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'
  17     _VALID_URL = r'https?://(?:www\.)?mdr\.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         # audio with alternative playerURL pattern
  39         'url': 'http://www.mdr.de/kultur/videos-und-audios/audio-radio/operation-mindfuck-robert-wilson100.html',
  40         'info_dict': {
  41             'id': '100',
  42             'ext': 'mp4',
  43             'title': 'Feature: Operation Mindfuck - Robert Anton Wilson',
  44             'duration': 3239,
  45             'uploader': 'MITTELDEUTSCHER RUNDFUNK',
  46         },
  47         'skip': '404 not found',
  48     }, {
  49         'url': 'http://www.mdr.de/mediathek/mdr-videos/a/video-1334.html',
  50         'only_matching': True,
  51     }]
  52
  53     def _real_extract(self, url):
  54         video_id = self._match_id(url)
  55
  56         webpage = self._download_webpage(url, video_id)
  57
  58         data_url = self._search_regex(
  59             r'(?:dataURL|playerXml(?:["\'])?)\s*:\s*(["\'])(?P<url>.+?-avCustom\.xml)\1',
  60             webpage, 'data url', group='url').replace(r'\/', '/')
  61
  62         doc = self._download_xml(
  63             urllib.parse.urljoin(url, data_url), video_id)
  64
  65         title = xpath_text(doc, ['./title', './broadcast/broadcastName'], 'title', fatal=True)
  66
  67         type_ = xpath_text(doc, './type', default=None)
  68
  69         formats = []
  70         processed_urls = []
  71         for asset in doc.findall('./assets/asset'):
  72             for source in (
  73                     'download',
  74                     'progressiveDownload',
  75                     'dynamicHttpStreamingRedirector',
  76                     'adaptiveHttpStreamingRedirector'):
  77                 url_el = asset.find(f'./{source}Url')
  78                 if url_el is None:
  79                     continue
  80
  81                 video_url = url_or_none(url_el.text)
  82                 if not video_url or video_url in processed_urls:
  83                     continue
  84
  85                 processed_urls.append(video_url)
  86
  87                 ext = determine_ext(video_url)
  88                 if ext == 'm3u8':
  89                     formats.extend(self._extract_m3u8_formats(
  90                         video_url, video_id, 'mp4', entry_protocol='m3u8_native',
  91                         quality=1, m3u8_id='HLS', fatal=False))
  92                 elif ext == 'f4m':
  93                     formats.extend(self._extract_f4m_formats(
  94                         video_url + '?hdcore=3.7.0&plugin=aasp-3.7.0.39.44', video_id,
  95                         quality=1, f4m_id='HDS', fatal=False))
  96                 else:
  97                     media_type = xpath_text(asset, './mediaType', 'media type', default='MP4')
  98                     vbr = int_or_none(xpath_text(asset, './bitrateVideo', 'vbr'), 1000)
  99                     abr = int_or_none(xpath_text(asset, './bitrateAudio', 'abr'), 1000)
 100                     filesize = int_or_none(xpath_text(asset, './fileSize', 'file size'))
 101
 102                     f = {
 103                         'url': video_url,
 104                         'format_id': join_nonempty(media_type, vbr or abr),
 105                         'filesize': filesize,
 106                         'abr': abr,
 107                         'vbr': vbr,
 108                     }
 109
 110                     if vbr:
 111                         f.update({
 112                             'width': int_or_none(xpath_text(asset, './frameWidth', 'width')),
 113                             'height': int_or_none(xpath_text(asset, './frameHeight', 'height')),
 114                         })
 115
 116                     if type_ == 'audio':
 117                         f['vcodec'] = 'none'
 118
 119                     formats.append(f)
 120
 121         description = xpath_text(doc, './broadcast/broadcastDescription', 'description')
 122         timestamp = parse_iso8601(
 123             xpath_text(
 124                 doc, [
 125                     './broadcast/broadcastDate',
 126                     './broadcast/broadcastStartDate',
 127                     './broadcast/broadcastEndDate'],
 128                 'timestamp', default=None))
 129         duration = parse_duration(xpath_text(doc, './duration', 'duration'))
 130         uploader = xpath_text(doc, './rights', 'uploader')
 131
 132         return {
 133             'id': video_id,
 134             'title': title,
 135             'description': description,
 136             'timestamp': timestamp,
 137             'duration': duration,
 138             'uploader': uploader,
 139             'formats': formats,
 140         }