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 (
   6     compat_str,
   7     compat_urlparse,
   8 )
   9 from ..utils import (
  10     determine_ext,
  11     int_or_none,
  12     parse_duration,
  13     parse_iso8601,
  14     url_or_none,
  15     xpath_text,
  16 )
  17
  18
  19 class MDRIE(InfoExtractor):
  20     IE_DESC = 'MDR.DE and KiKA'
  21     _VALID_URL = r'https?://(?:www\.)?(?:mdr|kika)\.de/(?:.*)/[a-z-]+-?(?P<id>\d+)(?:_.+?)?\.html'
  22
  23     _GEO_COUNTRIES = ['DE']
  24
  25     _TESTS = [{
  26         # MDR regularly deletes its videos
  27         'url': 'http://www.mdr.de/fakt/video189002.html',
  28         'only_matching': True,
  29     }, {
  30         # audio
  31         'url': 'http://www.mdr.de/kultur/audio1312272_zc-15948bad_zs-86171fdd.html',
  32         'md5': '64c4ee50f0a791deb9479cd7bbe9d2fa',
  33         'info_dict': {
  34             'id': '1312272',
  35             'ext': 'mp3',
  36             'title': 'Feuilleton vom 30. Oktober 2015',
  37             'duration': 250,
  38             'uploader': 'MITTELDEUTSCHER RUNDFUNK',
  39         },
  40         'skip': '404 not found',
  41     }, {
  42         'url': 'http://www.kika.de/baumhaus/videos/video19636.html',
  43         'md5': '4930515e36b06c111213e80d1e4aad0e',
  44         'info_dict': {
  45             'id': '19636',
  46             'ext': 'mp4',
  47             'title': 'Baumhaus vom 30. Oktober 2015',
  48             'duration': 134,
  49             'uploader': 'KIKA',
  50         },
  51         'skip': '404 not found',
  52     }, {
  53         'url': 'http://www.kika.de/sendungen/einzelsendungen/weihnachtsprogramm/videos/video8182.html',
  54         'md5': '5fe9c4dd7d71e3b238f04b8fdd588357',
  55         'info_dict': {
  56             'id': '8182',
  57             'ext': 'mp4',
  58             'title': 'Beutolomäus und der geheime Weihnachtswunsch',
  59             'description': 'md5:b69d32d7b2c55cbe86945ab309d39bbd',
  60             'timestamp': 1482541200,
  61             'upload_date': '20161224',
  62             'duration': 4628,
  63             'uploader': 'KIKA',
  64         },
  65     }, {
  66         # audio with alternative playerURL pattern
  67         'url': 'http://www.mdr.de/kultur/videos-und-audios/audio-radio/operation-mindfuck-robert-wilson100.html',
  68         'info_dict': {
  69             'id': '100',
  70             'ext': 'mp4',
  71             'title': 'Feature: Operation Mindfuck - Robert Anton Wilson',
  72             'duration': 3239,
  73             'uploader': 'MITTELDEUTSCHER RUNDFUNK',
  74         },
  75     }, {
  76         # empty bitrateVideo and bitrateAudio
  77         'url': 'https://www.kika.de/filme/sendung128372_zc-572e3f45_zs-1d9fb70e.html',
  78         'info_dict': {
  79             'id': '128372',
  80             'ext': 'mp4',
  81             'title': 'Der kleine Wichtel kehrt zurück',
  82             'description': 'md5:f77fafdff90f7aa1e9dca14f662c052a',
  83             'duration': 4876,
  84             'timestamp': 1607823300,
  85             'upload_date': '20201213',
  86             'uploader': 'ZDF',
  87         },
  88         'params': {
  89             'skip_download': True,
  90         },
  91     }, {
  92         'url': 'http://www.kika.de/baumhaus/sendungen/video19636_zc-fea7f8a0_zs-4bf89c60.html',
  93         'only_matching': True,
  94     }, {
  95         'url': 'http://www.kika.de/sendungen/einzelsendungen/weihnachtsprogramm/einzelsendung2534.html',
  96         'only_matching': True,
  97     }, {
  98         'url': 'http://www.mdr.de/mediathek/mdr-videos/a/video-1334.html',
  99         'only_matching': True,
 100     }]
 101
 102     def _real_extract(self, url):
 103         video_id = self._match_id(url)
 104
 105         webpage = self._download_webpage(url, video_id)
 106
 107         data_url = self._search_regex(
 108             r'(?:dataURL|playerXml(?:["\'])?)\s*:\s*(["\'])(?P<url>.+?-avCustom\.xml)\1',
 109             webpage, 'data url', group='url').replace(r'\/', '/')
 110
 111         doc = self._download_xml(
 112             compat_urlparse.urljoin(url, data_url), video_id)
 113
 114         title = xpath_text(doc, ['./title', './broadcast/broadcastName'], 'title', fatal=True)
 115
 116         type_ = xpath_text(doc, './type', default=None)
 117
 118         formats = []
 119         processed_urls = []
 120         for asset in doc.findall('./assets/asset'):
 121             for source in (
 122                     'download',
 123                     'progressiveDownload',
 124                     'dynamicHttpStreamingRedirector',
 125                     'adaptiveHttpStreamingRedirector'):
 126                 url_el = asset.find('./%sUrl' % source)
 127                 if url_el is None:
 128                     continue
 129
 130                 video_url = url_or_none(url_el.text)
 131                 if not video_url or video_url in processed_urls:
 132                     continue
 133
 134                 processed_urls.append(video_url)
 135
 136                 ext = determine_ext(video_url)
 137                 if ext == 'm3u8':
 138                     formats.extend(self._extract_m3u8_formats(
 139                         video_url, video_id, 'mp4', entry_protocol='m3u8_native',
 140                         preference=0, m3u8_id='HLS', fatal=False))
 141                 elif ext == 'f4m':
 142                     formats.extend(self._extract_f4m_formats(
 143                         video_url + '?hdcore=3.7.0&plugin=aasp-3.7.0.39.44', video_id,
 144                         preference=0, f4m_id='HDS', fatal=False))
 145                 else:
 146                     media_type = xpath_text(asset, './mediaType', 'media type', default='MP4')
 147                     vbr = int_or_none(xpath_text(asset, './bitrateVideo', 'vbr'), 1000)
 148                     abr = int_or_none(xpath_text(asset, './bitrateAudio', 'abr'), 1000)
 149                     filesize = int_or_none(xpath_text(asset, './fileSize', 'file size'))
 150
 151                     format_id = [media_type]
 152                     if vbr or abr:
 153                         format_id.append(compat_str(vbr or abr))
 154
 155                     f = {
 156                         'url': video_url,
 157                         'format_id': '-'.join(format_id),
 158                         'filesize': filesize,
 159                         'abr': abr,
 160                         'vbr': vbr,
 161                     }
 162
 163                     if vbr:
 164                         f.update({
 165                             'width': int_or_none(xpath_text(asset, './frameWidth', 'width')),
 166                             'height': int_or_none(xpath_text(asset, './frameHeight', 'height')),
 167                         })
 168
 169                     if type_ == 'audio':
 170                         f['vcodec'] = 'none'
 171
 172                     formats.append(f)
 173
 174         self._sort_formats(formats)
 175
 176         description = xpath_text(doc, './broadcast/broadcastDescription', 'description')
 177         timestamp = parse_iso8601(
 178             xpath_text(
 179                 doc, [
 180                     './broadcast/broadcastDate',
 181                     './broadcast/broadcastStartDate',
 182                     './broadcast/broadcastEndDate'],
 183                 'timestamp', default=None))
 184         duration = parse_duration(xpath_text(doc, './duration', 'duration'))
 185         uploader = xpath_text(doc, './rights', 'uploader')
 186
 187         return {
 188             'id': video_id,
 189             'title': title,
 190             'description': description,
 191             'timestamp': timestamp,
 192             'duration': duration,
 193             'uploader': uploader,
 194             'formats': formats,
 195         }