Raphaël G. Git Repositories - youtubedl/blob - 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 regularly 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': 1450950000,
  53             'upload_date': '20151224',
  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         'url': 'http://www.mdr.de/mediathek/mdr-videos/a/video-1334.html',
  65         'only_matching': True,
  66     }]
  67
  68     def _real_extract(self, url):
  69         video_id = self._match_id(url)
  70
  71         webpage = self._download_webpage(url, video_id)
  72
  73         data_url = self._search_regex(
  74             r'(?:dataURL|playerXml(?:["\'])?)\s*:\s*(["\'])(?P<url>.+/(?:video|audio)-?[0-9]+-avCustom\.xml)\1',
  75             webpage, 'data url', group='url').replace('\/', '/')
  76
  77         doc = self._download_xml(
  78             compat_urlparse.urljoin(url, data_url), video_id)
  79
  80         title = xpath_text(doc, ['./title', './broadcast/broadcastName'], 'title', fatal=True)
  81
  82         formats = []
  83         processed_urls = []
  84         for asset in doc.findall('./assets/asset'):
  85             for source in (
  86                     'progressiveDownload',
  87                     'dynamicHttpStreamingRedirector',
  88                     'adaptiveHttpStreamingRedirector'):
  89                 url_el = asset.find('./%sUrl' % source)
  90                 if url_el is None:
  91                     continue
  92
  93                 video_url = url_el.text
  94                 if video_url in processed_urls:
  95                     continue
  96
  97                 processed_urls.append(video_url)
  98
  99                 vbr = int_or_none(xpath_text(asset, './bitrateVideo', 'vbr'), 1000)
 100                 abr = int_or_none(xpath_text(asset, './bitrateAudio', 'abr'), 1000)
 101
 102                 ext = determine_ext(url_el.text)
 103                 if ext == 'm3u8':
 104                     url_formats = self._extract_m3u8_formats(
 105                         video_url, video_id, 'mp4', entry_protocol='m3u8_native',
 106                         preference=0, m3u8_id='HLS', fatal=False)
 107                 elif ext == 'f4m':
 108                     url_formats = self._extract_f4m_formats(
 109                         video_url + '?hdcore=3.7.0&plugin=aasp-3.7.0.39.44', video_id,
 110                         preference=0, f4m_id='HDS', fatal=False)
 111                 else:
 112                     media_type = xpath_text(asset, './mediaType', 'media type', default='MP4')
 113                     vbr = int_or_none(xpath_text(asset, './bitrateVideo', 'vbr'), 1000)
 114                     abr = int_or_none(xpath_text(asset, './bitrateAudio', 'abr'), 1000)
 115                     filesize = int_or_none(xpath_text(asset, './fileSize', 'file size'))
 116
 117                     f = {
 118                         'url': video_url,
 119                         'format_id': '%s-%d' % (media_type, vbr or abr),
 120                         'filesize': filesize,
 121                         'abr': abr,
 122                         'preference': 1,
 123                     }
 124
 125                     if vbr:
 126                         width = int_or_none(xpath_text(asset, './frameWidth', 'width'))
 127                         height = int_or_none(xpath_text(asset, './frameHeight', 'height'))
 128                         f.update({
 129                             'vbr': vbr,
 130                             'width': width,
 131                             'height': height,
 132                         })
 133
 134                     url_formats = [f]
 135
 136                 if not url_formats:
 137                     continue
 138
 139                 if not vbr:
 140                     for f in url_formats:
 141                         abr = f.get('tbr') or abr
 142                         if 'tbr' in f:
 143                             del f['tbr']
 144                         f.update({
 145                             'abr': abr,
 146                             'vcodec': 'none',
 147                         })
 148
 149                 formats.extend(url_formats)
 150
 151         self._sort_formats(formats)
 152
 153         description = xpath_text(doc, './broadcast/broadcastDescription', 'description')
 154         timestamp = parse_iso8601(
 155             xpath_text(
 156                 doc, [
 157                     './broadcast/broadcastDate',
 158                     './broadcast/broadcastStartDate',
 159                     './broadcast/broadcastEndDate'],
 160                 'timestamp', default=None))
 161         duration = parse_duration(xpath_text(doc, './duration', 'duration'))
 162         uploader = xpath_text(doc, './rights', 'uploader')
 163
 164         return {
 165             'id': video_id,
 166             'title': title,
 167             'description': description,
 168             'timestamp': timestamp,
 169             'duration': duration,
 170             'uploader': uploader,
 171             'formats': formats,
 172         }