_ 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
  64 class RaiTVIE(RaiBaseIE):
  65     _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'
  66     _TESTS = [
  67         {
  68             'url': 'http://www.rai.tv/dl/RaiTV/programmi/media/ContentItem-cb27157f-9dd0-4aee-b788-b1f67643a391.html',
  69             'md5': '8970abf8caf8aef4696e7b1f2adfc696',
  70             'info_dict': {
  71                 'id': 'cb27157f-9dd0-4aee-b788-b1f67643a391',
  72                 'ext': 'mp4',
  73                 'title': 'Report del 07/04/2014',
  74                 'description': 'md5:f27c544694cacb46a078db84ec35d2d9',
  75                 'upload_date': '20140407',
  76                 'duration': 6160,
  77                 'thumbnail': 're:^https?://.*\.jpg$',
  78             }
  79         },
  80         {
  81             # no m3u8 stream
  82             'url': 'http://www.raisport.rai.it/dl/raiSport/media/rassegna-stampa-04a9f4bd-b563-40cf-82a6-aad3529cb4a9.html',
  83             # HDS download, MD5 is unstable
  84             'info_dict': {
  85                 'id': '04a9f4bd-b563-40cf-82a6-aad3529cb4a9',
  86                 'ext': 'flv',
  87                 'title': 'TG PRIMO TEMPO',
  88                 'upload_date': '20140612',
  89                 'duration': 1758,
  90                 'thumbnail': 're:^https?://.*\.jpg$',
  91             },
  92             'skip': 'Geo-restricted to Italy',
  93         },
  94         {
  95             'url': 'http://www.rainews.it/dl/rainews/media/state-of-the-net-Antonella-La-Carpia-regole-virali-7aafdea9-0e5d-49d5-88a6-7e65da67ae13.html',
  96             'md5': '35cf7c229f22eeef43e48b5cf923bef0',
  97             'info_dict': {
  98                 'id': '7aafdea9-0e5d-49d5-88a6-7e65da67ae13',
  99                 'ext': 'mp4',
 100                 'title': 'State of the Net, Antonella La Carpia: regole virali',
 101                 'description': 'md5:b0ba04a324126903e3da7763272ae63c',
 102                 'upload_date': '20140613',
 103             },
 104             'skip': 'Error 404',
 105         },
 106         {
 107             'url': 'http://www.rai.tv/dl/RaiTV/programmi/media/ContentItem-b4a49761-e0cc-4b14-8736-2729f6f73132-tg2.html',
 108             'info_dict': {
 109                 'id': 'b4a49761-e0cc-4b14-8736-2729f6f73132',
 110                 'ext': 'mp4',
 111                 'title': 'Alluvione in Sardegna e dissesto idrogeologico',
 112                 'description': 'Edizione delle ore 20:30 ',
 113             },
 114             'skip': 'invalid urls',
 115         },
 116         {
 117             'url': 'http://www.ilcandidato.rai.it/dl/ray/media/Il-Candidato---Primo-episodio-Le-Primarie-28e5525a-b495-45e8-a7c3-bc48ba45d2b6.html',
 118             'md5': 'e57493e1cb8bc7c564663f363b171847',
 119             'info_dict': {
 120                 'id': '28e5525a-b495-45e8-a7c3-bc48ba45d2b6',
 121                 'ext': 'mp4',
 122                 'title': 'Il Candidato - Primo episodio: "Le Primarie"',
 123                 'description': 'md5:364b604f7db50594678f483353164fb8',
 124                 'upload_date': '20140923',
 125                 'duration': 386,
 126                 'thumbnail': 're:^https?://.*\.jpg$',
 127             }
 128         },
 129     ]
 130
 131     def _real_extract(self, url):
 132         video_id = self._match_id(url)
 133
 134         media = self._download_json(
 135             'http://www.rai.tv/dl/RaiTV/programmi/media/ContentItem-%s.html?json' % video_id,
 136             video_id, 'Downloading video JSON')
 137
 138         thumbnails = []
 139         for image_type in ('image', 'image_medium', 'image_300'):
 140             thumbnail_url = media.get(image_type)
 141             if thumbnail_url:
 142                 thumbnails.append({
 143                     'url': compat_urlparse.urljoin(url, thumbnail_url),
 144                 })
 145
 146         formats = []
 147         media_type = media['type']
 148         if 'Audio' in media_type:
 149             formats.append({
 150                 'format_id': media.get('formatoAudio'),
 151                 'url': media['audioUrl'],
 152                 'ext': media.get('formatoAudio'),
 153             })
 154         elif 'Video' in media_type:
 155             formats.extend(self._extract_relinker_formats(media['mediaUri'], video_id))
 156             self._sort_formats(formats)
 157         else:
 158             raise ExtractorError('not a media file')
 159
 160         subtitles = {}
 161         captions = media.get('subtitlesUrl')
 162         if captions:
 163             STL_EXT = '.stl'
 164             SRT_EXT = '.srt'
 165             if captions.endswith(STL_EXT):
 166                 captions = captions[:-len(STL_EXT)] + SRT_EXT
 167             subtitles['it'] = [{
 168                 'ext': 'srt',
 169                 'url': captions,
 170             }]
 171
 172         return {
 173             'id': video_id,
 174             'title': media['name'],
 175             'description': media.get('desc'),
 176             'thumbnails': thumbnails,
 177             'uploader': media.get('author'),
 178             'upload_date': unified_strdate(media.get('date')),
 179             'duration': parse_duration(media.get('length')),
 180             'formats': formats,
 181             'subtitles': subtitles,
 182         }
 183
 184
 185 class RaiIE(RaiBaseIE):
 186     _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'
 187     _TESTS = [
 188         {
 189             'url': 'http://www.report.rai.it/dl/Report/puntata/ContentItem-0c7a664b-d0f4-4b2c-8835-3f82e46f433e.html',
 190             'md5': '2dd727e61114e1ee9c47f0da6914e178',
 191             'info_dict': {
 192                 'id': '59d69d28-6bb6-409d-a4b5-ed44096560af',
 193                 'ext': 'mp4',
 194                 'title': 'Il pacco',
 195                 'description': 'md5:4b1afae1364115ce5d78ed83cd2e5b3a',
 196                 'upload_date': '20141221',
 197             },
 198         },
 199         {
 200             'url': 'http://www.rai.tv/dl/RaiTV/dirette/PublishingBlock-1912dbbf-3f96-44c3-b4cf-523681fbacbc.html?channel=EuroNews',
 201             # HDS live stream, MD5 is unstable
 202             'info_dict': {
 203                 'id': '1912dbbf-3f96-44c3-b4cf-523681fbacbc',
 204                 'ext': 'flv',
 205                 'title': 'EuroNews',
 206             },
 207             'skip': 'Geo-restricted to Italy',
 208         }
 209     ]
 210
 211     @classmethod
 212     def suitable(cls, url):
 213         return False if RaiTVIE.suitable(url) else super(RaiIE, cls).suitable(url)
 214
 215     def _real_extract(self, url):
 216         video_id = self._match_id(url)
 217         webpage = self._download_webpage(url, video_id)
 218
 219         iframe_url = self._search_regex(
 220             [r'<iframe[^>]+src="([^"]*/dl/[^"]+\?iframe\b[^"]*)"',
 221              r'drawMediaRaiTV\(["\'](.+?)["\']'],
 222             webpage, 'iframe', default=None)
 223         if iframe_url:
 224             if not iframe_url.startswith('http'):
 225                 iframe_url = compat_urlparse.urljoin(url, iframe_url)
 226             return self.url_result(iframe_url)
 227
 228         relinker_url = compat_urlparse.urljoin(url, self._search_regex(
 229             r'var\s+videoURL\s*=\s*(?P<q1>[\'"])(?P<url>(https?:)?//mediapolis\.rai\.it/relinker/relinkerServlet\.htm\?cont=\d+)(?P=q1)',
 230             webpage, 'relinker URL', group='url'))
 231         formats = self._extract_relinker_formats(relinker_url, video_id)
 232         self._sort_formats(formats)
 233
 234         title = self._search_regex(
 235             r'var\s+videoTitolo\s*=\s*([\'"])(?P<title>[^\'"]+)\1',
 236             webpage, 'title', group='title', default=None) or self._og_search_title(webpage)
 237
 238         return {
 239             'id': video_id,
 240             'title': title,
 241             'formats': formats,
 242         }