youtube_dl/extractor/mediaset.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .theplatform import ThePlatformBaseIE
   7 from ..compat import (
   8     compat_parse_qs,
   9     compat_urllib_parse_urlparse,
  10 )
  11 from ..utils import (
  12     ExtractorError,
  13     int_or_none,
  14     update_url_query,
  15 )
  16
  17
  18 class MediasetIE(ThePlatformBaseIE):
  19     _TP_TLD = 'eu'
  20     _VALID_URL = r'''(?x)
  21                     (?:
  22                         mediaset:|
  23                         https?://
  24                             (?:(?:www|static3)\.)?mediasetplay\.mediaset\.it/
  25                             (?:
  26                                 (?:video|on-demand|movie)/(?:[^/]+/)+[^/]+_|
  27                                 player/index\.html\?.*?\bprogramGuid=
  28                             )
  29                     )(?P<id>[0-9A-Z]{16,})
  30                     '''
  31     _TESTS = [{
  32         # full episode
  33         'url': 'https://www.mediasetplay.mediaset.it/video/hellogoodbye/quarta-puntata_FAFU000000661824',
  34         'md5': '9b75534d42c44ecef7bf1ffeacb7f85d',
  35         'info_dict': {
  36             'id': 'FAFU000000661824',
  37             'ext': 'mp4',
  38             'title': 'Quarta puntata',
  39             'description': 'md5:d41d8cd98f00b204e9800998ecf8427e',
  40             'thumbnail': r're:^https?://.*\.jpg$',
  41             'duration': 1414.26,
  42             'upload_date': '20161107',
  43             'series': 'Hello Goodbye',
  44             'timestamp': 1478532900,
  45             'uploader': 'Rete 4',
  46             'uploader_id': 'R4',
  47         },
  48     }, {
  49         'url': 'https://www.mediasetplay.mediaset.it/video/matrix/puntata-del-25-maggio_F309013801000501',
  50         'md5': '288532f0ad18307705b01e581304cd7b',
  51         'info_dict': {
  52             'id': 'F309013801000501',
  53             'ext': 'mp4',
  54             'title': 'Puntata del 25 maggio',
  55             'description': 'md5:d41d8cd98f00b204e9800998ecf8427e',
  56             'thumbnail': r're:^https?://.*\.jpg$',
  57             'duration': 6565.007,
  58             'upload_date': '20180526',
  59             'series': 'Matrix',
  60             'timestamp': 1527326245,
  61             'uploader': 'Canale 5',
  62             'uploader_id': 'C5',
  63         },
  64     }, {
  65         # clip
  66         'url': 'https://www.mediasetplay.mediaset.it/video/gogglebox/un-grande-classico-della-commedia-sexy_FAFU000000661680',
  67         'only_matching': True,
  68     }, {
  69         # iframe simple
  70         'url': 'https://static3.mediasetplay.mediaset.it/player/index.html?appKey=5ad3966b1de1c4000d5cec48&programGuid=FAFU000000665924&id=665924',
  71         'only_matching': True,
  72     }, {
  73         # iframe twitter (from http://www.wittytv.it/se-prima-mi-fidavo-zero/)
  74         'url': 'https://static3.mediasetplay.mediaset.it/player/index.html?appKey=5ad3966b1de1c4000d5cec48&programGuid=FAFU000000665104&id=665104',
  75         'only_matching': True,
  76     }, {
  77         'url': 'mediaset:FAFU000000665924',
  78         'only_matching': True,
  79     }, {
  80         'url': 'https://www.mediasetplay.mediaset.it/video/mediasethaacuoreilfuturo/palmieri-alicudi-lisola-dei-tre-bambini-felici--un-decreto-per-alicudi-e-tutte-le-microscuole_FD00000000102295',
  81         'only_matching': True,
  82     }, {
  83         'url': 'https://www.mediasetplay.mediaset.it/video/cherryseason/anticipazioni-degli-episodi-del-23-ottobre_F306837101005C02',
  84         'only_matching': True,
  85     }, {
  86         'url': 'https://www.mediasetplay.mediaset.it/video/tg5/ambiente-onda-umana-per-salvare-il-pianeta_F309453601079D01',
  87         'only_matching': True,
  88     }, {
  89         'url': 'https://www.mediasetplay.mediaset.it/video/grandefratellovip/benedetta-una-doccia-gelata_F309344401044C135',
  90         'only_matching': True,
  91     }, {
  92         'url': 'https://www.mediasetplay.mediaset.it/movie/herculeslaleggendahainizio/hercules-la-leggenda-ha-inizio_F305927501000102',
  93         'only_matching': True,
  94     }]
  95
  96     @staticmethod
  97     def _extract_urls(ie, webpage):
  98         def _qs(url):
  99             return compat_parse_qs(compat_urllib_parse_urlparse(url).query)
 100
 101         def _program_guid(qs):
 102             return qs.get('programGuid', [None])[0]
 103
 104         entries = []
 105         for mobj in re.finditer(
 106                 r'<iframe\b[^>]+\bsrc=(["\'])(?P<url>(?:https?:)?//(?:www\.)?video\.mediaset\.it/player/playerIFrame(?:Twitter)?\.shtml.*?)\1',
 107                 webpage):
 108             embed_url = mobj.group('url')
 109             embed_qs = _qs(embed_url)
 110             program_guid = _program_guid(embed_qs)
 111             if program_guid:
 112                 entries.append(embed_url)
 113                 continue
 114             video_id = embed_qs.get('id', [None])[0]
 115             if not video_id:
 116                 continue
 117             urlh = ie._request_webpage(
 118                 embed_url, video_id, note='Following embed URL redirect')
 119             embed_url = urlh.geturl()
 120             program_guid = _program_guid(_qs(embed_url))
 121             if program_guid:
 122                 entries.append(embed_url)
 123         return entries
 124
 125     def _parse_smil_formats(self, smil, smil_url, video_id, namespace=None, f4m_params=None, transform_rtmp_url=None):
 126         for video in smil.findall(self._xpath_ns('.//video', namespace)):
 127             video.attrib['src'] = re.sub(r'(https?://vod05)t(-mediaset-it\.akamaized\.net/.+?.mpd)\?.+', r'\1\2', video.attrib['src'])
 128         return super(MediasetIE, self)._parse_smil_formats(smil, smil_url, video_id, namespace, f4m_params, transform_rtmp_url)
 129
 130     def _real_extract(self, url):
 131         guid = self._match_id(url)
 132         tp_path = 'PR1GhC/media/guid/2702976343/' + guid
 133         info = self._extract_theplatform_metadata(tp_path, guid)
 134
 135         formats = []
 136         subtitles = {}
 137         first_e = None
 138         for asset_type in ('SD', 'HD'):
 139             # TODO: fixup ISM+none manifest URLs
 140             for f in ('MPEG4', 'MPEG-DASH+none', 'M3U+none'):
 141                 try:
 142                     tp_formats, tp_subtitles = self._extract_theplatform_smil(
 143                         update_url_query('http://link.theplatform.%s/s/%s' % (self._TP_TLD, tp_path), {
 144                             'mbr': 'true',
 145                             'formats': f,
 146                             'assetTypes': asset_type,
 147                         }), guid, 'Downloading %s %s SMIL data' % (f.split('+')[0], asset_type))
 148                 except ExtractorError as e:
 149                     if not first_e:
 150                         first_e = e
 151                     break
 152                 for tp_f in tp_formats:
 153                     tp_f['quality'] = 1 if asset_type == 'HD' else 0
 154                 formats.extend(tp_formats)
 155                 subtitles = self._merge_subtitles(subtitles, tp_subtitles)
 156         if first_e and not formats:
 157             raise first_e
 158         self._sort_formats(formats)
 159
 160         fields = []
 161         for templ, repls in (('tvSeason%sNumber', ('', 'Episode')), ('mediasetprogram$%s', ('brandTitle', 'numberOfViews', 'publishInfo'))):
 162             fields.extend(templ % repl for repl in repls)
 163         feed_data = self._download_json(
 164             'https://feed.entertainment.tv.theplatform.eu/f/PR1GhC/mediaset-prod-all-programs/guid/-/' + guid,
 165             guid, fatal=False, query={'fields': ','.join(fields)})
 166         if feed_data:
 167             publish_info = feed_data.get('mediasetprogram$publishInfo') or {}
 168             info.update({
 169                 'episode_number': int_or_none(feed_data.get('tvSeasonEpisodeNumber')),
 170                 'season_number': int_or_none(feed_data.get('tvSeasonNumber')),
 171                 'series': feed_data.get('mediasetprogram$brandTitle'),
 172                 'uploader': publish_info.get('description'),
 173                 'uploader_id': publish_info.get('channel'),
 174                 'view_count': int_or_none(feed_data.get('mediasetprogram$numberOfViews')),
 175             })
 176
 177         info.update({
 178             'id': guid,
 179             'formats': formats,
 180             'subtitles': subtitles,
 181         })
 182         return info