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                 'uploader': 'BR',
  30                 'upload_date': '20140802',
  31             }
  32         },
  33         {
  34             'url': 'http://www.br.de/nachrichten/schaeuble-haushaltsentwurf-bundestag-100.html',
  35             'md5': '3db0df1a9a9cd9fa0c70e6ea8aa8e820',
  36             'info_dict': {
  37                 'id': 'c6aae3de-2cf9-43f2-957f-f17fef9afaab',
  38                 'ext': 'aac',
  39                 'title': '"Keine neuen Schulden im nächsten Jahr"',
  40                 'description': 'Haushaltsentwurf: "Keine neuen Schulden im nächsten Jahr"',
  41                 'duration': 64,
  42             }
  43         },
  44         {
  45             'url': 'http://www.br.de/radio/bayern1/service/team/videos/team-video-erdelt100.html',
  46             'md5': 'dbab0aef2e047060ea7a21fc1ce1078a',
  47             'info_dict': {
  48                 'id': '6ba73750-d405-45d3-861d-1ce8c524e059',
  49                 'ext': 'mp4',
  50                 'title': 'Umweltbewusster Häuslebauer',
  51                 'description': 'Uwe Erdelt: Umweltbewusster Häuslebauer',
  52                 'duration': 116,
  53             }
  54         },
  55         {
  56             'url': 'http://www.br.de/fernsehen/br-alpha/sendungen/kant-fuer-anfaenger/kritik-der-reinen-vernunft/kant-kritik-01-metaphysik100.html',
  57             'md5': '23bca295f1650d698f94fc570977dae3',
  58             'info_dict': {
  59                 'id': 'd982c9ce-8648-4753-b358-98abb8aec43d',
  60                 'ext': 'mp4',
  61                 'title': 'Folge 1 - Metaphysik',
  62                 'description': 'Kant für Anfänger: Folge 1 - Metaphysik',
  63                 'duration': 893,
  64                 'uploader': 'Eva Maria Steimle',
  65                 'upload_date': '20140117',
  66             }
  67         },
  68     ]
  69
  70     def _real_extract(self, url):
  71         display_id = self._match_id(url)
  72         page = self._download_webpage(url, display_id)
  73         xml_url = self._search_regex(
  74             r"return BRavFramework\.register\(BRavFramework\('avPlayer_(?:[a-f0-9-]{36})'\)\.setup\({dataURL:'(/(?:[a-z0-9\-]+/)+[a-z0-9/~_.-]+)'}\)\);", page, 'XMLURL')
  75         xml = self._download_xml(self._BASE_URL + xml_url, None)
  76
  77         medias = []
  78
  79         for xml_media in xml.findall('video') + xml.findall('audio'):
  80             media = {
  81                 'id': xml_media.get('externalId'),
  82                 'title': xml_media.find('title').text,
  83                 'duration': parse_duration(xml_media.find('duration').text),
  84                 'formats': self._extract_formats(xml_media.find('assets')),
  85                 'thumbnails': self._extract_thumbnails(xml_media.find('teaserImage/variants')),
  86                 'description': ' '.join(xml_media.find('shareTitle').text.splitlines()),
  87                 'webpage_url': xml_media.find('permalink').text
  88             }
  89             if xml_media.find('author').text:
  90                 media['uploader'] = xml_media.find('author').text
  91             if xml_media.find('broadcastDate').text:
  92                 media['upload_date'] = ''.join(reversed(xml_media.find('broadcastDate').text.split('.')))
  93             medias.append(media)
  94
  95         if len(medias) > 1:
  96             self._downloader.report_warning(
  97                 'found multiple medias; please '
  98                 'report this with the video URL to http://yt-dl.org/bug')
  99         if not medias:
 100             raise ExtractorError('No media entries found')
 101         return medias[0]
 102
 103     def _extract_formats(self, assets):
 104
 105         def text_or_none(asset, tag):
 106             elem = asset.find(tag)
 107             return None if elem is None else elem.text
 108
 109         formats = [{
 110             'url': text_or_none(asset, 'downloadUrl'),
 111             'ext': text_or_none(asset, 'mediaType'),
 112             'format_id': asset.get('type'),
 113             'width': int_or_none(text_or_none(asset, 'frameWidth')),
 114             'height': int_or_none(text_or_none(asset, 'frameHeight')),
 115             'tbr': int_or_none(text_or_none(asset, 'bitrateVideo')),
 116             'abr': int_or_none(text_or_none(asset, 'bitrateAudio')),
 117             'vcodec': text_or_none(asset, 'codecVideo'),
 118             'acodec': text_or_none(asset, 'codecAudio'),
 119             'container': text_or_none(asset, 'mediaType'),
 120             'filesize': int_or_none(text_or_none(asset, 'size')),
 121         } for asset in assets.findall('asset')
 122             if asset.find('downloadUrl') is not None]
 123
 124         self._sort_formats(formats)
 125         return formats
 126
 127     def _extract_thumbnails(self, variants):
 128         thumbnails = [{
 129             'url': self._BASE_URL + variant.find('url').text,
 130             'width': int_or_none(variant.find('width').text),
 131             'height': int_or_none(variant.find('height').text),
 132         } for variant in variants.findall('variant')]
 133         thumbnails.sort(key=lambda x: x['width'] * x['height'], reverse=True)
 134         return thumbnails