Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/rtlnl.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 from .common import InfoExtractor
   5 from ..utils import (
   6     int_or_none,
   7     parse_duration,
   8 )
   9
  10
  11 class RtlNlIE(InfoExtractor):
  12     IE_NAME = 'rtl.nl'
  13     IE_DESC = 'rtl.nl and rtlxl.nl'
  14     _VALID_URL = r'''(?x)
  15         https?://(www\.)?
  16         (?:
  17             rtlxl\.nl/\#!/[^/]+/|
  18             rtl\.nl/system/videoplayer/[^?#]+?/video_embed\.html\#uuid=
  19         )
  20         (?P<id>[0-9a-f-]+)'''
  21
  22     _TESTS = [{
  23         'url': 'http://www.rtlxl.nl/#!/rtl-nieuws-132237/6e4203a6-0a5e-3596-8424-c599a59e0677',
  24         'md5': 'cc16baa36a6c169391f0764fa6b16654',
  25         'info_dict': {
  26             'id': '6e4203a6-0a5e-3596-8424-c599a59e0677',
  27             'ext': 'mp4',
  28             'title': 'RTL Nieuws - Laat',
  29             'description': 'md5:6b61f66510c8889923b11f2778c72dc5',
  30             'timestamp': 1408051800,
  31             'upload_date': '20140814',
  32             'duration': 576.880,
  33         },
  34     }, {
  35         'url': 'http://www.rtl.nl/system/videoplayer/derden/rtlnieuws/video_embed.html#uuid=84ae5571-ac25-4225-ae0c-ef8d9efb2aed/autoplay=false',
  36         'md5': 'dea7474214af1271d91ef332fb8be7ea',
  37         'info_dict': {
  38             'id': '84ae5571-ac25-4225-ae0c-ef8d9efb2aed',
  39             'ext': 'mp4',
  40             'timestamp': 1424039400,
  41             'title': 'RTL Nieuws - Nieuwe beelden Kopenhagen: chaos direct na aanslag',
  42             'thumbnail': 're:^https?://screenshots\.rtl\.nl/system/thumb/sz=[0-9]+x[0-9]+/uuid=84ae5571-ac25-4225-ae0c-ef8d9efb2aed$',
  43             'upload_date': '20150215',
  44             'description': 'Er zijn nieuwe beelden vrijgegeven die vlak na de aanslag in Kopenhagen zijn gemaakt. Op de video is goed te zien hoe omstanders zich bekommeren om één van de slachtoffers, terwijl de eerste agenten ter plaatse komen.',
  45         }
  46     }]
  47
  48     def _real_extract(self, url):
  49         uuid = self._match_id(url)
  50         info = self._download_json(
  51             'http://www.rtl.nl/system/s4m/vfd/version=2/uuid=%s/fmt=flash/' % uuid,
  52             uuid)
  53
  54         material = info['material'][0]
  55         progname = info['abstracts'][0]['name']
  56         subtitle = material['title'] or info['episodes'][0]['name']
  57         description = material.get('synopsis') or info['episodes'][0]['synopsis']
  58
  59         # Use unencrypted m3u8 streams (See https://github.com/rg3/youtube-dl/issues/4118)
  60         videopath = material['videopath'].replace('.f4m', '.m3u8')
  61         m3u8_url = 'http://manifest.us.rtl.nl' + videopath
  62
  63         formats = self._extract_m3u8_formats(m3u8_url, uuid, ext='mp4')
  64
  65         video_urlpart = videopath.split('/flash/')[1][:-5]
  66         PG_URL_TEMPLATE = 'http://pg.us.rtl.nl/rtlxl/network/%s/progressive/%s.mp4'
  67
  68         formats.extend([
  69             {
  70                 'url': PG_URL_TEMPLATE % ('a2m', video_urlpart),
  71                 'format_id': 'pg-sd',
  72             },
  73             {
  74                 'url': PG_URL_TEMPLATE % ('a3m', video_urlpart),
  75                 'format_id': 'pg-hd',
  76                 'quality': 0,
  77             }
  78         ])
  79         self._sort_formats(formats)
  80
  81         thumbnails = []
  82         meta = info.get('meta', {})
  83         for p in ('poster_base_url', '"thumb_base_url"'):
  84             if not meta.get(p):
  85                 continue
  86
  87             thumbnails.append({
  88                 'url': self._proto_relative_url(meta[p] + uuid),
  89                 'width': int_or_none(self._search_regex(
  90                     r'/sz=([0-9]+)', meta[p], 'thumbnail width', fatal=False)),
  91                 'height': int_or_none(self._search_regex(
  92                     r'/sz=[0-9]+x([0-9]+)',
  93                     meta[p], 'thumbnail height', fatal=False))
  94             })
  95
  96         return {
  97             'id': uuid,
  98             'title': '%s - %s' % (progname, subtitle),
  99             'formats': formats,
 100             'timestamp': material['original_date'],
 101             'description': description,
 102             'duration': parse_duration(material.get('duration')),
 103             'thumbnails': thumbnails,
 104         }