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 regularly deletes its videos
  21         'url': 'http://www.mdr.de/fakt/video189002.html',
  22         'only_matching': True,
  23     }, {
  24         # audio
  25         'url': 'http://www.mdr.de/kultur/audio1312272_zc-15948bad_zs-86171fdd.html',
  26         'md5': '64c4ee50f0a791deb9479cd7bbe9d2fa',
  27         'info_dict': {
  28             'id': '1312272',
  29             'ext': 'mp3',
  30             'title': 'Feuilleton vom 30. Oktober 2015',
  31             'duration': 250,
  32             'uploader': 'MITTELDEUTSCHER RUNDFUNK',
  33         },
  34         'skip': '404 not found',
  35     }, {
  36         'url': 'http://www.kika.de/baumhaus/videos/video19636.html',
  37         'md5': '4930515e36b06c111213e80d1e4aad0e',
  38         'info_dict': {
  39             'id': '19636',
  40             'ext': 'mp4',
  41             'title': 'Baumhaus vom 30. Oktober 2015',
  42             'duration': 134,
  43             'uploader': 'KIKA',
  44         },
  45         'skip': '404 not found',
  46     }, {
  47         'url': 'http://www.kika.de/sendungen/einzelsendungen/weihnachtsprogramm/videos/video8182.html',
  48         'md5': '5fe9c4dd7d71e3b238f04b8fdd588357',
  49         'info_dict': {
  50             'id': '8182',
  51             'ext': 'mp4',
  52             'title': 'Beutolomäus und der geheime Weihnachtswunsch',
  53             'description': 'md5:b69d32d7b2c55cbe86945ab309d39bbd',
  54             'timestamp': 1482541200,
  55             'upload_date': '20161224',
  56             'duration': 4628,
  57             'uploader': 'KIKA',
  58         },
  59     }, {
  60         # audio with alternative playerURL pattern
  61         'url': 'http://www.mdr.de/kultur/videos-und-audios/audio-radio/operation-mindfuck-robert-wilson100.html',
  62         'info_dict': {
  63             'id': '100',
  64             'ext': 'mp4',
  65             'title': 'Feature: Operation Mindfuck - Robert Anton Wilson',
  66             'duration': 3239,
  67             'uploader': 'MITTELDEUTSCHER RUNDFUNK',
  68         },
  69     }, {
  70         'url': 'http://www.kika.de/baumhaus/sendungen/video19636_zc-fea7f8a0_zs-4bf89c60.html',
  71         'only_matching': True,
  72     }, {
  73         'url': 'http://www.kika.de/sendungen/einzelsendungen/weihnachtsprogramm/einzelsendung2534.html',
  74         'only_matching': True,
  75     }, {
  76         'url': 'http://www.mdr.de/mediathek/mdr-videos/a/video-1334.html',
  77         'only_matching': True,
  78     }]
  79
  80     def _real_extract(self, url):
  81         video_id = self._match_id(url)
  82
  83         webpage = self._download_webpage(url, video_id)
  84
  85         data_url = self._search_regex(
  86             r'(?:dataURL|playerXml(?:["\'])?)\s*:\s*(["\'])(?P<url>.+?-avCustom\.xml)\1',
  87             webpage, 'data url', group='url').replace(r'\/', '/')
  88
  89         doc = self._download_xml(
  90             compat_urlparse.urljoin(url, data_url), video_id)
  91
  92         title = xpath_text(doc, ['./title', './broadcast/broadcastName'], 'title', fatal=True)
  93
  94         formats = []
  95         processed_urls = []
  96         for asset in doc.findall('./assets/asset'):
  97             for source in (
  98                     'progressiveDownload',
  99                     'dynamicHttpStreamingRedirector',
 100                     'adaptiveHttpStreamingRedirector'):
 101                 url_el = asset.find('./%sUrl' % source)
 102                 if url_el is None:
 103                     continue
 104
 105                 video_url = url_el.text
 106                 if video_url in processed_urls:
 107                     continue
 108
 109                 processed_urls.append(video_url)
 110
 111                 vbr = int_or_none(xpath_text(asset, './bitrateVideo', 'vbr'), 1000)
 112                 abr = int_or_none(xpath_text(asset, './bitrateAudio', 'abr'), 1000)
 113
 114                 ext = determine_ext(url_el.text)
 115                 if ext == 'm3u8':
 116                     url_formats = self._extract_m3u8_formats(
 117                         video_url, video_id, 'mp4', entry_protocol='m3u8_native',
 118                         preference=0, m3u8_id='HLS', fatal=False)
 119                 elif ext == 'f4m':
 120                     url_formats = self._extract_f4m_formats(
 121                         video_url + '?hdcore=3.7.0&plugin=aasp-3.7.0.39.44', video_id,
 122                         preference=0, f4m_id='HDS', fatal=False)
 123                 else:
 124                     media_type = xpath_text(asset, './mediaType', 'media type', default='MP4')
 125                     vbr = int_or_none(xpath_text(asset, './bitrateVideo', 'vbr'), 1000)
 126                     abr = int_or_none(xpath_text(asset, './bitrateAudio', 'abr'), 1000)
 127                     filesize = int_or_none(xpath_text(asset, './fileSize', 'file size'))
 128
 129                     f = {
 130                         'url': video_url,
 131                         'format_id': '%s-%d' % (media_type, vbr or abr),
 132                         'filesize': filesize,
 133                         'abr': abr,
 134                         'preference': 1,
 135                     }
 136
 137                     if vbr:
 138                         width = int_or_none(xpath_text(asset, './frameWidth', 'width'))
 139                         height = int_or_none(xpath_text(asset, './frameHeight', 'height'))
 140                         f.update({
 141                             'vbr': vbr,
 142                             'width': width,
 143                             'height': height,
 144                         })
 145
 146                     url_formats = [f]
 147
 148                 if not url_formats:
 149                     continue
 150
 151                 if not vbr:
 152                     for f in url_formats:
 153                         abr = f.get('tbr') or abr
 154                         if 'tbr' in f:
 155                             del f['tbr']
 156                         f.update({
 157                             'abr': abr,
 158                             'vcodec': 'none',
 159                         })
 160
 161                 formats.extend(url_formats)
 162
 163         self._sort_formats(formats)
 164
 165         description = xpath_text(doc, './broadcast/broadcastDescription', 'description')
 166         timestamp = parse_iso8601(
 167             xpath_text(
 168                 doc, [
 169                     './broadcast/broadcastDate',
 170                     './broadcast/broadcastStartDate',
 171                     './broadcast/broadcastEndDate'],
 172                 'timestamp', default=None))
 173         duration = parse_duration(xpath_text(doc, './duration', 'duration'))
 174         uploader = xpath_text(doc, './rights', 'uploader')
 175
 176         return {
 177             'id': video_id,
 178             'title': title,
 179             'description': description,
 180             'timestamp': timestamp,
 181             'duration': duration,
 182             'uploader': uploader,
 183             'formats': formats,
 184         }