]> gitweb @ CieloNegro.org - youtube-dl.git/blob - youtube_dl/extractor/sbs.py
[ign] Fix extraction of video in articles
[youtube-dl.git] / youtube_dl / extractor / sbs.py
1 # -*- coding: utf-8 -*-
2 from __future__ import unicode_literals
3
4 import json
5 import re
6 from .common import InfoExtractor
7 from ..utils import (
8     js_to_json,
9     remove_end,
10 )
11
12
13 class SBSIE(InfoExtractor):
14     IE_DESC = 'sbs.com.au'
15     _VALID_URL = r'https?://(?:www\.)?sbs\.com\.au/ondemand/video/single/(?P<id>[0-9]+)/'
16
17     _TESTS = [{
18         # Original URL is handled by the generic IE which finds the iframe:
19         # http://www.sbs.com.au/thefeed/blog/2014/08/21/dingo-conservation
20         'url': 'http://www.sbs.com.au/ondemand/video/single/320403011771/?source=drupal&vertical=thefeed',
21         'md5': '3150cf278965eeabb5b4cea1c963fe0a',
22         'info_dict': {
23             'id': '320403011771',
24             'ext': 'flv',
25             'title': 'Dingo Conservation',
26             'description': 'Dingoes are on the brink of extinction; most of the animals we think are dingoes are in fact crossbred with wild dogs. This family run a dingo conservation park to prevent their extinction',
27             'thumbnail': 're:http://.*\.jpg',
28         },
29         'add_ies': ['generic'],
30     }]
31
32     def _real_extract(self, url):
33         mobj = re.match(self._VALID_URL, url)
34         video_id = mobj.group('id')
35         webpage = self._download_webpage(url, video_id)
36
37         release_urls_json = js_to_json(self._search_regex(
38             r'(?s)playerParams\.releaseUrls\s*=\s*(\{.*?\n\});\n',
39             webpage, ''))
40         release_urls = json.loads(release_urls_json)
41         theplatform_url = (
42             release_urls.get('progressive') or release_urls.get('standard'))
43
44         title = remove_end(self._og_search_title(webpage), ' (The Feed)')
45         description = self._html_search_meta('description', webpage)
46         thumbnail = self._og_search_thumbnail(webpage)
47
48         return {
49             '_type': 'url_transparent',
50             'id': video_id,
51             'url': theplatform_url,
52
53             'title': title,
54             'description': description,
55             'thumbnail': thumbnail,
56         }