yt_dlp/extractor/mdr.py

   1 import urllib.parse
   2
   3 from .common import InfoExtractor
   4 from ..utils import (
   5     determine_ext,
   6     int_or_none,
   7     join_nonempty,
   8     parse_duration,
   9     parse_iso8601,
  10     url_or_none,
  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     _GEO_COUNTRIES = ['DE']
  20
  21     _TESTS = [{
  22         # MDR regularly deletes its videos
  23         'url': 'http://www.mdr.de/fakt/video189002.html',
  24         'only_matching': True,
  25     }, {
  26         # audio
  27         'url': 'http://www.mdr.de/kultur/audio1312272_zc-15948bad_zs-86171fdd.html',
  28         'md5': '64c4ee50f0a791deb9479cd7bbe9d2fa',
  29         'info_dict': {
  30             'id': '1312272',
  31             'ext': 'mp3',
  32             'title': 'Feuilleton vom 30. Oktober 2015',
  33             'duration': 250,
  34             'uploader': 'MITTELDEUTSCHER RUNDFUNK',
  35         },
  36         'skip': '404 not found',
  37     }, {
  38         'url': 'http://www.kika.de/baumhaus/videos/video19636.html',
  39         'md5': '4930515e36b06c111213e80d1e4aad0e',
  40         'info_dict': {
  41             'id': '19636',
  42             'ext': 'mp4',
  43             'title': 'Baumhaus vom 30. Oktober 2015',
  44             'duration': 134,
  45             'uploader': 'KIKA',
  46         },
  47         'skip': '404 not found',
  48     }, {
  49         'url': 'http://www.kika.de/sendungen/einzelsendungen/weihnachtsprogramm/videos/video8182.html',
  50         'md5': '5fe9c4dd7d71e3b238f04b8fdd588357',
  51         'info_dict': {
  52             'id': '8182',
  53             'ext': 'mp4',
  54             'title': 'Beutolomäus und der geheime Weihnachtswunsch',
  55             'description': 'md5:b69d32d7b2c55cbe86945ab309d39bbd',
  56             'timestamp': 1482541200,
  57             'upload_date': '20161224',
  58             'duration': 4628,
  59             'uploader': 'KIKA',
  60         },
  61     }, {
  62         # audio with alternative playerURL pattern
  63         'url': 'http://www.mdr.de/kultur/videos-und-audios/audio-radio/operation-mindfuck-robert-wilson100.html',
  64         'info_dict': {
  65             'id': '100',
  66             'ext': 'mp4',
  67             'title': 'Feature: Operation Mindfuck - Robert Anton Wilson',
  68             'duration': 3239,
  69             'uploader': 'MITTELDEUTSCHER RUNDFUNK',
  70         },
  71     }, {
  72         # empty bitrateVideo and bitrateAudio
  73         'url': 'https://www.kika.de/filme/sendung128372_zc-572e3f45_zs-1d9fb70e.html',
  74         'info_dict': {
  75             'id': '128372',
  76             'ext': 'mp4',
  77             'title': 'Der kleine Wichtel kehrt zurück',
  78             'description': 'md5:f77fafdff90f7aa1e9dca14f662c052a',
  79             'duration': 4876,
  80             'timestamp': 1607823300,
  81             'upload_date': '20201213',
  82             'uploader': 'ZDF',
  83         },
  84         'params': {
  85             'skip_download': True,
  86         },
  87     }, {
  88         'url': 'http://www.kika.de/baumhaus/sendungen/video19636_zc-fea7f8a0_zs-4bf89c60.html',
  89         'only_matching': True,
  90     }, {
  91         'url': 'http://www.kika.de/sendungen/einzelsendungen/weihnachtsprogramm/einzelsendung2534.html',
  92         'only_matching': True,
  93     }, {
  94         'url': 'http://www.mdr.de/mediathek/mdr-videos/a/video-1334.html',
  95         'only_matching': True,
  96     }]
  97
  98     def _real_extract(self, url):
  99         video_id = self._match_id(url)
 100
 101         webpage = self._download_webpage(url, video_id)
 102
 103         data_url = self._search_regex(
 104             r'(?:dataURL|playerXml(?:["\'])?)\s*:\s*(["\'])(?P<url>.+?-avCustom\.xml)\1',
 105             webpage, 'data url', group='url').replace(r'\/', '/')
 106
 107         doc = self._download_xml(
 108             urllib.parse.urljoin(url, data_url), video_id)
 109
 110         title = xpath_text(doc, ['./title', './broadcast/broadcastName'], 'title', fatal=True)
 111
 112         type_ = xpath_text(doc, './type', default=None)
 113
 114         formats = []
 115         processed_urls = []
 116         for asset in doc.findall('./assets/asset'):
 117             for source in (
 118                     'download',
 119                     'progressiveDownload',
 120                     'dynamicHttpStreamingRedirector',
 121                     'adaptiveHttpStreamingRedirector'):
 122                 url_el = asset.find(f'./{source}Url')
 123                 if url_el is None:
 124                     continue
 125
 126                 video_url = url_or_none(url_el.text)
 127                 if not video_url or video_url in processed_urls:
 128                     continue
 129
 130                 processed_urls.append(video_url)
 131
 132                 ext = determine_ext(video_url)
 133                 if ext == 'm3u8':
 134                     formats.extend(self._extract_m3u8_formats(
 135                         video_url, video_id, 'mp4', entry_protocol='m3u8_native',
 136                         quality=1, m3u8_id='HLS', fatal=False))
 137                 elif ext == 'f4m':
 138                     formats.extend(self._extract_f4m_formats(
 139                         video_url + '?hdcore=3.7.0&plugin=aasp-3.7.0.39.44', video_id,
 140                         quality=1, f4m_id='HDS', fatal=False))
 141                 else:
 142                     media_type = xpath_text(asset, './mediaType', 'media type', default='MP4')
 143                     vbr = int_or_none(xpath_text(asset, './bitrateVideo', 'vbr'), 1000)
 144                     abr = int_or_none(xpath_text(asset, './bitrateAudio', 'abr'), 1000)
 145                     filesize = int_or_none(xpath_text(asset, './fileSize', 'file size'))
 146
 147                     f = {
 148                         'url': video_url,
 149                         'format_id': join_nonempty(media_type, vbr or abr),
 150                         'filesize': filesize,
 151                         'abr': abr,
 152                         'vbr': vbr,
 153                     }
 154
 155                     if vbr:
 156                         f.update({
 157                             'width': int_or_none(xpath_text(asset, './frameWidth', 'width')),
 158                             'height': int_or_none(xpath_text(asset, './frameHeight', 'height')),
 159                         })
 160
 161                     if type_ == 'audio':
 162                         f['vcodec'] = 'none'
 163
 164                     formats.append(f)
 165
 166         description = xpath_text(doc, './broadcast/broadcastDescription', 'description')
 167         timestamp = parse_iso8601(
 168             xpath_text(
 169                 doc, [
 170                     './broadcast/broadcastDate',
 171                     './broadcast/broadcastStartDate',
 172                     './broadcast/broadcastEndDate'],
 173                 'timestamp', default=None))
 174         duration = parse_duration(xpath_text(doc, './duration', 'duration'))
 175         uploader = xpath_text(doc, './rights', 'uploader')
 176
 177         return {
 178             'id': video_id,
 179             'title': title,
 180             'description': description,
 181             'timestamp': timestamp,
 182             'duration': duration,
 183             'uploader': uploader,
 184             'formats': formats,
 185         }