2 from __future__ import unicode_literals
6 from .common import InfoExtractor
7 from ..utils import float_or_none
10 class VRTIE(InfoExtractor):
11 _VALID_URL = r'https?://(?:deredactie|sporza|cobra(?:\.canvas)?)\.be/cm/(?:[^/]+/)+(?P<id>[^/]+)/*'
15 'url': 'http://deredactie.be/cm/vrtnieuws/videozone/programmas/journaal/EP_141025_JOL',
16 'md5': '4cebde1eb60a53782d4f3992cbd46ec8',
20 'title': 'Het journaal L - 25/10/14',
22 'timestamp': 1414271750.949,
23 'upload_date': '20141025',
29 'url': 'http://sporza.be/cm/sporza/videozone/programmas/extratime/EP_141020_Extra_time',
30 'md5': '11f53088da9bf8e7cfc42456697953ff',
34 'title': 'Bekijk Extra Time van 20 oktober',
35 'description': 'md5:83ac5415a4f1816c6a93f8138aef2426',
36 'timestamp': 1413835980.560,
37 'upload_date': '20141020',
43 'url': 'http://cobra.be/cm/cobra/videozone/rubriek/film-videozone/141022-mv-ellis-cafecorsari',
44 'md5': '78a2b060a5083c4f055449a72477409d',
48 'title': 'Bret Easton Ellis in Café Corsari',
49 'description': 'md5:f699986e823f32fd6036c1855a724ee9',
50 'timestamp': 1413967500.494,
51 'upload_date': '20141022',
56 'url': 'http://cobra.canvas.be/cm/cobra/videozone/rubriek/film-videozone/1.2377055',
57 'only_matching': True,
61 def _real_extract(self, url):
62 video_id = self._match_id(url)
64 webpage = self._download_webpage(url, video_id)
66 video_id = self._search_regex(
67 r'data-video-id="([^"]+)_[^"]+"', webpage, 'video id', fatal=False)
71 r'data-video-iphone-server="(?P<server>[^"]+)"\s+data-video-iphone-path="(?P<path>[^"]+)"',
74 formats.extend(self._extract_m3u8_formats(
75 '%s/%s' % (mobj.group('server'), mobj.group('path')),
77 mobj = re.search(r'data-video-src="(?P<src>[^"]+)"', webpage)
79 formats.extend(self._extract_f4m_formats(
80 '%s/manifest.f4m' % mobj.group('src'), video_id))
81 self._sort_formats(formats)
83 title = self._og_search_title(webpage)
84 description = self._og_search_description(webpage, default=None)
85 thumbnail = self._og_search_thumbnail(webpage)
86 timestamp = float_or_none(self._search_regex(
87 r'data-video-sitestat-pubdate="(\d+)"', webpage, 'timestamp', fatal=False), 1000)
88 duration = float_or_none(self._search_regex(
89 r'data-video-duration="(\d+)"', webpage, 'duration', fatal=False), 1000)
94 'description': description,
95 'thumbnail': thumbnail,
96 'timestamp': timestamp,