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 regularily 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     }, {
  35         'url': 'http://www.kika.de/baumhaus/videos/video19636.html',
  36         'md5': '4930515e36b06c111213e80d1e4aad0e',
  37         'info_dict': {
  38             'id': '19636',
  39             'ext': 'mp4',
  40             'title': 'Baumhaus vom 30. Oktober 2015',
  41             'duration': 134,
  42             'uploader': 'KIKA',
  43         },
  44     }, {
  45         'url': 'http://www.kika.de/sendungen/einzelsendungen/weihnachtsprogramm/videos/video8182.html',
  46         'md5': '5fe9c4dd7d71e3b238f04b8fdd588357',
  47         'info_dict': {
  48             'id': '8182',
  49             'ext': 'mp4',
  50             'title': 'Beutolomäus und der geheime Weihnachtswunsch',
  51             'description': 'md5:b69d32d7b2c55cbe86945ab309d39bbd',
  52             'timestamp': 1419047100,
  53             'upload_date': '20141220',
  54             'duration': 4628,
  55             'uploader': 'KIKA',
  56         },
  57     }, {
  58         'url': 'http://www.kika.de/baumhaus/sendungen/video19636_zc-fea7f8a0_zs-4bf89c60.html',
  59         'only_matching': True,
  60     }, {
  61         'url': 'http://www.kika.de/sendungen/einzelsendungen/weihnachtsprogramm/einzelsendung2534.html',
  62         'only_matching': True,
  63     }]
  64
  65     def _real_extract(self, url):
  66         video_id = self._match_id(url)
  67
  68         webpage = self._download_webpage(url, video_id)
  69
  70         data_url = self._search_regex(
  71             r'dataURL\s*:\s*(["\'])(?P<url>/.+/(?:video|audio)[0-9]+-avCustom\.xml)\1',
  72             webpage, 'data url', group='url')
  73
  74         doc = self._download_xml(
  75             compat_urlparse.urljoin(url, data_url), video_id)
  76
  77         title = xpath_text(doc, ['./title', './broadcast/broadcastName'], 'title', fatal=True)
  78
  79         formats = []
  80         processed_urls = []
  81         for asset in doc.findall('./assets/asset'):
  82             for source in (
  83                     'progressiveDownload',
  84                     'dynamicHttpStreamingRedirector',
  85                     'adaptiveHttpStreamingRedirector'):
  86                 url_el = asset.find('./%sUrl' % source)
  87                 if url_el is None:
  88                     continue
  89
  90                 video_url = url_el.text
  91                 if video_url in processed_urls:
  92                     continue
  93
  94                 processed_urls.append(video_url)
  95
  96                 vbr = int_or_none(xpath_text(asset, './bitrateVideo', 'vbr'), 1000)
  97                 abr = int_or_none(xpath_text(asset, './bitrateAudio', 'abr'), 1000)
  98
  99                 url_formats = []
 100
 101                 ext = determine_ext(url_el.text)
 102                 if ext == 'm3u8':
 103                     url_formats = self._extract_m3u8_formats(
 104                         video_url, video_id, 'mp4', entry_protocol='m3u8_native',
 105                         preference=0, m3u8_id='HLS', fatal=False)
 106                 elif ext == 'f4m':
 107                     url_formats = self._extract_f4m_formats(
 108                         video_url + '?hdcore=3.7.0&plugin=aasp-3.7.0.39.44', video_id,
 109                         preference=0, f4m_id='HDS', fatal=False)
 110                 else:
 111                     media_type = xpath_text(asset, './mediaType', 'media type', default='MP4')
 112                     vbr = int_or_none(xpath_text(asset, './bitrateVideo', 'vbr'), 1000)
 113                     abr = int_or_none(xpath_text(asset, './bitrateAudio', 'abr'), 1000)
 114                     filesize = int_or_none(xpath_text(asset, './fileSize', 'file size'))
 115
 116                     f = {
 117                         'url': video_url,
 118                         'format_id': '%s-%d' % (media_type, vbr or abr),
 119                         'filesize': filesize,
 120                         'abr': abr,
 121                         'preference': 1,
 122                     }
 123
 124                     if vbr:
 125                         width = int_or_none(xpath_text(asset, './frameWidth', 'width'))
 126                         height = int_or_none(xpath_text(asset, './frameHeight', 'height'))
 127                         f.update({
 128                             'vbr': vbr,
 129                             'width': width,
 130                             'height': height,
 131                         })
 132
 133                     url_formats.append(f)
 134
 135                 if not vbr:
 136                     for f in url_formats:
 137                         abr = f.get('tbr') or abr
 138                         if 'tbr' in f:
 139                             del f['tbr']
 140                         f.update({
 141                             'abr': abr,
 142                             'vcodec': 'none',
 143                         })
 144
 145                 if url_formats:
 146                     formats.extend(url_formats)
 147         self._sort_formats(formats)
 148
 149         description = xpath_text(doc, './broadcast/broadcastDescription', 'description')
 150         timestamp = parse_iso8601(
 151             xpath_text(
 152                 doc, [
 153                     './broadcast/broadcastDate',
 154                     './broadcast/broadcastStartDate',
 155                     './broadcast/broadcastEndDate'],
 156                 'timestamp', default=None))
 157         duration = parse_duration(xpath_text(doc, './duration', 'duration'))
 158         uploader = xpath_text(doc, './rights', 'uploader')
 159
 160         return {
 161             'id': video_id,
 162             'title': title,
 163             'description': description,
 164             'timestamp': timestamp,
 165             'duration': duration,
 166             'uploader': uploader,
 167             'formats': formats,
 168         }