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     _GEO_COUNTRIES = ['SE']
  28     _TESTS = [
  29         {
  30             'url': 'http://www.tv4.se/kalla-fakta/klipp/kalla-fakta-5-english-subtitles-2491650',
  31             'md5': 'cb837212f342d77cec06e6dad190e96d',
  32             'info_dict': {
  33                 'id': '2491650',
  34                 'ext': 'mp4',
  35                 'title': 'Kalla Fakta 5 (english subtitles)',
  36                 'thumbnail': r're:^https?://.*\.jpg$',
  37                 'timestamp': int,
  38                 'upload_date': '20131125',
  39             },
  40         },
  41         {
  42             'url': 'http://www.tv4play.se/iframe/video/3054113',
  43             'md5': 'cb837212f342d77cec06e6dad190e96d',
  44             'info_dict': {
  45                 'id': '3054113',
  46                 'ext': 'mp4',
  47                 'title': 'Så här jobbar ficktjuvarna - se avslöjande bilder',
  48                 'thumbnail': r're:^https?://.*\.jpg$',
  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             'url': 'http://www.tv4play.se/program/farang/3922081',
  68             'only_matching': True,
  69         }
  70     ]
  71
  72     def _real_extract(self, url):
  73         video_id = self._match_id(url)
  74
  75         info = self._download_json(
  76             'http://www.tv4play.se/player/assets/%s.json' % video_id,
  77             video_id, 'Downloading video info JSON')
  78
  79         title = info['title']
  80
  81         subtitles = {}
  82         formats = []
  83         # http formats are linked with unresolvable host
  84         for kind in ('hls3', ''):
  85             data = self._download_json(
  86                 'https://prima.tv4play.se/api/web/asset/%s/play.json' % video_id,
  87                 video_id, 'Downloading sources JSON', query={
  88                     'protocol': kind,
  89                     'videoFormat': 'MP4+WEBVTT',
  90                 })
  91             items = try_get(data, lambda x: x['playback']['items']['item'])
  92             if not items:
  93                 continue
  94             if isinstance(items, dict):
  95                 items = [items]
  96             for item in items:
  97                 manifest_url = item.get('url')
  98                 if not isinstance(manifest_url, compat_str):
  99                     continue
 100                 ext = determine_ext(manifest_url)
 101                 if ext == 'm3u8':
 102                     formats.extend(self._extract_m3u8_formats(
 103                         manifest_url, video_id, 'mp4', entry_protocol='m3u8_native',
 104                         m3u8_id=kind, fatal=False))
 105                 elif ext == 'f4m':
 106                     formats.extend(self._extract_akamai_formats(
 107                         manifest_url, video_id, {
 108                             'hls': 'tv4play-i.akamaihd.net',
 109                         }))
 110                 elif ext == 'webvtt':
 111                     subtitles = self._merge_subtitles(
 112                         subtitles, {
 113                             'sv': [{
 114                                 'url': manifest_url,
 115                                 'ext': 'vtt',
 116                             }]})
 117
 118         if not formats and info.get('is_geo_restricted'):
 119             self.raise_geo_restricted(countries=self._GEO_COUNTRIES)
 120
 121         self._sort_formats(formats)
 122
 123         return {
 124             'id': video_id,
 125             'title': title,
 126             'formats': formats,
 127             'subtitles': subtitles,
 128             'description': info.get('description'),
 129             'timestamp': parse_iso8601(info.get('broadcast_date_time')),
 130             'duration': int_or_none(info.get('duration')),
 131             'thumbnail': info.get('image'),
 132             'is_live': info.get('is_live') is True,
 133         }