[rtlxl] Strip additional dot from video URL (#4115)
[youtube-dl] / youtube_dl / extractor / rtlnl.py
1 from __future__ import unicode_literals
2
3 import re
4
5 from .common import InfoExtractor
6 from ..utils import parse_duration
7
8
9 class RtlXlIE(InfoExtractor):
10     IE_NAME = 'rtlxl.nl'
11     _VALID_URL = r'https?://www\.rtlxl\.nl/#!/[^/]+/(?P<uuid>[^/?]+)'
12
13     _TEST = {
14         'url': 'http://www.rtlxl.nl/#!/rtl-nieuws-132237/6e4203a6-0a5e-3596-8424-c599a59e0677',
15         'md5': 'cc16baa36a6c169391f0764fa6b16654',
16         'info_dict': {
17             'id': '6e4203a6-0a5e-3596-8424-c599a59e0677',
18             'ext': 'mp4',
19             'title': 'RTL Nieuws - Laat',
20             'description': 'md5:6b61f66510c8889923b11f2778c72dc5',
21             'timestamp': 1408051800,
22             'upload_date': '20140814',
23             'duration': 576.880,
24         },
25     }
26
27     def _real_extract(self, url):
28         mobj = re.match(self._VALID_URL, url)
29         uuid = mobj.group('uuid')
30
31         info = self._download_json(
32             'http://www.rtl.nl/system/s4m/vfd/version=2/uuid=%s/fmt=flash/' % uuid,
33             uuid)
34
35         material = info['material'][0]
36         episode_info = info['episodes'][0]
37
38         progname = info['abstracts'][0]['name']
39         subtitle = material['title'] or info['episodes'][0]['name']
40
41         # Use unencrypted m3u8 streams (See https://github.com/rg3/youtube-dl/issues/4118)
42         videopath = material['videopath'].replace('.f4m', '.m3u8')
43         m3u8_url = 'http://manifest.us.rtl.nl' + videopath
44
45         formats = self._extract_m3u8_formats(m3u8_url, uuid, ext='mp4')
46
47         video_urlpart = videopath.split('/flash/')[1][:-5]
48         PG_URL_TEMPLATE = 'http://pg.us.rtl.nl/rtlxl/network/%s/progressive/%s.mp4'
49
50         formats.extend([
51             {
52                 'url': PG_URL_TEMPLATE % ('a2m', video_urlpart),
53                 'format_id': 'pg-sd',
54             },
55             {
56                 'url': PG_URL_TEMPLATE % ('a3m', video_urlpart),
57                 'format_id': 'pg-hd',
58                 'quality': 0,
59             }
60         ])
61
62         self._sort_formats(formats)
63
64         return {
65             'id': uuid,
66             'title': '%s - %s' % (progname, subtitle),
67             'formats': formats,
68             'timestamp': material['original_date'],
69             'description': episode_info['synopsis'],
70             'duration': parse_duration(material.get('duration')),
71         }