yt_dlp/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     join_nonempty,
  10     parse_duration,
  11     parse_iso8601,
  12     url_or_none,
  13     xpath_text,
  14 )
  15
  16
  17 class MDRIE(InfoExtractor):
  18     IE_DESC = 'MDR.DE and KiKA'
  19     _VALID_URL = r'https?://(?:www\.)?(?:mdr|kika)\.de/(?:.*)/[a-z-]+-?(?P<id>\d+)(?:_.+?)?\.html'
  20
  21     _GEO_COUNTRIES = ['DE']
  22
  23     _TESTS = [{
  24         # MDR regularly deletes its videos
  25         'url': 'http://www.mdr.de/fakt/video189002.html',
  26         'only_matching': True,
  27     }, {
  28         # audio
  29         'url': 'http://www.mdr.de/kultur/audio1312272_zc-15948bad_zs-86171fdd.html',
  30         'md5': '64c4ee50f0a791deb9479cd7bbe9d2fa',
  31         'info_dict': {
  32             'id': '1312272',
  33             'ext': 'mp3',
  34             'title': 'Feuilleton vom 30. Oktober 2015',
  35             'duration': 250,
  36             'uploader': 'MITTELDEUTSCHER RUNDFUNK',
  37         },
  38         'skip': '404 not found',
  39     }, {
  40         'url': 'http://www.kika.de/baumhaus/videos/video19636.html',
  41         'md5': '4930515e36b06c111213e80d1e4aad0e',
  42         'info_dict': {
  43             'id': '19636',
  44             'ext': 'mp4',
  45             'title': 'Baumhaus vom 30. Oktober 2015',
  46             'duration': 134,
  47             'uploader': 'KIKA',
  48         },
  49         'skip': '404 not found',
  50     }, {
  51         'url': 'http://www.kika.de/sendungen/einzelsendungen/weihnachtsprogramm/videos/video8182.html',
  52         'md5': '5fe9c4dd7d71e3b238f04b8fdd588357',
  53         'info_dict': {
  54             'id': '8182',
  55             'ext': 'mp4',
  56             'title': 'Beutolomäus und der geheime Weihnachtswunsch',
  57             'description': 'md5:b69d32d7b2c55cbe86945ab309d39bbd',
  58             'timestamp': 1482541200,
  59             'upload_date': '20161224',
  60             'duration': 4628,
  61             'uploader': 'KIKA',
  62         },
  63     }, {
  64         # audio with alternative playerURL pattern
  65         'url': 'http://www.mdr.de/kultur/videos-und-audios/audio-radio/operation-mindfuck-robert-wilson100.html',
  66         'info_dict': {
  67             'id': '100',
  68             'ext': 'mp4',
  69             'title': 'Feature: Operation Mindfuck - Robert Anton Wilson',
  70             'duration': 3239,
  71             'uploader': 'MITTELDEUTSCHER RUNDFUNK',
  72         },
  73     }, {
  74         # empty bitrateVideo and bitrateAudio
  75         'url': 'https://www.kika.de/filme/sendung128372_zc-572e3f45_zs-1d9fb70e.html',
  76         'info_dict': {
  77             'id': '128372',
  78             'ext': 'mp4',
  79             'title': 'Der kleine Wichtel kehrt zurück',
  80             'description': 'md5:f77fafdff90f7aa1e9dca14f662c052a',
  81             'duration': 4876,
  82             'timestamp': 1607823300,
  83             'upload_date': '20201213',
  84             'uploader': 'ZDF',
  85         },
  86         'params': {
  87             'skip_download': True,
  88         },
  89     }, {
  90         'url': 'http://www.kika.de/baumhaus/sendungen/video19636_zc-fea7f8a0_zs-4bf89c60.html',
  91         'only_matching': True,
  92     }, {
  93         'url': 'http://www.kika.de/sendungen/einzelsendungen/weihnachtsprogramm/einzelsendung2534.html',
  94         'only_matching': True,
  95     }, {
  96         'url': 'http://www.mdr.de/mediathek/mdr-videos/a/video-1334.html',
  97         'only_matching': True,
  98     }]
  99
 100     def _real_extract(self, url):
 101         video_id = self._match_id(url)
 102
 103         webpage = self._download_webpage(url, video_id)
 104
 105         data_url = self._search_regex(
 106             r'(?:dataURL|playerXml(?:["\'])?)\s*:\s*(["\'])(?P<url>.+?-avCustom\.xml)\1',
 107             webpage, 'data url', group='url').replace(r'\/', '/')
 108
 109         doc = self._download_xml(
 110             compat_urlparse.urljoin(url, data_url), video_id)
 111
 112         title = xpath_text(doc, ['./title', './broadcast/broadcastName'], 'title', fatal=True)
 113
 114         type_ = xpath_text(doc, './type', default=None)
 115
 116         formats = []
 117         processed_urls = []
 118         for asset in doc.findall('./assets/asset'):
 119             for source in (
 120                     'download',
 121                     'progressiveDownload',
 122                     'dynamicHttpStreamingRedirector',
 123                     'adaptiveHttpStreamingRedirector'):
 124                 url_el = asset.find('./%sUrl' % source)
 125                 if url_el is None:
 126                     continue
 127
 128                 video_url = url_or_none(url_el.text)
 129                 if not video_url or video_url in processed_urls:
 130                     continue
 131
 132                 processed_urls.append(video_url)
 133
 134                 ext = determine_ext(video_url)
 135                 if ext == 'm3u8':
 136                     formats.extend(self._extract_m3u8_formats(
 137                         video_url, video_id, 'mp4', entry_protocol='m3u8_native',
 138                         quality=1, m3u8_id='HLS', fatal=False))
 139                 elif ext == 'f4m':
 140                     formats.extend(self._extract_f4m_formats(
 141                         video_url + '?hdcore=3.7.0&plugin=aasp-3.7.0.39.44', video_id,
 142                         quality=1, f4m_id='HDS', fatal=False))
 143                 else:
 144                     media_type = xpath_text(asset, './mediaType', 'media type', default='MP4')
 145                     vbr = int_or_none(xpath_text(asset, './bitrateVideo', 'vbr'), 1000)
 146                     abr = int_or_none(xpath_text(asset, './bitrateAudio', 'abr'), 1000)
 147                     filesize = int_or_none(xpath_text(asset, './fileSize', 'file size'))
 148
 149                     f = {
 150                         'url': video_url,
 151                         'format_id': join_nonempty(media_type, vbr or abr),
 152                         'filesize': filesize,
 153                         'abr': abr,
 154                         'vbr': vbr,
 155                     }
 156
 157                     if vbr:
 158                         f.update({
 159                             'width': int_or_none(xpath_text(asset, './frameWidth', 'width')),
 160                             'height': int_or_none(xpath_text(asset, './frameHeight', 'height')),
 161                         })
 162
 163                     if type_ == 'audio':
 164                         f['vcodec'] = 'none'
 165
 166                     formats.append(f)
 167
 168         self._sort_formats(formats)
 169
 170         description = xpath_text(doc, './broadcast/broadcastDescription', 'description')
 171         timestamp = parse_iso8601(
 172             xpath_text(
 173                 doc, [
 174                     './broadcast/broadcastDate',
 175                     './broadcast/broadcastStartDate',
 176                     './broadcast/broadcastEndDate'],
 177                 'timestamp', default=None))
 178         duration = parse_duration(xpath_text(doc, './duration', 'duration'))
 179         uploader = xpath_text(doc, './rights', 'uploader')
 180
 181         return {
 182             'id': video_id,
 183             'title': title,
 184             'description': description,
 185             'timestamp': timestamp,
 186             'duration': duration,
 187             'uploader': uploader,
 188             'formats': formats,
 189         }