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     int_or_none,
   8     parse_iso8601,
   9     try_get,
  10     determine_ext,
  11 )
  12
  13
  14 class TV4IE(InfoExtractor):
  15     IE_DESC = 'tv4.se and tv4play.se'
  16     _VALID_URL = r'''(?x)https?://(?:www\.)?
  17         (?:
  18             tv4\.se/(?:[^/]+)/klipp/(?:.*)-|
  19             tv4play\.se/
  20             (?:
  21                 (?:program|barn)/(?:[^\?]+)\?video_id=|
  22                 iframe/video/|
  23                 film/|
  24                 sport/|
  25             )
  26         )(?P<id>[0-9]+)'''
  27     _TESTS = [
  28         {
  29             'url': 'http://www.tv4.se/kalla-fakta/klipp/kalla-fakta-5-english-subtitles-2491650',
  30             'md5': 'cb837212f342d77cec06e6dad190e96d',
  31             'info_dict': {
  32                 'id': '2491650',
  33                 'ext': 'mp4',
  34                 'title': 'Kalla Fakta 5 (english subtitles)',
  35                 'thumbnail': r're:^https?://.*\.jpg$',
  36                 'timestamp': int,
  37                 'upload_date': '20131125',
  38             },
  39         },
  40         {
  41             'url': 'http://www.tv4play.se/iframe/video/3054113',
  42             'md5': 'cb837212f342d77cec06e6dad190e96d',
  43             'info_dict': {
  44                 'id': '3054113',
  45                 'ext': 'mp4',
  46                 'title': 'Så här jobbar ficktjuvarna - se avslöjande bilder',
  47                 'thumbnail': r're:^https?://.*\.jpg$',
  48                 'description': 'Unika bilder avslöjar hur turisternas fickor vittjas mitt på Stockholms central. Två experter på ficktjuvarna avslöjar knepen du ska se upp för.',
  49                 'timestamp': int,
  50                 'upload_date': '20150130',
  51             },
  52         },
  53         {
  54             'url': 'http://www.tv4play.se/sport/3060959',
  55             'only_matching': True,
  56         },
  57         {
  58             'url': 'http://www.tv4play.se/film/2378136',
  59             'only_matching': True,
  60         },
  61         {
  62             'url': 'http://www.tv4play.se/barn/looney-tunes?video_id=3062412',
  63             'only_matching': True,
  64         },
  65     ]
  66
  67     def _real_extract(self, url):
  68         video_id = self._match_id(url)
  69
  70         info = self._download_json(
  71             'http://www.tv4play.se/player/assets/%s.json' % video_id,
  72             video_id, 'Downloading video info JSON')
  73
  74         # If is_geo_restricted is true, it doesn't necessarily mean we can't download it
  75         if info.get('is_geo_restricted'):
  76             self.report_warning('This content might not be available in your country due to licensing restrictions.')
  77
  78         title = info['title']
  79
  80         subtitles = {}
  81         formats = []
  82         # http formats are linked with unresolvable host
  83         for kind in ('hls', ''):
  84             data = self._download_json(
  85                 'https://prima.tv4play.se/api/web/asset/%s/play.json' % video_id,
  86                 video_id, 'Downloading sources JSON', query={
  87                     'protocol': kind,
  88                     'videoFormat': 'MP4+WEBVTT',
  89                 })
  90             items = try_get(data, lambda x: x['playback']['items']['item'])
  91             if not items:
  92                 continue
  93             if isinstance(items, dict):
  94                 items = [items]
  95             for item in items:
  96                 manifest_url = item.get('url')
  97                 if not isinstance(manifest_url, compat_str):
  98                     continue
  99                 ext = determine_ext(manifest_url)
 100                 if ext == 'm3u8':
 101                     formats.extend(self._extract_m3u8_formats(
 102                         manifest_url, video_id, 'mp4', entry_protocol='m3u8_native',
 103                         m3u8_id=kind, fatal=False))
 104                 elif ext == 'f4m':
 105                     formats.extend(self._extract_akamai_formats(
 106                         manifest_url, video_id, {
 107                             'hls': 'tv4play-i.akamaihd.net',
 108                         }))
 109                 elif ext == 'webvtt':
 110                     subtitles = self._merge_subtitles(
 111                         subtitles, {
 112                             'sv': [{
 113                                 'url': manifest_url,
 114                                 'ext': 'vtt',
 115                             }]})
 116         self._sort_formats(formats)
 117
 118         return {
 119             'id': video_id,
 120             'title': title,
 121             'formats': formats,
 122             'subtitles': subtitles,
 123             'description': info.get('description'),
 124             'timestamp': parse_iso8601(info.get('broadcast_date_time')),
 125             'duration': int_or_none(info.get('duration')),
 126             'thumbnail': info.get('image'),
 127             'is_live': info.get('is_live') is True,
 128         }