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     update_url_query,
  17 )
  18 from ..compat import compat_etree_fromstring
  19
  20
  21 class ARDMediathekIE(InfoExtractor):
  22     IE_NAME = 'ARD:mediathek'
  23     _VALID_URL = r'^https?://(?:(?:www\.)?ardmediathek\.de|mediathek\.(?:daserste|rbb-online)\.de)/(?:.*/)(?P<video_id>[0-9]+|[^0-9][^/\?]+)[^/\?]*(?:\?.*)?'
  24
  25     _TESTS = [{
  26         'url': 'http://www.ardmediathek.de/tv/Dokumentation-und-Reportage/Ich-liebe-das-Leben-trotzdem/rbb-Fernsehen/Video?documentId=29582122&bcastId=3822114',
  27         'info_dict': {
  28             'id': '29582122',
  29             'ext': 'mp4',
  30             'title': 'Ich liebe das Leben trotzdem',
  31             'description': 'md5:45e4c225c72b27993314b31a84a5261c',
  32             'duration': 4557,
  33         },
  34         'params': {
  35             # m3u8 download
  36             'skip_download': True,
  37         },
  38         'skip': 'HTTP Error 404: Not Found',
  39     }, {
  40         'url': 'http://www.ardmediathek.de/tv/Tatort/Tatort-Scheinwelten-H%C3%B6rfassung-Video/Das-Erste/Video?documentId=29522730&bcastId=602916',
  41         'md5': 'f4d98b10759ac06c0072bbcd1f0b9e3e',
  42         'info_dict': {
  43             'id': '29522730',
  44             'ext': 'mp4',
  45             'title': 'Tatort: Scheinwelten - Hörfassung (Video tgl. ab 20 Uhr)',
  46             'description': 'md5:196392e79876d0ac94c94e8cdb2875f1',
  47             'duration': 5252,
  48         },
  49         'skip': 'HTTP Error 404: Not Found',
  50     }, {
  51         # audio
  52         'url': 'http://www.ardmediathek.de/tv/WDR-H%C3%B6rspiel-Speicher/Tod-eines-Fu%C3%9Fballers/WDR-3/Audio-Podcast?documentId=28488308&bcastId=23074086',
  53         'md5': '219d94d8980b4f538c7fcb0865eb7f2c',
  54         'info_dict': {
  55             'id': '28488308',
  56             'ext': 'mp3',
  57             'title': 'Tod eines Fußballers',
  58             'description': 'md5:f6e39f3461f0e1f54bfa48c8875c86ef',
  59             'duration': 3240,
  60         },
  61         'skip': 'HTTP Error 404: Not Found',
  62     }, {
  63         'url': 'http://mediathek.daserste.de/sendungen_a-z/328454_anne-will/22429276_vertrauen-ist-gut-spionieren-ist-besser-geht',
  64         'only_matching': True,
  65     }, {
  66         # audio
  67         'url': 'http://mediathek.rbb-online.de/radio/Hörspiel/Vor-dem-Fest/kulturradio/Audio?documentId=30796318&topRessort=radio&bcastId=9839158',
  68         'md5': '4e8f00631aac0395fee17368ac0e9867',
  69         'info_dict': {
  70             'id': '30796318',
  71             'ext': 'mp3',
  72             'title': 'Vor dem Fest',
  73             'description': 'md5:c0c1c8048514deaed2a73b3a60eecacb',
  74             'duration': 3287,
  75         },
  76         'skip': 'Video is no longer available',
  77     }]
  78
  79     def _extract_media_info(self, media_info_url, webpage, video_id):
  80         media_info = self._download_json(
  81             media_info_url, video_id, 'Downloading media JSON')
  82
  83         formats = self._extract_formats(media_info, video_id)
  84
  85         if not formats:
  86             if '"fsk"' in webpage:
  87                 raise ExtractorError(
  88                     'This video is only available after 20:00', expected=True)
  89             elif media_info.get('_geoblocked'):
  90                 raise ExtractorError('This video is not available due to geo restriction', expected=True)
  91
  92         self._sort_formats(formats)
  93
  94         duration = int_or_none(media_info.get('_duration'))
  95         thumbnail = media_info.get('_previewImage')
  96
  97         subtitles = {}
  98         subtitle_url = media_info.get('_subtitleUrl')
  99         if subtitle_url:
 100             subtitles['de'] = [{
 101                 'ext': 'ttml',
 102                 'url': subtitle_url,
 103             }]
 104
 105         return {
 106             'id': video_id,
 107             'duration': duration,
 108             'thumbnail': thumbnail,
 109             'formats': formats,
 110             'subtitles': subtitles,
 111         }
 112
 113     def _extract_formats(self, media_info, video_id):
 114         type_ = media_info.get('_type')
 115         media_array = media_info.get('_mediaArray', [])
 116         formats = []
 117         for num, media in enumerate(media_array):
 118             for stream in media.get('_mediaStreamArray', []):
 119                 stream_urls = stream.get('_stream')
 120                 if not stream_urls:
 121                     continue
 122                 if not isinstance(stream_urls, list):
 123                     stream_urls = [stream_urls]
 124                 quality = stream.get('_quality')
 125                 server = stream.get('_server')
 126                 for stream_url in stream_urls:
 127                     ext = determine_ext(stream_url)
 128                     if quality != 'auto' and ext in ('f4m', 'm3u8'):
 129                         continue
 130                     if ext == 'f4m':
 131                         formats.extend(self._extract_f4m_formats(
 132                             update_url_query(stream_url, {
 133                                 'hdcore': '3.1.1',
 134                                 'plugin': 'aasp-3.1.1.69.124'
 135                             }),
 136                             video_id, f4m_id='hds', fatal=False))
 137                     elif ext == 'm3u8':
 138                         formats.extend(self._extract_m3u8_formats(
 139                             stream_url, video_id, 'mp4', m3u8_id='hls', fatal=False))
 140                     else:
 141                         if server and server.startswith('rtmp'):
 142                             f = {
 143                                 'url': server,
 144                                 'play_path': stream_url,
 145                                 'format_id': 'a%s-rtmp-%s' % (num, quality),
 146                             }
 147                         elif stream_url.startswith('http'):
 148                             f = {
 149                                 'url': stream_url,
 150                                 'format_id': 'a%s-%s-%s' % (num, ext, quality)
 151                             }
 152                         else:
 153                             continue
 154                         m = re.search(r'_(?P<width>\d+)x(?P<height>\d+)\.mp4$', stream_url)
 155                         if m:
 156                             f.update({
 157                                 'width': int(m.group('width')),
 158                                 'height': int(m.group('height')),
 159                             })
 160                         if type_ == 'audio':
 161                             f['vcodec'] = 'none'
 162                         formats.append(f)
 163         return formats
 164
 165     def _real_extract(self, url):
 166         # determine video id from url
 167         m = re.match(self._VALID_URL, url)
 168
 169         numid = re.search(r'documentId=([0-9]+)', url)
 170         if numid:
 171             video_id = numid.group(1)
 172         else:
 173             video_id = m.group('video_id')
 174
 175         webpage = self._download_webpage(url, video_id)
 176
 177         if '>Der gewünschte Beitrag ist nicht mehr verfügbar.<' in webpage:
 178             raise ExtractorError('Video %s is no longer available' % video_id, expected=True)
 179
 180         if 'Diese Sendung ist für Jugendliche unter 12 Jahren nicht geeignet. Der Clip ist deshalb nur von 20 bis 6 Uhr verfügbar.' in webpage:
 181             raise ExtractorError('This program is only suitable for those aged 12 and older. Video %s is therefore only available between 20 pm and 6 am.' % video_id, expected=True)
 182
 183         if re.search(r'[\?&]rss($|[=&])', url):
 184             doc = compat_etree_fromstring(webpage.encode('utf-8'))
 185             if doc.tag == 'rss':
 186                 return GenericIE()._extract_rss(url, video_id, doc)
 187
 188         title = self._html_search_regex(
 189             [r'<h1(?:\s+class="boxTopHeadline")?>(.*?)</h1>',
 190              r'<meta name="dcterms.title" content="(.*?)"/>',
 191              r'<h4 class="headline">(.*?)</h4>'],
 192             webpage, 'title')
 193         description = self._html_search_meta(
 194             'dcterms.abstract', webpage, 'description', default=None)
 195         if description is None:
 196             description = self._html_search_meta(
 197                 'description', webpage, 'meta description')
 198
 199         # Thumbnail is sometimes not present.
 200         # It is in the mobile version, but that seems to use a different URL
 201         # structure altogether.
 202         thumbnail = self._og_search_thumbnail(webpage, default=None)
 203
 204         media_streams = re.findall(r'''(?x)
 205             mediaCollection\.addMediaStream\([0-9]+,\s*[0-9]+,\s*"[^"]*",\s*
 206             "([^"]+)"''', webpage)
 207
 208         if media_streams:
 209             QUALITIES = qualities(['lo', 'hi', 'hq'])
 210             formats = []
 211             for furl in set(media_streams):
 212                 if furl.endswith('.f4m'):
 213                     fid = 'f4m'
 214                 else:
 215                     fid_m = re.match(r'.*\.([^.]+)\.[^.]+$', furl)
 216                     fid = fid_m.group(1) if fid_m else None
 217                 formats.append({
 218                     'quality': QUALITIES(fid),
 219                     'format_id': fid,
 220                     'url': furl,
 221                 })
 222             self._sort_formats(formats)
 223             info = {
 224                 'formats': formats,
 225             }
 226         else:  # request JSON file
 227             info = self._extract_media_info(
 228                 'http://www.ardmediathek.de/play/media/%s' % video_id, webpage, video_id)
 229
 230         info.update({
 231             'id': video_id,
 232             'title': title,
 233             'description': description,
 234             'thumbnail': thumbnail,
 235         })
 236
 237         return info
 238
 239
 240 class ARDIE(InfoExtractor):
 241     _VALID_URL = '(?P<mainurl>https?://(www\.)?daserste\.de/[^?#]+/videos/(?P<display_id>[^/?#]+)-(?P<id>[0-9]+))\.html'
 242     _TEST = {
 243         'url': 'http://www.daserste.de/information/reportage-dokumentation/dokus/videos/die-story-im-ersten-mission-unter-falscher-flagge-100.html',
 244         'md5': 'd216c3a86493f9322545e045ddc3eb35',
 245         'info_dict': {
 246             'display_id': 'die-story-im-ersten-mission-unter-falscher-flagge',
 247             'id': '100',
 248             'ext': 'mp4',
 249             'duration': 2600,
 250             'title': 'Die Story im Ersten: Mission unter falscher Flagge',
 251             'upload_date': '20140804',
 252             'thumbnail': 're:^https?://.*\.jpg$',
 253         },
 254         'skip': 'HTTP Error 404: Not Found',
 255     }
 256
 257     def _real_extract(self, url):
 258         mobj = re.match(self._VALID_URL, url)
 259         display_id = mobj.group('display_id')
 260
 261         player_url = mobj.group('mainurl') + '~playerXml.xml'
 262         doc = self._download_xml(player_url, display_id)
 263         video_node = doc.find('./video')
 264         upload_date = unified_strdate(xpath_text(
 265             video_node, './broadcastDate'))
 266         thumbnail = xpath_text(video_node, './/teaserImage//variant/url')
 267
 268         formats = []
 269         for a in video_node.findall('.//asset'):
 270             f = {
 271                 'format_id': a.attrib['type'],
 272                 'width': int_or_none(a.find('./frameWidth').text),
 273                 'height': int_or_none(a.find('./frameHeight').text),
 274                 'vbr': int_or_none(a.find('./bitrateVideo').text),
 275                 'abr': int_or_none(a.find('./bitrateAudio').text),
 276                 'vcodec': a.find('./codecVideo').text,
 277                 'tbr': int_or_none(a.find('./totalBitrate').text),
 278             }
 279             if a.find('./serverPrefix').text:
 280                 f['url'] = a.find('./serverPrefix').text
 281                 f['playpath'] = a.find('./fileName').text
 282             else:
 283                 f['url'] = a.find('./fileName').text
 284             formats.append(f)
 285         self._sort_formats(formats)
 286
 287         return {
 288             'id': mobj.group('id'),
 289             'formats': formats,
 290             'display_id': display_id,
 291             'title': video_node.find('./title').text,
 292             'duration': parse_duration(video_node.find('./duration').text),
 293             'upload_date': upload_date,
 294             'thumbnail': thumbnail,
 295         }