_ Git - youtube-dl/blob - youtube_dl/extractor/rai.py

   1 from __future__ import unicode_literals
   2
   3 from .common import InfoExtractor
   4 from ..compat import compat_urlparse
   5 from ..utils import (
   6     determine_ext,
   7     ExtractorError,
   8     find_xpath_attr,
   9     fix_xml_ampersands,
  10     int_or_none,
  11     parse_duration,
  12     unified_strdate,
  13     update_url_query,
  14     xpath_text,
  15 )
  16
  17
  18 class RaiBaseIE(InfoExtractor):
  19     def _extract_relinker_formats(self, relinker_url, video_id):
  20         formats = []
  21
  22         for platform in ('mon', 'flash', 'native'):
  23             headers = {}
  24             # TODO: rename --cn-verification-proxy
  25             cn_verification_proxy = self._downloader.params.get('cn_verification_proxy')
  26             if cn_verification_proxy:
  27                 headers['Ytdl-request-proxy'] = cn_verification_proxy
  28
  29             relinker = self._download_xml(
  30                 relinker_url, video_id,
  31                 note='Downloading XML metadata for platform %s' % platform,
  32                 transform_source=fix_xml_ampersands,
  33                 query={'output': 45, 'pl': platform}, headers=headers)
  34
  35             media_url = find_xpath_attr(relinker, './url', 'type', 'content').text
  36             if media_url == 'http://download.rai.it/video_no_available.mp4':
  37                 self.raise_geo_restricted()
  38
  39             ext = determine_ext(media_url)
  40             if (ext == 'm3u8' and platform != 'mon') or (ext == 'f4m' and platform != 'flash'):
  41                 continue
  42
  43             if ext == 'm3u8':
  44                 formats.extend(self._extract_m3u8_formats(
  45                     media_url, video_id, 'mp4', 'm3u8_native',
  46                     m3u8_id='hls', fatal=False))
  47             elif ext == 'f4m':
  48                 manifest_url = update_url_query(
  49                     media_url.replace('manifest#live_hds.f4m', 'manifest.f4m'),
  50                     {'hdcore': '3.7.0', 'plugin': 'aasp-3.7.0.39.44'})
  51                 formats.extend(self._extract_f4m_formats(
  52                     manifest_url, video_id, f4m_id='hds', fatal=False))
  53             else:
  54                 bitrate = int_or_none(xpath_text(relinker, 'bitrate'))
  55                 formats.append({
  56                     'url': media_url,
  57                     'tbr': bitrate if bitrate > 0 else None,
  58                     'format_id': 'http-%d' % bitrate if bitrate > 0 else 'http',
  59                 })
  60
  61         return formats
  62
  63     def _extract_from_content_id(self, content_id, base_url):
  64         media = self._download_json(
  65             'http://www.rai.tv/dl/RaiTV/programmi/media/ContentItem-%s.html?json' % content_id,
  66             content_id, 'Downloading video JSON')
  67
  68         thumbnails = []
  69         for image_type in ('image', 'image_medium', 'image_300'):
  70             thumbnail_url = media.get(image_type)
  71             if thumbnail_url:
  72                 thumbnails.append({
  73                     'url': compat_urlparse.urljoin(base_url, thumbnail_url),
  74                 })
  75
  76         formats = []
  77         media_type = media['type']
  78         if 'Audio' in media_type:
  79             formats.append({
  80                 'format_id': media.get('formatoAudio'),
  81                 'url': media['audioUrl'],
  82                 'ext': media.get('formatoAudio'),
  83             })
  84         elif 'Video' in media_type:
  85             formats.extend(self._extract_relinker_formats(media['mediaUri'], content_id))
  86             self._sort_formats(formats)
  87         else:
  88             raise ExtractorError('not a media file')
  89
  90         subtitles = {}
  91         captions = media.get('subtitlesUrl')
  92         if captions:
  93             STL_EXT = '.stl'
  94             SRT_EXT = '.srt'
  95             if captions.endswith(STL_EXT):
  96                 captions = captions[:-len(STL_EXT)] + SRT_EXT
  97             subtitles['it'] = [{
  98                 'ext': 'srt',
  99                 'url': captions,
 100             }]
 101
 102         return {
 103             'id': content_id,
 104             'title': media['name'],
 105             'description': media.get('desc'),
 106             'thumbnails': thumbnails,
 107             'uploader': media.get('author'),
 108             'upload_date': unified_strdate(media.get('date')),
 109             'duration': parse_duration(media.get('length')),
 110             'formats': formats,
 111             'subtitles': subtitles,
 112         }
 113
 114
 115 class RaiTVIE(RaiBaseIE):
 116     _VALID_URL = r'https?://(?:.+?\.)?(?:rai\.it|rai\.tv|rainews\.it)/dl/(?:[^/]+/)+(?:media|ondemand)/.+?-(?P<id>[\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})(?:-.+?)?\.html'
 117     _TESTS = [
 118         {
 119             'url': 'http://www.rai.tv/dl/RaiTV/programmi/media/ContentItem-cb27157f-9dd0-4aee-b788-b1f67643a391.html',
 120             'md5': '8970abf8caf8aef4696e7b1f2adfc696',
 121             'info_dict': {
 122                 'id': 'cb27157f-9dd0-4aee-b788-b1f67643a391',
 123                 'ext': 'mp4',
 124                 'title': 'Report del 07/04/2014',
 125                 'description': 'md5:f27c544694cacb46a078db84ec35d2d9',
 126                 'upload_date': '20140407',
 127                 'duration': 6160,
 128                 'thumbnail': 're:^https?://.*\.jpg$',
 129             }
 130         },
 131         {
 132             # no m3u8 stream
 133             'url': 'http://www.raisport.rai.it/dl/raiSport/media/rassegna-stampa-04a9f4bd-b563-40cf-82a6-aad3529cb4a9.html',
 134             # HDS download, MD5 is unstable
 135             'info_dict': {
 136                 'id': '04a9f4bd-b563-40cf-82a6-aad3529cb4a9',
 137                 'ext': 'flv',
 138                 'title': 'TG PRIMO TEMPO',
 139                 'upload_date': '20140612',
 140                 'duration': 1758,
 141                 'thumbnail': 're:^https?://.*\.jpg$',
 142             },
 143             'skip': 'Geo-restricted to Italy',
 144         },
 145         {
 146             'url': 'http://www.rainews.it/dl/rainews/media/state-of-the-net-Antonella-La-Carpia-regole-virali-7aafdea9-0e5d-49d5-88a6-7e65da67ae13.html',
 147             'md5': '35cf7c229f22eeef43e48b5cf923bef0',
 148             'info_dict': {
 149                 'id': '7aafdea9-0e5d-49d5-88a6-7e65da67ae13',
 150                 'ext': 'mp4',
 151                 'title': 'State of the Net, Antonella La Carpia: regole virali',
 152                 'description': 'md5:b0ba04a324126903e3da7763272ae63c',
 153                 'upload_date': '20140613',
 154             },
 155             'skip': 'Error 404',
 156         },
 157         {
 158             'url': 'http://www.rai.tv/dl/RaiTV/programmi/media/ContentItem-b4a49761-e0cc-4b14-8736-2729f6f73132-tg2.html',
 159             'info_dict': {
 160                 'id': 'b4a49761-e0cc-4b14-8736-2729f6f73132',
 161                 'ext': 'mp4',
 162                 'title': 'Alluvione in Sardegna e dissesto idrogeologico',
 163                 'description': 'Edizione delle ore 20:30 ',
 164             },
 165             'skip': 'invalid urls',
 166         },
 167         {
 168             'url': 'http://www.ilcandidato.rai.it/dl/ray/media/Il-Candidato---Primo-episodio-Le-Primarie-28e5525a-b495-45e8-a7c3-bc48ba45d2b6.html',
 169             'md5': 'e57493e1cb8bc7c564663f363b171847',
 170             'info_dict': {
 171                 'id': '28e5525a-b495-45e8-a7c3-bc48ba45d2b6',
 172                 'ext': 'mp4',
 173                 'title': 'Il Candidato - Primo episodio: "Le Primarie"',
 174                 'description': 'md5:364b604f7db50594678f483353164fb8',
 175                 'upload_date': '20140923',
 176                 'duration': 386,
 177                 'thumbnail': 're:^https?://.*\.jpg$',
 178             }
 179         },
 180     ]
 181
 182     def _real_extract(self, url):
 183         video_id = self._match_id(url)
 184
 185         return self._extract_from_content_id(video_id, url)
 186
 187
 188 class RaiIE(RaiBaseIE):
 189     _VALID_URL = r'https?://(?:.+?\.)?(?:rai\.it|rai\.tv|rainews\.it)/dl/.+?-(?P<id>[\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})(?:-.+?)?\.html'
 190     _TESTS = [
 191         {
 192             'url': 'http://www.report.rai.it/dl/Report/puntata/ContentItem-0c7a664b-d0f4-4b2c-8835-3f82e46f433e.html',
 193             'md5': '2dd727e61114e1ee9c47f0da6914e178',
 194             'info_dict': {
 195                 'id': '59d69d28-6bb6-409d-a4b5-ed44096560af',
 196                 'ext': 'mp4',
 197                 'title': 'Il pacco',
 198                 'description': 'md5:4b1afae1364115ce5d78ed83cd2e5b3a',
 199                 'upload_date': '20141221',
 200             },
 201         },
 202         {
 203             # Direct relinker URL
 204             'url': 'http://www.rai.tv/dl/RaiTV/dirette/PublishingBlock-1912dbbf-3f96-44c3-b4cf-523681fbacbc.html?channel=EuroNews',
 205             # HDS live stream, MD5 is unstable
 206             'info_dict': {
 207                 'id': '1912dbbf-3f96-44c3-b4cf-523681fbacbc',
 208                 'ext': 'flv',
 209                 'title': 'EuroNews',
 210             },
 211             'skip': 'Geo-restricted to Italy',
 212         },
 213         {
 214             # Embedded content item ID
 215             'url': 'http://www.tg1.rai.it/dl/tg1/2010/edizioni/ContentSet-9b6e0cba-4bef-4aef-8cf0-9f7f665b7dfb-tg1.html?item=undefined',
 216             'md5': '84c1135ce960e8822ae63cec34441d63',
 217             'info_dict': {
 218                 'id': '0960e765-62c8-474a-ac4b-7eb3e2be39c8',
 219                 'ext': 'mp4',
 220                 'title': 'TG1 ore 20:00 del 02/07/2016',
 221                 'upload_date': '20160702',
 222             },
 223         },
 224     ]
 225
 226     @classmethod
 227     def suitable(cls, url):
 228         return False if RaiTVIE.suitable(url) else super(RaiIE, cls).suitable(url)
 229
 230     def _real_extract(self, url):
 231         video_id = self._match_id(url)
 232         webpage = self._download_webpage(url, video_id)
 233
 234         iframe_url = self._search_regex(
 235             [r'<iframe[^>]+src="([^"]*/dl/[^"]+\?iframe\b[^"]*)"',
 236              r'drawMediaRaiTV\(["\'](.+?)["\']'],
 237             webpage, 'iframe', default=None)
 238         if iframe_url:
 239             if not iframe_url.startswith('http'):
 240                 iframe_url = compat_urlparse.urljoin(url, iframe_url)
 241             return self.url_result(iframe_url)
 242
 243         content_item_id = self._search_regex(
 244             r'initEdizione\((?P<q1>[\'"])ContentItem-(?P<content_id>[^\'"]+)(?P=q1)',
 245             webpage, 'content item ID', group='content_id', default=None)
 246         if content_item_id:
 247             return self._extract_from_content_id(content_item_id, url)
 248
 249         relinker_url = compat_urlparse.urljoin(url, self._search_regex(
 250             r'var\s+videoURL\s*=\s*(?P<q1>[\'"])(?P<url>(https?:)?//mediapolis\.rai\.it/relinker/relinkerServlet\.htm\?cont=\d+)(?P=q1)',
 251             webpage, 'relinker URL', group='url'))
 252         formats = self._extract_relinker_formats(relinker_url, video_id)
 253         self._sort_formats(formats)
 254
 255         title = self._search_regex(
 256             r'var\s+videoTitolo\s*=\s*([\'"])(?P<title>[^\'"]+)\1',
 257             webpage, 'title', group='title', default=None) or self._og_search_title(webpage)
 258
 259         return {
 260             'id': video_id,
 261             'title': title,
 262             'formats': formats,
 263         }