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

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 from .common import InfoExtractor
   5 from ..compat import compat_str
   6 from ..utils import (
   7     determine_ext,
   8     parse_duration,
   9     try_get,
  10     unified_strdate,
  11 )
  12
  13
  14 class MediasetIE(InfoExtractor):
  15     _VALID_URL = r'''(?x)
  16                     https?://
  17                         (?:www\.)?video\.mediaset\.it/
  18                         (?:
  19                             (?:video|on-demand)/(?:[^/]+/)+[^/]+_|
  20                             player/playerIFrame(?:Twitter)?\.shtml\?.*?\bid=
  21                         )(?P<id>[0-9]+)
  22                     '''
  23     _TESTS = [{
  24         # full episode
  25         'url': 'http://www.video.mediaset.it/video/hello_goodbye/full/quarta-puntata_661824.html',
  26         'md5': '9b75534d42c44ecef7bf1ffeacb7f85d',
  27         'info_dict': {
  28             'id': '661824',
  29             'ext': 'mp4',
  30             'title': 'Quarta puntata',
  31             'description': 'md5:7183696d6df570e3412a5ef74b27c5e2',
  32             'thumbnail': r're:^https?://.*\.jpg$',
  33             'duration': 1414,
  34             'creator': 'mediaset',
  35             'release_date': '20161107',
  36             'series': 'Hello Goodbye',
  37             'categories': ['reality'],
  38         },
  39         'expected_warnings': ['is not a supported codec'],
  40     }, {
  41         # clip
  42         'url': 'http://www.video.mediaset.it/video/gogglebox/clip/un-grande-classico-della-commedia-sexy_661680.html',
  43         'only_matching': True,
  44     }, {
  45         # iframe simple
  46         'url': 'http://www.video.mediaset.it/player/playerIFrame.shtml?id=665924&autoplay=true',
  47         'only_matching': True,
  48     }, {
  49         # iframe twitter (from http://www.wittytv.it/se-prima-mi-fidavo-zero/)
  50         'url': 'https://www.video.mediaset.it/player/playerIFrameTwitter.shtml?id=665104&playrelated=false&autoplay=false&related=true&hidesocial=true',
  51         'only_matching': True,
  52     }]
  53
  54     def _real_extract(self, url):
  55         video_id = self._match_id(url)
  56
  57         video_list = self._download_json(
  58             'http://cdnsel01.mediaset.net/GetCdn.aspx',
  59             video_id, 'Downloading video CDN JSON', query={
  60                 'streamid': video_id,
  61                 'format': 'json',
  62             })['videoList']
  63
  64         formats = []
  65         for format_url in video_list:
  66             if '.ism' in format_url:
  67                 formats.extend(self._extract_ism_formats(
  68                     format_url, video_id, ism_id='mss', fatal=False))
  69             else:
  70                 formats.append({
  71                     'url': format_url,
  72                     'format_id': determine_ext(format_url),
  73                 })
  74         self._sort_formats(formats)
  75
  76         mediainfo = self._download_json(
  77             'http://plr.video.mediaset.it/html/metainfo.sjson',
  78             video_id, 'Downloading video info JSON', query={
  79                 'id': video_id,
  80             })['video']
  81
  82         title = mediainfo['title']
  83
  84         creator = try_get(
  85             mediainfo, lambda x: x['brand-info']['publisher'], compat_str)
  86         category = try_get(
  87             mediainfo, lambda x: x['brand-info']['category'], compat_str)
  88         categories = [category] if category else None
  89
  90         return {
  91             'id': video_id,
  92             'title': title,
  93             'description': mediainfo.get('short-description'),
  94             'thumbnail': mediainfo.get('thumbnail'),
  95             'duration': parse_duration(mediainfo.get('duration')),
  96             'creator': creator,
  97             'release_date': unified_strdate(mediainfo.get('production-date')),
  98             'webpage_url': mediainfo.get('url'),
  99             'series': mediainfo.get('brand-value'),
 100             'categories': categories,
 101             'formats': formats,
 102         }