Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/tv4.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     ExtractorError,
   8     int_or_none,
   9     parse_iso8601,
  10     try_get,
  11     update_url_query,
  12 )
  13
  14
  15 class TV4IE(InfoExtractor):
  16     IE_DESC = 'tv4.se and tv4play.se'
  17     _VALID_URL = r'''(?x)https?://(?:www\.)?
  18         (?:
  19             tv4\.se/(?:[^/]+)/klipp/(?:.*)-|
  20             tv4play\.se/
  21             (?:
  22                 (?:program|barn)/(?:[^\?]+)\?video_id=|
  23                 iframe/video/|
  24                 film/|
  25                 sport/|
  26             )
  27         )(?P<id>[0-9]+)'''
  28     _TESTS = [
  29         {
  30             'url': 'http://www.tv4.se/kalla-fakta/klipp/kalla-fakta-5-english-subtitles-2491650',
  31             'md5': '909d6454b87b10a25aa04c4bdd416a9b',
  32             'info_dict': {
  33                 'id': '2491650',
  34                 'ext': 'mp4',
  35                 'title': 'Kalla Fakta 5 (english subtitles)',
  36                 'thumbnail': 're:^https?://.*\.jpg$',
  37                 'timestamp': int,
  38                 'upload_date': '20131125',
  39             },
  40         },
  41         {
  42             'url': 'http://www.tv4play.se/iframe/video/3054113',
  43             'md5': '77f851c55139ffe0ebd41b6a5552489b',
  44             'info_dict': {
  45                 'id': '3054113',
  46                 'ext': 'mp4',
  47                 'title': 'Så här jobbar ficktjuvarna - se avslöjande bilder',
  48                 'thumbnail': 're:^https?://.*\.jpg$',
  49                 'description': 'Unika bilder avslöjar hur turisternas fickor vittjas mitt på Stockholms central. Två experter på ficktjuvarna avslöjar knepen du ska se upp för.',
  50                 'timestamp': int,
  51                 'upload_date': '20150130',
  52             },
  53         },
  54         {
  55             'url': 'http://www.tv4play.se/sport/3060959',
  56             'only_matching': True,
  57         },
  58         {
  59             'url': 'http://www.tv4play.se/film/2378136',
  60             'only_matching': True,
  61         },
  62         {
  63             'url': 'http://www.tv4play.se/barn/looney-tunes?video_id=3062412',
  64             'only_matching': True,
  65         },
  66     ]
  67
  68     def _real_extract(self, url):
  69         video_id = self._match_id(url)
  70
  71         info = self._download_json(
  72             'http://www.tv4play.se/player/assets/%s.json' % video_id,
  73             video_id, 'Downloading video info JSON')
  74
  75         # If is_geo_restricted is true, it doesn't necessarily mean we can't download it
  76         if info.get('is_geo_restricted'):
  77             self.report_warning('This content might not be available in your country due to licensing restrictions.')
  78         if info.get('requires_subscription'):
  79             raise ExtractorError('This content requires subscription.', expected=True)
  80
  81         title = info['title']
  82
  83         formats = []
  84         # http formats are linked with unresolvable host
  85         for kind in ('hls', ''):
  86             data = self._download_json(
  87                 'https://prima.tv4play.se/api/web/asset/%s/play.json' % video_id,
  88                 video_id, 'Downloading sources JSON', query={
  89                     'protocol': kind,
  90                     'videoFormat': 'MP4+WEBVTTS+WEBVTT',
  91                 })
  92             item = try_get(data, lambda x: x['playback']['items']['item'], dict)
  93             manifest_url = item.get('url')
  94             if not isinstance(manifest_url, compat_str):
  95                 continue
  96             if kind == 'hls':
  97                 formats.extend(self._extract_m3u8_formats(
  98                     manifest_url, video_id, 'mp4', entry_protocol='m3u8_native',
  99                     m3u8_id=kind, fatal=False))
 100             else:
 101                 formats.extend(self._extract_f4m_formats(
 102                     update_url_query(manifest_url, {'hdcore': '3.8.0'}),
 103                     video_id, f4m_id='hds', fatal=False))
 104         self._sort_formats(formats)
 105
 106         return {
 107             'id': video_id,
 108             'title': title,
 109             'formats': formats,
 110             'description': info.get('description'),
 111             'timestamp': parse_iso8601(info.get('broadcast_date_time')),
 112             'duration': int_or_none(info.get('duration')),
 113             'thumbnail': info.get('image'),
 114             'is_live': info.get('is_live') is True,
 115         }