[youtube] Fix extraction.
[youtube-dl] / youtube_dl / extractor / filmweb.py
1 from __future__ import unicode_literals
2
3 import re
4
5 from .common import InfoExtractor
6
7
8 class FilmwebIE(InfoExtractor):
9     _VALID_URL = r'https?://(?:www\.)?filmweb\.no/(?P<type>trailere|filmnytt)/article(?P<id>\d+)\.ece'
10     _TEST = {
11         'url': 'http://www.filmweb.no/trailere/article1264921.ece',
12         'md5': 'e353f47df98e557d67edaceda9dece89',
13         'info_dict': {
14             'id': '13033574',
15             'ext': 'mp4',
16             'title': 'Det som en gang var',
17             'upload_date': '20160316',
18             'timestamp': 1458140101,
19             'uploader_id': '12639966',
20             'uploader': 'Live Roaldset',
21         }
22     }
23
24     def _real_extract(self, url):
25         article_type, article_id = re.match(self._VALID_URL, url).groups()
26         if article_type == 'filmnytt':
27             webpage = self._download_webpage(url, article_id)
28             article_id = self._search_regex(r'data-videoid="(\d+)"', webpage, 'article id')
29         embed_code = self._download_json(
30             'https://www.filmweb.no/template_v2/ajax/json_trailerEmbed.jsp',
31             article_id, query={
32                 'articleId': article_id,
33             })['embedCode']
34         iframe_url = self._proto_relative_url(self._search_regex(
35             r'<iframe[^>]+src="([^"]+)', embed_code, 'iframe url'))
36
37         return {
38             '_type': 'url_transparent',
39             'id': article_id,
40             'url': iframe_url,
41             'ie_key': 'TwentyThreeVideo',
42         }