_ Git - youtube-dl/blob - youtube_dl/extractor/mdr.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 from .common import InfoExtractor
   5 from ..compat import compat_urlparse
   6 from ..utils import (
   7     determine_ext,
   8     int_or_none,
   9     parse_duration,
  10     parse_iso8601,
  11     xpath_text,
  12 )
  13
  14
  15 class MDRIE(InfoExtractor):
  16     IE_DESC = 'MDR.DE and KiKA'
  17     _VALID_URL = r'https?://(?:www\.)?(?:mdr|kika)\.de/(?:.*)/[a-z]+(?P<id>\d+)(?:_.+?)?\.html'
  18
  19     _TESTS = [{
  20         # MDR regularily deletes its videos
  21         'url': 'http://www.mdr.de/fakt/video189002.html',
  22         'only_matching': True,
  23     }, {
  24         'url': 'http://www.kika.de/baumhaus/videos/video19636.html',
  25         'md5': '4930515e36b06c111213e80d1e4aad0e',
  26         'info_dict': {
  27             'id': '19636',
  28             'ext': 'mp4',
  29             'title': 'Baumhaus vom 30. Oktober 2015',
  30             'duration': 134,
  31             'uploader': 'KIKA',
  32         },
  33     }, {
  34         'url': 'http://www.kika.de/sendungen/einzelsendungen/weihnachtsprogramm/videos/video8182.html',
  35         'md5': '5fe9c4dd7d71e3b238f04b8fdd588357',
  36         'info_dict': {
  37             'id': '8182',
  38             'ext': 'mp4',
  39             'title': 'Beutolomäus und der geheime Weihnachtswunsch',
  40             'description': 'md5:b69d32d7b2c55cbe86945ab309d39bbd',
  41             'timestamp': 1419047100,
  42             'upload_date': '20141220',
  43             'duration': 4628,
  44             'uploader': 'KIKA',
  45         },
  46     }, {
  47         'url': 'http://www.kika.de/baumhaus/sendungen/video19636_zc-fea7f8a0_zs-4bf89c60.html',
  48         'only_matching': True,
  49     }, {
  50         'url': 'http://www.kika.de/sendungen/einzelsendungen/weihnachtsprogramm/einzelsendung2534.html',
  51         'only_matching': True,
  52     }]
  53
  54     def _real_extract(self, url):
  55         video_id = self._match_id(url)
  56
  57         webpage = self._download_webpage(url, video_id)
  58
  59         data_url = self._search_regex(
  60             r'dataURL\s*:\s*(["\'])(?P<url>/.+/(?:video|audio)[0-9]+-avCustom\.xml)\1',
  61             webpage, 'data url', group='url')
  62
  63         doc = self._download_xml(
  64             compat_urlparse.urljoin(url, data_url), video_id)
  65
  66         title = (xpath_text(doc, './title', 'title', default=None) or
  67                  xpath_text(doc, './broadcast/broadcastName', 'title'))
  68
  69         formats = []
  70         processed_urls = []
  71         for asset in doc.findall('./assets/asset'):
  72             for source in (
  73                     'progressiveDownload',
  74                     'dynamicHttpStreamingRedirector',
  75                     'adaptiveHttpStreamingRedirector'):
  76                 url_el = asset.find('./%sUrl' % source)
  77                 if url_el is None:
  78                     continue
  79
  80                 video_url = url_el.text
  81                 if video_url in processed_urls:
  82                     continue
  83
  84                 processed_urls.append(video_url)
  85
  86                 vbr = int_or_none(xpath_text(asset, './bitrateVideo', 'vbr'), 1000)
  87                 abr = int_or_none(xpath_text(asset, './bitrateAudio', 'abr'), 1000)
  88
  89                 url_formats = []
  90
  91                 ext = determine_ext(url_el.text)
  92                 if ext == 'm3u8':
  93                     url_formats = self._extract_m3u8_formats(
  94                         video_url, video_id, 'mp4', entry_protocol='m3u8_native',
  95                         preference=0, m3u8_id='HLS', fatal=False)
  96                 elif ext == 'f4m':
  97                     url_formats = self._extract_f4m_formats(
  98                         video_url + '?hdcore=3.7.0&plugin=aasp-3.7.0.39.44', video_id,
  99                         preference=0, f4m_id='HDS', fatal=False)
 100                 else:
 101                     media_type = xpath_text(asset, './mediaType', 'media type', default='MP4')
 102                     vbr = int_or_none(xpath_text(asset, './bitrateVideo', 'vbr'), 1000)
 103                     abr = int_or_none(xpath_text(asset, './bitrateAudio', 'abr'), 1000)
 104                     filesize = int_or_none(xpath_text(asset, './fileSize', 'file size'))
 105
 106                     f = {
 107                         'url': video_url,
 108                         'format_id': '%s-%d' % (media_type, vbr or abr),
 109                         'filesize': filesize,
 110                         'abr': abr,
 111                         'preference': 1,
 112                     }
 113
 114                     if vbr:
 115                         width = int_or_none(xpath_text(asset, './frameWidth', 'width'))
 116                         height = int_or_none(xpath_text(asset, './frameHeight', 'height'))
 117                         f.update({
 118                             'vbr': vbr,
 119                             'width': width,
 120                             'height': height,
 121                         })
 122
 123                     url_formats.append(f)
 124
 125                 if not vbr:
 126                     for f in url_formats:
 127                         abr = f.get('tbr') or abr
 128                         if 'tbr' in f:
 129                             del f['tbr']
 130                         f.update({
 131                             'abr': abr,
 132                             'vcodec': 'none',
 133                         })
 134
 135                 if url_formats:
 136                     formats.extend(url_formats)
 137         self._sort_formats(formats)
 138
 139         description = xpath_text(doc, './broadcast/broadcastDescription', 'description')
 140         timestamp = parse_iso8601(
 141             xpath_text(doc, './broadcast/broadcastDate', 'timestamp', default=None) or
 142             xpath_text(doc, './broadcast/broadcastStartDate', 'timestamp', default=None))
 143         duration = parse_duration(xpath_text(doc, './duration', 'duration'))
 144         uploader = xpath_text(doc, './rights', 'uploader')
 145
 146         return {
 147             'id': video_id,
 148             'title': title,
 149             'description': description,
 150             'timestamp': timestamp,
 151             'duration': duration,
 152             'uploader': uploader,
 153             'formats': formats,
 154         }