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/anselm-gruen-114.html',
  21             'md5': 'c4f83cf0f023ba5875aba0bf46860df2',
  22             'info_dict': {
  23                 'id': '2c8d81c5-6fb7-4a74-88d4-e768e5856532',
  24                 'ext': 'mp4',
  25                 'title': 'Feiern und Verzichten',
  26                 'description': 'Anselm Grün: Feiern und Verzichten',
  27                 'uploader': 'BR/Birgit Baier',
  28                 'upload_date': '20140301',
  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             }
  40         },
  41         {
  42             'url': 'http://www.br.de/nachrichten/schaeuble-haushaltsentwurf-bundestag-100.html',
  43             'md5': '3db0df1a9a9cd9fa0c70e6ea8aa8e820',
  44             'info_dict': {
  45                 'id': 'c6aae3de-2cf9-43f2-957f-f17fef9afaab',
  46                 'ext': 'aac',
  47                 'title': '"Keine neuen Schulden im nächsten Jahr"',
  48                 'description': 'Haushaltsentwurf: "Keine neuen Schulden im nächsten Jahr"',
  49             }
  50         },
  51         {
  52             'url': 'http://www.br.de/radio/bayern1/service/team/videos/team-video-erdelt100.html',
  53             'md5': 'dbab0aef2e047060ea7a21fc1ce1078a',
  54             'info_dict': {
  55                 'id': '6ba73750-d405-45d3-861d-1ce8c524e059',
  56                 'ext': 'mp4',
  57                 'title': 'Umweltbewusster Häuslebauer',
  58                 'description': 'Uwe Erdelt: Umweltbewusster Häuslebauer',
  59             }
  60         },
  61         {
  62             'url': 'http://www.br.de/fernsehen/br-alpha/sendungen/kant-fuer-anfaenger/kritik-der-reinen-vernunft/kant-kritik-01-metaphysik100.html',
  63             'md5': '23bca295f1650d698f94fc570977dae3',
  64             'info_dict': {
  65                 'id': 'd982c9ce-8648-4753-b358-98abb8aec43d',
  66                 'ext': 'mp4',
  67                 'title': 'Folge 1 - Metaphysik',
  68                 'description': 'Kant für Anfänger: Folge 1 - Metaphysik',
  69                 'uploader': 'Eva Maria Steimle',
  70                 'upload_date': '20140117',
  71             }
  72         },
  73     ]
  74
  75     def _real_extract(self, url):
  76         mobj = re.match(self._VALID_URL, url)
  77         display_id = mobj.group('id')
  78         page = self._download_webpage(url, display_id)
  79         xml_url = self._search_regex(
  80             r"return BRavFramework\.register\(BRavFramework\('avPlayer_(?:[a-f0-9-]{36})'\)\.setup\({dataURL:'(/(?:[a-z0-9\-]+/)+[a-z0-9/~_.-]+)'}\)\);", page, 'XMLURL')
  81         xml = self._download_xml(self._BASE_URL + xml_url, None)
  82
  83         medias = []
  84
  85         for xml_media in xml.findall('video') + xml.findall('audio'):
  86             media = {
  87                 'id': xml_media.get('externalId'),
  88                 'title': xml_media.find('title').text,
  89                 'formats': self._extract_formats(xml_media.find('assets')),
  90                 'thumbnails': self._extract_thumbnails(xml_media.find('teaserImage/variants')),
  91                 'description': ' '.join(xml_media.find('shareTitle').text.splitlines()),
  92                 'webpage_url': xml_media.find('permalink').text
  93             }
  94             if xml_media.find('author').text:
  95                 media['uploader'] = xml_media.find('author').text
  96             if xml_media.find('broadcastDate').text:
  97                 media['upload_date'] = ''.join(reversed(xml_media.find('broadcastDate').text.split('.')))
  98             medias.append(media)
  99
 100         if len(medias) > 1:
 101             self._downloader.report_warning(
 102                 'found multiple medias; please '
 103                 'report this with the video URL to http://yt-dl.org/bug')
 104         if not medias:
 105             raise ExtractorError('No media entries found')
 106         return medias[0]
 107
 108     def _extract_formats(self, assets):
 109
 110         def text_or_none(asset, tag):
 111             elem = asset.find(tag)
 112             return None if elem is None else elem.text
 113
 114         formats = [{
 115             'url': text_or_none(asset, 'downloadUrl'),
 116             'ext': text_or_none(asset, 'mediaType'),
 117             'format_id': asset.get('type'),
 118             'width': int_or_none(text_or_none(asset, 'frameWidth')),
 119             'height': int_or_none(text_or_none(asset, 'frameHeight')),
 120             'tbr': int_or_none(text_or_none(asset, 'bitrateVideo')),
 121             'abr': int_or_none(text_or_none(asset, 'bitrateAudio')),
 122             'vcodec': text_or_none(asset, 'codecVideo'),
 123             'acodec': text_or_none(asset, 'codecAudio'),
 124             'container': text_or_none(asset, 'mediaType'),
 125             'filesize': int_or_none(text_or_none(asset, 'size')),
 126         } for asset in assets.findall('asset')
 127             if asset.find('downloadUrl') is not None]
 128
 129         self._sort_formats(formats)
 130         return formats
 131
 132     def _extract_thumbnails(self, variants):
 133         thumbnails = [{
 134             'url': self._BASE_URL + variant.find('url').text,
 135             'width': int_or_none(variant.find('width').text),
 136             'height': int_or_none(variant.find('height').text),
 137         } for variant in variants.findall('variant')]
 138         thumbnails.sort(key=lambda x: x['width'] * x['height'], reverse=True)
 139         return thumbnails