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

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..utils import (
   8     determine_ext,
   9     ExtractorError,
  10     js_to_json,
  11     strip_jsonp,
  12     unified_strdate,
  13     update_url_query,
  14     urlhandle_detect_ext,
  15 )
  16
  17
  18 class WDRBaseIE(InfoExtractor):
  19     def _extract_wdr_video(self, webpage, display_id):
  20         # for wdr.de the data-extension is in a tag with the class "mediaLink"
  21         # for wdr.de radio players, in a tag with the class "wdrrPlayerPlayBtn"
  22         # for wdrmaus its in a link to the page in a multiline "videoLink"-tag
  23         json_metadata = self._html_search_regex(
  24             r'class=(?:"(?:mediaLink|wdrrPlayerPlayBtn)\b[^"]*"[^>]+|"videoLink\b[^"]*"[\s]*>\n[^\n]*)data-extension="([^"]+)"',
  25             webpage, 'media link', default=None, flags=re.MULTILINE)
  26
  27         if not json_metadata:
  28             return
  29
  30         media_link_obj = self._parse_json(json_metadata, display_id,
  31                                           transform_source=js_to_json)
  32         jsonp_url = media_link_obj['mediaObj']['url']
  33
  34         metadata = self._download_json(
  35             jsonp_url, 'metadata', transform_source=strip_jsonp)
  36
  37         metadata_tracker_data = metadata['trackerData']
  38         metadata_media_resource = metadata['mediaResource']
  39
  40         formats = []
  41
  42         # check if the metadata contains a direct URL to a file
  43         for kind, media_resource in metadata_media_resource.items():
  44             if kind not in ('dflt', 'alt'):
  45                 continue
  46
  47             for tag_name, medium_url in media_resource.items():
  48                 if tag_name not in ('videoURL', 'audioURL'):
  49                     continue
  50
  51                 ext = determine_ext(medium_url)
  52                 if ext == 'm3u8':
  53                     formats.extend(self._extract_m3u8_formats(
  54                         medium_url, display_id, 'mp4', 'm3u8_native',
  55                         m3u8_id='hls'))
  56                 elif ext == 'f4m':
  57                     manifest_url = update_url_query(
  58                         medium_url, {'hdcore': '3.2.0', 'plugin': 'aasp-3.2.0.77.18'})
  59                     formats.extend(self._extract_f4m_formats(
  60                         manifest_url, display_id, f4m_id='hds', fatal=False))
  61                 elif ext == 'smil':
  62                     formats.extend(self._extract_smil_formats(
  63                         medium_url, 'stream', fatal=False))
  64                 else:
  65                     a_format = {
  66                         'url': medium_url
  67                     }
  68                     if ext == 'unknown_video':
  69                         urlh = self._request_webpage(
  70                             medium_url, display_id, note='Determining extension')
  71                         ext = urlhandle_detect_ext(urlh)
  72                         a_format['ext'] = ext
  73                     formats.append(a_format)
  74
  75         self._sort_formats(formats)
  76
  77         subtitles = {}
  78         caption_url = metadata_media_resource.get('captionURL')
  79         if caption_url:
  80             subtitles['de'] = [{
  81                 'url': caption_url,
  82                 'ext': 'ttml',
  83             }]
  84
  85         title = metadata_tracker_data['trackerClipTitle']
  86
  87         return {
  88             'id': metadata_tracker_data.get('trackerClipId', display_id),
  89             'display_id': display_id,
  90             'title': title,
  91             'alt_title': metadata_tracker_data.get('trackerClipSubcategory'),
  92             'formats': formats,
  93             'subtitles': subtitles,
  94             'upload_date': unified_strdate(metadata_tracker_data.get('trackerClipAirTime')),
  95         }
  96
  97
  98 class WDRIE(WDRBaseIE):
  99     _CURRENT_MAUS_URL = r'https?://(?:www\.)wdrmaus.de/(?:[^/]+/){1,2}[^/?#]+\.php5'
 100     _PAGE_REGEX = r'/(?:mediathek/)?[^/]+/(?P<type>[^/]+)/(?P<display_id>.+)\.html'
 101     _VALID_URL = r'(?P<page_url>https?://(?:www\d\.)?wdr\d?\.de)' + _PAGE_REGEX + '|' + _CURRENT_MAUS_URL
 102
 103     _TESTS = [
 104         {
 105             'url': 'http://www1.wdr.de/mediathek/video/sendungen/doku-am-freitag/video-geheimnis-aachener-dom-100.html',
 106             # HDS download, MD5 is unstable
 107             'info_dict': {
 108                 'id': 'mdb-1058683',
 109                 'ext': 'flv',
 110                 'display_id': 'doku-am-freitag/video-geheimnis-aachener-dom-100',
 111                 'title': 'Geheimnis Aachener Dom',
 112                 'alt_title': 'Doku am Freitag',
 113                 'upload_date': '20160304',
 114                 'description': 'md5:87be8ff14d8dfd7a7ee46f0299b52318',
 115                 'is_live': False,
 116                 'subtitles': {'de': [{
 117                     'url': 'http://ondemand-ww.wdr.de/medp/fsk0/105/1058683/1058683_12220974.xml',
 118                     'ext': 'ttml',
 119                 }]},
 120             },
 121         },
 122         {
 123             'url': 'http://www1.wdr.de/mediathek/audio/wdr3/wdr3-gespraech-am-samstag/audio-schriftstellerin-juli-zeh-100.html',
 124             'md5': 'f4c1f96d01cf285240f53ea4309663d8',
 125             'info_dict': {
 126                 'id': 'mdb-1072000',
 127                 'ext': 'mp3',
 128                 'display_id': 'wdr3-gespraech-am-samstag/audio-schriftstellerin-juli-zeh-100',
 129                 'title': 'Schriftstellerin Juli Zeh',
 130                 'alt_title': 'WDR 3 Gespräch am Samstag',
 131                 'upload_date': '20160312',
 132                 'description': 'md5:e127d320bc2b1f149be697ce044a3dd7',
 133                 'is_live': False,
 134                 'subtitles': {}
 135             },
 136         },
 137         {
 138             'url': 'http://www1.wdr.de/mediathek/video/live/index.html',
 139             'info_dict': {
 140                 'id': 'mdb-103364',
 141                 'ext': 'mp4',
 142                 'display_id': 'index',
 143                 'title': r're:^WDR Fernsehen im Livestream [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
 144                 'alt_title': 'WDR Fernsehen Live',
 145                 'upload_date': None,
 146                 'description': 'md5:ae2ff888510623bf8d4b115f95a9b7c9',
 147                 'is_live': True,
 148                 'subtitles': {}
 149             },
 150             'params': {
 151                 'skip_download': True,  # m3u8 download
 152             },
 153         },
 154         {
 155             'url': 'http://www1.wdr.de/mediathek/video/sendungen/aktuelle-stunde/aktuelle-stunde-120.html',
 156             'playlist_mincount': 8,
 157             'info_dict': {
 158                 'id': 'aktuelle-stunde/aktuelle-stunde-120',
 159             },
 160         },
 161         {
 162             'url': 'http://www.wdrmaus.de/aktuelle-sendung/index.php5',
 163             'info_dict': {
 164                 'id': 'mdb-1096487',
 165                 'ext': 'flv',
 166                 'upload_date': 're:^[0-9]{8}$',
 167                 'title': 're:^Die Sendung mit der Maus vom [0-9.]{10}$',
 168                 'description': '- Die Sendung mit der Maus -',
 169             },
 170             'skip': 'The id changes from week to week because of the new episode'
 171         },
 172         {
 173             'url': 'http://www.wdrmaus.de/sachgeschichten/sachgeschichten/achterbahn.php5',
 174             'md5': '803138901f6368ee497b4d195bb164f2',
 175             'info_dict': {
 176                 'id': 'mdb-186083',
 177                 'ext': 'mp4',
 178                 'upload_date': '20130919',
 179                 'title': 'Sachgeschichte - Achterbahn ',
 180                 'description': '- Die Sendung mit der Maus -',
 181             },
 182         },
 183         {
 184             'url': 'http://www1.wdr.de/radio/player/radioplayer116~_layout-popupVersion.html',
 185             # Live stream, MD5 unstable
 186             'info_dict': {
 187                 'id': 'mdb-869971',
 188                 'ext': 'flv',
 189                 'title': 'Funkhaus Europa Livestream',
 190                 'description': 'md5:2309992a6716c347891c045be50992e4',
 191                 'upload_date': '20160101',
 192             },
 193         }
 194     ]
 195
 196     def _real_extract(self, url):
 197         mobj = re.match(self._VALID_URL, url)
 198         url_type = mobj.group('type')
 199         page_url = mobj.group('page_url')
 200         display_id = mobj.group('display_id')
 201         webpage = self._download_webpage(url, display_id)
 202
 203         info_dict = self._extract_wdr_video(webpage, display_id)
 204
 205         if not info_dict:
 206             entries = [
 207                 self.url_result(page_url + href[0], 'WDR')
 208                 for href in re.findall(
 209                     r'<a href="(%s)"[^>]+data-extension=' % self._PAGE_REGEX,
 210                     webpage)
 211             ]
 212
 213             if entries:  # Playlist page
 214                 return self.playlist_result(entries, playlist_id=display_id)
 215
 216             raise ExtractorError('No downloadable streams found', expected=True)
 217
 218         is_live = url_type == 'live'
 219
 220         if is_live:
 221             info_dict.update({
 222                 'title': self._live_title(info_dict['title']),
 223                 'upload_date': None,
 224             })
 225         elif 'upload_date' not in info_dict:
 226             info_dict['upload_date'] = unified_strdate(self._html_search_meta('DC.Date', webpage, 'upload date'))
 227
 228         info_dict.update({
 229             'description': self._html_search_meta('Description', webpage),
 230             'is_live': is_live,
 231         })
 232
 233         return info_dict
 234
 235
 236 class WDRMobileIE(InfoExtractor):
 237     _VALID_URL = r'''(?x)
 238         https?://mobile-ondemand\.wdr\.de/
 239         .*?/fsk(?P<age_limit>[0-9]+)
 240         /[0-9]+/[0-9]+/
 241         (?P<id>[0-9]+)_(?P<title>[0-9]+)'''
 242     IE_NAME = 'wdr:mobile'
 243     _TEST = {
 244         'url': 'http://mobile-ondemand.wdr.de/CMS2010/mdb/ondemand/weltweit/fsk0/42/421735/421735_4283021.mp4',
 245         'info_dict': {
 246             'title': '4283021',
 247             'id': '421735',
 248             'ext': 'mp4',
 249             'age_limit': 0,
 250         },
 251         'skip': 'Problems with loading data.'
 252     }
 253
 254     def _real_extract(self, url):
 255         mobj = re.match(self._VALID_URL, url)
 256         return {
 257             'id': mobj.group('id'),
 258             'title': mobj.group('title'),
 259             'age_limit': int(mobj.group('age_limit')),
 260             'url': url,
 261             'http_headers': {
 262                 'User-Agent': 'mobile',
 263             },
 264         }