _TESTS = [
  20         {
  21             'url': 'http://www.br.de/mediathek/video/sendungen/heimatsound/heimatsound-festival-2014-trailer-100.html',
  22             'md5': '93556dd2bcb2948d9259f8670c516d59',
  23             'info_dict': {
  24                 'id': '25e279aa-1ffd-40fd-9955-5325bd48a53a',
  25                 'ext': 'mp4',
  26                 'title': 'Wenn das Traditions-Theater wackelt',
  27                 'description': 'Heimatsound-Festival 2014: Wenn das Traditions-Theater wackelt',
  28                 'duration': 34,
  29             }
  30         },
  31         {
  32             'url': 'http://www.br.de/mediathek/video/sendungen/unter-unserem-himmel/unter-unserem-himmel-alpen-ueber-den-pass-100.html',
  33             'md5': 'ab451b09d861dbed7d7cc9ab0be19ebe',
  34             'info_dict': {
  35                 'id': '2c060e69-3a27-4e13-b0f0-668fac17d812',
  36                 'ext': 'mp4',
  37                 'title': 'Über den Pass',
  38                 'description': 'Die Eroberung der Alpen: Über den Pass',
  39                 'duration': 2588,
  40             }
  41         },
  42         {
  43             'url': 'http://www.br.de/nachrichten/schaeuble-haushaltsentwurf-bundestag-100.html',
  44             'md5': '3db0df1a9a9cd9fa0c70e6ea8aa8e820',
  45             'info_dict': {
  46                 'id': 'c6aae3de-2cf9-43f2-957f-f17fef9afaab',
  47                 'ext': 'aac',
  48                 'title': '"Keine neuen Schulden im nächsten Jahr"',
  49                 'description': 'Haushaltsentwurf: "Keine neuen Schulden im nächsten Jahr"',
  50                 'duration': 64,
  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         mobj = re.match(self._VALID_URL, url)
  81         display_id = mobj.group('id')
  82         page = self._download_webpage(url, display_id)
  83         xml_url = self._search_regex(
  84             r"return BRavFramework\.register\(BRavFramework\('avPlayer_(?:[a-f0-9-]{36})'\)\.setup\({dataURL:'(/(?:[a-z0-9\-]+/)+[a-z0-9/~_.-]+)'}\)\);", page, 'XMLURL')
  85         xml = self._download_xml(self._BASE_URL + xml_url, None)
  86
  87         medias = []
  88
  89         for xml_media in xml.findall('video') + xml.findall('audio'):
  90             media = {
  91                 'id': xml_media.get('externalId'),
  92                 'title': xml_media.find('title').text,
  93                 'duration': parse_duration(xml_media.find('duration').text),
  94                 'formats': self._extract_formats(xml_media.find('assets')),
  95                 'thumbnails': self._extract_thumbnails(xml_media.find('teaserImage/variants')),
  96                 'description': ' '.join(xml_media.find('shareTitle').text.splitlines()),
  97                 'webpage_url': xml_media.find('permalink').text
  98             }
  99             if xml_media.find('author').text:
 100                 media['uploader'] = xml_media.find('author').text
 101             if xml_media.find('broadcastDate').text:
 102                 media['upload_date'] = ''.join(reversed(xml_media.find('broadcastDate').text.split('.')))
 103             medias.append(media)
 104
 105         if len(medias) > 1:
 106             self._downloader.report_warning(
 107                 'found multiple medias; please '
 108                 'report this with the video URL to http://yt-dl.org/bug')
 109         if not medias:
 110             raise ExtractorError('No media entries found')
 111         return medias[0]
 112
 113     def _extract_formats(self, assets):
 114
 115         def text_or_none(asset, tag):
 116             elem = asset.find(tag)
 117             return None if elem is None else elem.text
 118
 119         formats = [{
 120             'url': text_or_none(asset, 'downloadUrl'),
 121             'ext': text_or_none(asset, 'mediaType'),
 122             'format_id': asset.get('type'),
 123             'width': int_or_none(text_or_none(asset, 'frameWidth')),
 124             'height': int_or_none(text_or_none(asset, 'frameHeight')),
 125             'tbr': int_or_none(text_or_none(asset, 'bitrateVideo')),
 126             'abr': int_or_none(text_or_none(asset, 'bitrateAudio')),
 127             'vcodec': text_or_none(asset, 'codecVideo'),
 128             'acodec': text_or_none(asset, 'codecAudio'),
 129             'container': text_or_none(asset, 'mediaType'),
 130             'filesize': int_or_none(text_or_none(asset, 'size')),
 131         } for asset in assets.findall('asset')
 132             if asset.find('downloadUrl') is not None]
 133
 134         self._sort_formats(formats)
 135         return formats
 136
 137     def _extract_thumbnails(self, variants):
 138         thumbnails = [{
 139             'url': self._BASE_URL + variant.find('url').text,
 140             'width': int_or_none(variant.find('width').text),
 141             'height': int_or_none(variant.find('height').text),
 142         } for variant in variants.findall('variant')]
 143         thumbnails.sort(key=lambda x: x['width'] * x['height'], reverse=True)
 144         return thumbnails