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, in a tag with the class "videoButton" (previously a link
  23         # to the page in a multiline "videoLink"-tag)
  24         json_metadata = self._html_search_regex(
  25             r'class=(?:"(?:mediaLink|wdrrPlayerPlayBtn|videoButton)\b[^"]*"[^>]+|"videoLink\b[^"]*"[\s]*>\n[^\n]*)data-extension="([^"]+)"',
  26             webpage, 'media link', default=None, flags=re.MULTILINE)
  27
  28         if not json_metadata:
  29             return
  30
  31         media_link_obj = self._parse_json(json_metadata, display_id,
  32                                           transform_source=js_to_json)
  33         jsonp_url = media_link_obj['mediaObj']['url']
  34
  35         metadata = self._download_json(
  36             jsonp_url, display_id, transform_source=strip_jsonp)
  37
  38         metadata_tracker_data = metadata['trackerData']
  39         metadata_media_resource = metadata['mediaResource']
  40
  41         formats = []
  42
  43         # check if the metadata contains a direct URL to a file
  44         for kind, media_resource in metadata_media_resource.items():
  45             if kind not in ('dflt', 'alt'):
  46                 continue
  47
  48             for tag_name, medium_url in media_resource.items():
  49                 if tag_name not in ('videoURL', 'audioURL'):
  50                     continue
  51
  52                 ext = determine_ext(medium_url)
  53                 if ext == 'm3u8':
  54                     formats.extend(self._extract_m3u8_formats(
  55                         medium_url, display_id, 'mp4', 'm3u8_native',
  56                         m3u8_id='hls'))
  57                 elif ext == 'f4m':
  58                     manifest_url = update_url_query(
  59                         medium_url, {'hdcore': '3.2.0', 'plugin': 'aasp-3.2.0.77.18'})
  60                     formats.extend(self._extract_f4m_formats(
  61                         manifest_url, display_id, f4m_id='hds', fatal=False))
  62                 elif ext == 'smil':
  63                     formats.extend(self._extract_smil_formats(
  64                         medium_url, 'stream', fatal=False))
  65                 else:
  66                     a_format = {
  67                         'url': medium_url
  68                     }
  69                     if ext == 'unknown_video':
  70                         urlh = self._request_webpage(
  71                             medium_url, display_id, note='Determining extension')
  72                         ext = urlhandle_detect_ext(urlh)
  73                         a_format['ext'] = ext
  74                     formats.append(a_format)
  75
  76         self._sort_formats(formats)
  77
  78         subtitles = {}
  79         caption_url = metadata_media_resource.get('captionURL')
  80         if caption_url:
  81             subtitles['de'] = [{
  82                 'url': caption_url,
  83                 'ext': 'ttml',
  84             }]
  85
  86         title = metadata_tracker_data['trackerClipTitle']
  87
  88         return {
  89             'id': metadata_tracker_data.get('trackerClipId', display_id),
  90             'display_id': display_id,
  91             'title': title,
  92             'alt_title': metadata_tracker_data.get('trackerClipSubcategory'),
  93             'formats': formats,
  94             'subtitles': subtitles,
  95             'upload_date': unified_strdate(metadata_tracker_data.get('trackerClipAirTime')),
  96         }
  97
  98
  99 class WDRIE(WDRBaseIE):
 100     _CURRENT_MAUS_URL = r'https?://(?:www\.)wdrmaus.de/(?:[^/]+/){1,2}[^/?#]+\.php5'
 101     _PAGE_REGEX = r'/(?:mediathek/)?[^/]+/(?P<type>[^/]+)/(?P<display_id>.+)\.html'
 102     _VALID_URL = r'(?P<page_url>https?://(?:www\d\.)?wdr\d?\.de)' + _PAGE_REGEX + '|' + _CURRENT_MAUS_URL
 103
 104     _TESTS = [
 105         {
 106             'url': 'http://www1.wdr.de/mediathek/video/sendungen/doku-am-freitag/video-geheimnis-aachener-dom-100.html',
 107             # HDS download, MD5 is unstable
 108             'info_dict': {
 109                 'id': 'mdb-1058683',
 110                 'ext': 'flv',
 111                 'display_id': 'doku-am-freitag/video-geheimnis-aachener-dom-100',
 112                 'title': 'Geheimnis Aachener Dom',
 113                 'alt_title': 'Doku am Freitag',
 114                 'upload_date': '20160304',
 115                 'description': 'md5:87be8ff14d8dfd7a7ee46f0299b52318',
 116                 'is_live': False,
 117                 'subtitles': {'de': [{
 118                     'url': 'http://ondemand-ww.wdr.de/medp/fsk0/105/1058683/1058683_12220974.xml',
 119                     'ext': 'ttml',
 120                 }]},
 121             },
 122         },
 123         {
 124             'url': 'http://www1.wdr.de/mediathek/audio/wdr3/wdr3-gespraech-am-samstag/audio-schriftstellerin-juli-zeh-100.html',
 125             'md5': 'f4c1f96d01cf285240f53ea4309663d8',
 126             'info_dict': {
 127                 'id': 'mdb-1072000',
 128                 'ext': 'mp3',
 129                 'display_id': 'wdr3-gespraech-am-samstag/audio-schriftstellerin-juli-zeh-100',
 130                 'title': 'Schriftstellerin Juli Zeh',
 131                 'alt_title': 'WDR 3 Gespräch am Samstag',
 132                 'upload_date': '20160312',
 133                 'description': 'md5:e127d320bc2b1f149be697ce044a3dd7',
 134                 'is_live': False,
 135                 'subtitles': {}
 136             },
 137         },
 138         {
 139             'url': 'http://www1.wdr.de/mediathek/video/live/index.html',
 140             'info_dict': {
 141                 'id': 'mdb-103364',
 142                 'ext': 'mp4',
 143                 'display_id': 'index',
 144                 'title': r're:^WDR Fernsehen im Livestream [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
 145                 'alt_title': 'WDR Fernsehen Live',
 146                 'upload_date': None,
 147                 'description': 'md5:ae2ff888510623bf8d4b115f95a9b7c9',
 148                 'is_live': True,
 149                 'subtitles': {}
 150             },
 151             'params': {
 152                 'skip_download': True,  # m3u8 download
 153             },
 154         },
 155         {
 156             'url': 'http://www1.wdr.de/mediathek/video/sendungen/aktuelle-stunde/aktuelle-stunde-120.html',
 157             'playlist_mincount': 8,
 158             'info_dict': {
 159                 'id': 'aktuelle-stunde/aktuelle-stunde-120',
 160             },
 161         },
 162         {
 163             'url': 'http://www.wdrmaus.de/aktuelle-sendung/index.php5',
 164             'info_dict': {
 165                 'id': 'mdb-1323501',
 166                 'ext': 'mp4',
 167                 'upload_date': 're:^[0-9]{8}$',
 168                 'title': 're:^Die Sendung mit der Maus vom [0-9.]{10}$',
 169                 'description': 'Die Seite mit der Maus -',
 170             },
 171             'skip': 'The id changes from week to week because of the new episode'
 172         },
 173         {
 174             'url': 'http://www.wdrmaus.de/filme/sachgeschichten/achterbahn.php5',
 175             'md5': '803138901f6368ee497b4d195bb164f2',
 176             'info_dict': {
 177                 'id': 'mdb-186083',
 178                 'ext': 'mp4',
 179                 'upload_date': '20130919',
 180                 'title': 'Sachgeschichte - Achterbahn ',
 181                 'description': 'Die Seite mit der Maus -',
 182             },
 183         },
 184         {
 185             'url': 'http://www1.wdr.de/radio/player/radioplayer116~_layout-popupVersion.html',
 186             # Live stream, MD5 unstable
 187             'info_dict': {
 188                 'id': 'mdb-869971',
 189                 'ext': 'flv',
 190                 'title': 'COSMO Livestream',
 191                 'description': 'md5:2309992a6716c347891c045be50992e4',
 192                 'upload_date': '20160101',
 193             },
 194         }
 195     ]
 196
 197     def _real_extract(self, url):
 198         mobj = re.match(self._VALID_URL, url)
 199         url_type = mobj.group('type')
 200         page_url = mobj.group('page_url')
 201         display_id = mobj.group('display_id')
 202         webpage = self._download_webpage(url, display_id)
 203
 204         info_dict = self._extract_wdr_video(webpage, display_id)
 205
 206         if not info_dict:
 207             entries = [
 208                 self.url_result(page_url + href[0], 'WDR')
 209                 for href in re.findall(
 210                     r'<a href="(%s)"[^>]+data-extension=' % self._PAGE_REGEX,
 211                     webpage)
 212             ]
 213
 214             if entries:  # Playlist page
 215                 return self.playlist_result(entries, playlist_id=display_id)
 216
 217             raise ExtractorError('No downloadable streams found', expected=True)
 218
 219         is_live = url_type == 'live'
 220
 221         if is_live:
 222             info_dict.update({
 223                 'title': self._live_title(info_dict['title']),
 224                 'upload_date': None,
 225             })
 226         elif 'upload_date' not in info_dict:
 227             info_dict['upload_date'] = unified_strdate(self._html_search_meta('DC.Date', webpage, 'upload date'))
 228
 229         info_dict.update({
 230             'description': self._html_search_meta('Description', webpage),
 231             'is_live': is_live,
 232         })
 233
 234         return info_dict
 235
 236
 237 class WDRMobileIE(InfoExtractor):
 238     _VALID_URL = r'''(?x)
 239         https?://mobile-ondemand\.wdr\.de/
 240         .*?/fsk(?P<age_limit>[0-9]+)
 241         /[0-9]+/[0-9]+/
 242         (?P<id>[0-9]+)_(?P<title>[0-9]+)'''
 243     IE_NAME = 'wdr:mobile'
 244     _TEST = {
 245         'url': 'http://mobile-ondemand.wdr.de/CMS2010/mdb/ondemand/weltweit/fsk0/42/421735/421735_4283021.mp4',
 246         'info_dict': {
 247             'title': '4283021',
 248             'id': '421735',
 249             'ext': 'mp4',
 250             'age_limit': 0,
 251         },
 252         'skip': 'Problems with loading data.'
 253     }
 254
 255     def _real_extract(self, url):
 256         mobj = re.match(self._VALID_URL, url)
 257         return {
 258             'id': mobj.group('id'),
 259             'title': mobj.group('title'),
 260             'age_limit': int(mobj.group('age_limit')),
 261             'url': url,
 262             'http_headers': {
 263                 'User-Agent': 'mobile',
 264             },
 265         }