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 )
  11
  12
  13 class BRIE(InfoExtractor):
  14     IE_DESC = 'Bayerischer Rundfunk Mediathek'
  15     _VALID_URL = r'https?://(?:www\.)?br\.de/(?:[a-z0-9\-]+/)+(?P<id>[a-z0-9\-]+)\.html'
  16     _BASE_URL = 'http://www.br.de'
  17
  18     _TESTS = [
  19         {
  20             'url': 'http://www.br.de/mediathek/video/sendungen/heimatsound/heimatsound-festival-2014-trailer-100.html',
  21             'md5': '93556dd2bcb2948d9259f8670c516d59',
  22             'info_dict': {
  23                 'id': '25e279aa-1ffd-40fd-9955-5325bd48a53a',
  24                 'ext': 'mp4',
  25                 'title': 'Am 1. und 2. August in Oberammergau',
  26                 'description': 'md5:dfd224e5aa6819bc1fcbb7826a932021',
  27             }
  28         },
  29         {
  30             'url': 'http://www.br.de/mediathek/video/sendungen/unter-unserem-himmel/unter-unserem-himmel-alpen-ueber-den-pass-100.html',
  31             'md5': 'ab451b09d861dbed7d7cc9ab0be19ebe',
  32             'info_dict': {
  33                 'id': '2c060e69-3a27-4e13-b0f0-668fac17d812',
  34                 'ext': 'mp4',
  35                 'title': 'Über den Pass',
  36                 'description': 'Die Eroberung der Alpen: Über den Pass',
  37             }
  38         },
  39         {
  40             'url': 'http://www.br.de/nachrichten/schaeuble-haushaltsentwurf-bundestag-100.html',
  41             'md5': '3db0df1a9a9cd9fa0c70e6ea8aa8e820',
  42             'info_dict': {
  43                 'id': 'c6aae3de-2cf9-43f2-957f-f17fef9afaab',
  44                 'ext': 'aac',
  45                 'title': '"Keine neuen Schulden im nächsten Jahr"',
  46                 'description': 'Haushaltsentwurf: "Keine neuen Schulden im nächsten Jahr"',
  47             }
  48         },
  49         {
  50             'url': 'http://www.br.de/radio/bayern1/service/team/videos/team-video-erdelt100.html',
  51             'md5': 'dbab0aef2e047060ea7a21fc1ce1078a',
  52             'info_dict': {
  53                 'id': '6ba73750-d405-45d3-861d-1ce8c524e059',
  54                 'ext': 'mp4',
  55                 'title': 'Umweltbewusster Häuslebauer',
  56                 'description': 'Uwe Erdelt: Umweltbewusster Häuslebauer',
  57             }
  58         },
  59         {
  60             'url': 'http://www.br.de/fernsehen/br-alpha/sendungen/kant-fuer-anfaenger/kritik-der-reinen-vernunft/kant-kritik-01-metaphysik100.html',
  61             'md5': '23bca295f1650d698f94fc570977dae3',
  62             'info_dict': {
  63                 'id': 'd982c9ce-8648-4753-b358-98abb8aec43d',
  64                 'ext': 'mp4',
  65                 'title': 'Folge 1 - Metaphysik',
  66                 'description': 'Kant für Anfänger: Folge 1 - Metaphysik',
  67                 'uploader': 'Eva Maria Steimle',
  68                 'upload_date': '20140117',
  69             }
  70         },
  71     ]
  72
  73     def _real_extract(self, url):
  74         mobj = re.match(self._VALID_URL, url)
  75         display_id = mobj.group('id')
  76         page = self._download_webpage(url, display_id)
  77         xml_url = self._search_regex(
  78             r"return BRavFramework\.register\(BRavFramework\('avPlayer_(?:[a-f0-9-]{36})'\)\.setup\({dataURL:'(/(?:[a-z0-9\-]+/)+[a-z0-9/~_.-]+)'}\)\);", page, 'XMLURL')
  79         xml = self._download_xml(self._BASE_URL + xml_url, None)
  80
  81         medias = []
  82
  83         for xml_media in xml.findall('video') + xml.findall('audio'):
  84             media = {
  85                 'id': xml_media.get('externalId'),
  86                 'title': xml_media.find('title').text,
  87                 'formats': self._extract_formats(xml_media.find('assets')),
  88                 'thumbnails': self._extract_thumbnails(xml_media.find('teaserImage/variants')),
  89                 'description': ' '.join(xml_media.find('shareTitle').text.splitlines()),
  90                 'webpage_url': xml_media.find('permalink').text
  91             }
  92             if xml_media.find('author').text:
  93                 media['uploader'] = xml_media.find('author').text
  94             if xml_media.find('broadcastDate').text:
  95                 media['upload_date'] = ''.join(reversed(xml_media.find('broadcastDate').text.split('.')))
  96             medias.append(media)
  97
  98         if len(medias) > 1:
  99             self._downloader.report_warning(
 100                 'found multiple medias; please '
 101                 'report this with the video URL to http://yt-dl.org/bug')
 102         if not medias:
 103             raise ExtractorError('No media entries found')
 104         return medias[0]
 105
 106     def _extract_formats(self, assets):
 107
 108         def text_or_none(asset, tag):
 109             elem = asset.find(tag)
 110             return None if elem is None else elem.text
 111
 112         formats = [{
 113             'url': text_or_none(asset, 'downloadUrl'),
 114             'ext': text_or_none(asset, 'mediaType'),
 115             'format_id': asset.get('type'),
 116             'width': int_or_none(text_or_none(asset, 'frameWidth')),
 117             'height': int_or_none(text_or_none(asset, 'frameHeight')),
 118             'tbr': int_or_none(text_or_none(asset, 'bitrateVideo')),
 119             'abr': int_or_none(text_or_none(asset, 'bitrateAudio')),
 120             'vcodec': text_or_none(asset, 'codecVideo'),
 121             'acodec': text_or_none(asset, 'codecAudio'),
 122             'container': text_or_none(asset, 'mediaType'),
 123             'filesize': int_or_none(text_or_none(asset, 'size')),
 124         } for asset in assets.findall('asset')
 125             if asset.find('downloadUrl') is not None]
 126
 127         self._sort_formats(formats)
 128         return formats
 129
 130     def _extract_thumbnails(self, variants):
 131         thumbnails = [{
 132             'url': self._BASE_URL + variant.find('url').text,
 133             'width': int_or_none(variant.find('width').text),
 134             'height': int_or_none(variant.find('height').text),
 135         } for variant in variants.findall('variant')]
 136         thumbnails.sort(key=lambda x: x['width'] * x['height'], reverse=True)
 137         return thumbnails