Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/ard.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from .generic import GenericIE
   8 from ..utils import (
   9     determine_ext,
  10     ExtractorError,
  11     qualities,
  12     int_or_none,
  13     parse_duration,
  14     unified_strdate,
  15     xpath_text,
  16     parse_xml,
  17 )
  18
  19
  20 class ARDMediathekIE(InfoExtractor):
  21     IE_NAME = 'ARD:mediathek'
  22     _VALID_URL = r'^https?://(?:(?:www\.)?ardmediathek\.de|mediathek\.daserste\.de)/(?:.*/)(?P<video_id>[0-9]+|[^0-9][^/\?]+)[^/\?]*(?:\?.*)?'
  23
  24     _TESTS = [{
  25         'url': 'http://mediathek.daserste.de/sendungen_a-z/328454_anne-will/22429276_vertrauen-ist-gut-spionieren-ist-besser-geht',
  26         'only_matching': True,
  27     }, {
  28         'url': 'http://www.ardmediathek.de/tv/Tatort/Das-Wunder-von-Wolbeck-Video-tgl-ab-20/Das-Erste/Video?documentId=22490580&bcastId=602916',
  29         'info_dict': {
  30             'id': '22490580',
  31             'ext': 'mp4',
  32             'title': 'Das Wunder von Wolbeck (Video tgl. ab 20 Uhr)',
  33             'description': 'Auf einem restaurierten Hof bei Wolbeck wird der Heilpraktiker Raffael Lembeck eines morgens von seiner Frau Stella tot aufgefunden. Das Opfer war offensichtlich in seiner Praxis zu Fall gekommen und ist dann verblutet, erklärt Prof. Boerne am Tatort.',
  34         },
  35         'skip': 'Blocked outside of Germany',
  36     }]
  37
  38     def _real_extract(self, url):
  39         # determine video id from url
  40         m = re.match(self._VALID_URL, url)
  41
  42         numid = re.search(r'documentId=([0-9]+)', url)
  43         if numid:
  44             video_id = numid.group(1)
  45         else:
  46             video_id = m.group('video_id')
  47
  48         webpage = self._download_webpage(url, video_id)
  49
  50         if '>Der gewünschte Beitrag ist nicht mehr verfügbar.<' in webpage:
  51             raise ExtractorError('Video %s is no longer available' % video_id, expected=True)
  52
  53         if 'Diese Sendung ist für Jugendliche unter 12 Jahren nicht geeignet. Der Clip ist deshalb nur von 20 bis 6 Uhr verfügbar.' in webpage:
  54             raise ExtractorError('This program is only suitable for those aged 12 and older. Video %s is therefore only available between 20 pm and 6 am.' % video_id, expected=True)
  55
  56         if re.search(r'[\?&]rss($|[=&])', url):
  57             doc = parse_xml(webpage)
  58             if doc.tag == 'rss':
  59                 return GenericIE()._extract_rss(url, video_id, doc)
  60
  61         title = self._html_search_regex(
  62             [r'<h1(?:\s+class="boxTopHeadline")?>(.*?)</h1>',
  63              r'<meta name="dcterms.title" content="(.*?)"/>',
  64              r'<h4 class="headline">(.*?)</h4>'],
  65             webpage, 'title')
  66         description = self._html_search_meta(
  67             'dcterms.abstract', webpage, 'description', default=None)
  68         if description is None:
  69             description = self._html_search_meta(
  70                 'description', webpage, 'meta description')
  71
  72         # Thumbnail is sometimes not present.
  73         # It is in the mobile version, but that seems to use a different URL
  74         # structure altogether.
  75         thumbnail = self._og_search_thumbnail(webpage, default=None)
  76
  77         media_streams = re.findall(r'''(?x)
  78             mediaCollection\.addMediaStream\([0-9]+,\s*[0-9]+,\s*"[^"]*",\s*
  79             "([^"]+)"''', webpage)
  80
  81         if media_streams:
  82             QUALITIES = qualities(['lo', 'hi', 'hq'])
  83             formats = []
  84             for furl in set(media_streams):
  85                 if furl.endswith('.f4m'):
  86                     fid = 'f4m'
  87                 else:
  88                     fid_m = re.match(r'.*\.([^.]+)\.[^.]+$', furl)
  89                     fid = fid_m.group(1) if fid_m else None
  90                 formats.append({
  91                     'quality': QUALITIES(fid),
  92                     'format_id': fid,
  93                     'url': furl,
  94                 })
  95         else:  # request JSON file
  96             media_info = self._download_json(
  97                 'http://www.ardmediathek.de/play/media/%s' % video_id, video_id)
  98             # The second element of the _mediaArray contains the standard http urls
  99             streams = media_info['_mediaArray'][1]['_mediaStreamArray']
 100             if not streams:
 101                 if '"fsk"' in webpage:
 102                     raise ExtractorError('This video is only available after 20:00')
 103
 104             formats = []
 105             for s in streams:
 106                 if type(s['_stream']) == list:
 107                     for index, url in enumerate(s['_stream'][::-1]):
 108                         quality = s['_quality'] + index
 109                         formats.append({
 110                             'quality': quality,
 111                             'url': url,
 112                             'format_id': '%s-%s' % (determine_ext(url), quality)
 113                         })
 114                     continue
 115
 116                 format = {
 117                     'quality': s['_quality'],
 118                     'url': s['_stream'],
 119                 }
 120
 121                 format['format_id'] = '%s-%s' % (
 122                     determine_ext(format['url']), format['quality'])
 123
 124                 formats.append(format)
 125
 126         self._sort_formats(formats)
 127
 128         return {
 129             'id': video_id,
 130             'title': title,
 131             'description': description,
 132             'formats': formats,
 133             'thumbnail': thumbnail,
 134         }
 135
 136
 137 class ARDIE(InfoExtractor):
 138     _VALID_URL = '(?P<mainurl>https?://(www\.)?daserste\.de/[^?#]+/videos/(?P<display_id>[^/?#]+)-(?P<id>[0-9]+))\.html'
 139     _TEST = {
 140         'url': 'http://www.daserste.de/information/reportage-dokumentation/dokus/videos/die-story-im-ersten-mission-unter-falscher-flagge-100.html',
 141         'md5': 'd216c3a86493f9322545e045ddc3eb35',
 142         'info_dict': {
 143             'display_id': 'die-story-im-ersten-mission-unter-falscher-flagge',
 144             'id': '100',
 145             'ext': 'mp4',
 146             'duration': 2600,
 147             'title': 'Die Story im Ersten: Mission unter falscher Flagge',
 148             'upload_date': '20140804',
 149             'thumbnail': 're:^https?://.*\.jpg$',
 150         }
 151     }
 152
 153     def _real_extract(self, url):
 154         mobj = re.match(self._VALID_URL, url)
 155         display_id = mobj.group('display_id')
 156
 157         player_url = mobj.group('mainurl') + '~playerXml.xml'
 158         doc = self._download_xml(player_url, display_id)
 159         video_node = doc.find('./video')
 160         upload_date = unified_strdate(xpath_text(
 161             video_node, './broadcastDate'))
 162         thumbnail = xpath_text(video_node, './/teaserImage//variant/url')
 163
 164         formats = []
 165         for a in video_node.findall('.//asset'):
 166             f = {
 167                 'format_id': a.attrib['type'],
 168                 'width': int_or_none(a.find('./frameWidth').text),
 169                 'height': int_or_none(a.find('./frameHeight').text),
 170                 'vbr': int_or_none(a.find('./bitrateVideo').text),
 171                 'abr': int_or_none(a.find('./bitrateAudio').text),
 172                 'vcodec': a.find('./codecVideo').text,
 173                 'tbr': int_or_none(a.find('./totalBitrate').text),
 174             }
 175             if a.find('./serverPrefix').text:
 176                 f['url'] = a.find('./serverPrefix').text
 177                 f['playpath'] = a.find('./fileName').text
 178             else:
 179                 f['url'] = a.find('./fileName').text
 180             formats.append(f)
 181         self._sort_formats(formats)
 182
 183         return {
 184             'id': mobj.group('id'),
 185             'formats': formats,
 186             'display_id': display_id,
 187             'title': video_node.find('./title').text,
 188             'duration': parse_duration(video_node.find('./duration').text),
 189             'upload_date': upload_date,
 190             'thumbnail': thumbnail,
 191         }