1 # coding: utf-8
2 from __future__ import unicode_literals
3
4 import re
5
6 from .common import InfoExtractor
7 from ..utils import (
8 float_or_none,
9 )
10
11
12 class VRTIE(InfoExtractor):
13 _VALID_URL = r'https?://(?:deredactie|sporza|cobra(?:\.canvas)?)\.be/cm/(?:[^/]+/)+(?P<id>[^/]+)/*'
14 _TESTS = [
15 # deredactie.be
16 {
17 'url': 'http://deredactie.be/cm/vrtnieuws/videozone/programmas/journaal/EP_141025_JOL',
18 'md5': '4cebde1eb60a53782d4f3992cbd46ec8',
19 'info_dict': {
20 'id': '2129880',
21 'ext': 'flv',
22 'title': 'Het journaal L - 25/10/14',
23 'description': None,
24 'timestamp': 1414271750.949,
25 'upload_date': '20141025',
26 'duration': 929,
27 },
28 'skip': 'HTTP Error 404: Not Found',
29 },
30 # sporza.be
31 {
32 'url': 'http://sporza.be/cm/sporza/videozone/programmas/extratime/EP_141020_Extra_time',
33 'md5': '11f53088da9bf8e7cfc42456697953ff',
34 'info_dict': {
35 'id': '2124639',
36 'ext': 'flv',
37 'title': 'Bekijk Extra Time van 20 oktober',
38 'description': 'md5:83ac5415a4f1816c6a93f8138aef2426',
39 'timestamp': 1413835980.560,
40 'upload_date': '20141020',
41 'duration': 3238,
42 },
43 'skip': 'HTTP Error 404: Not Found',
44 },
45 # cobra.be
46 {
47 'url': 'http://cobra.be/cm/cobra/videozone/rubriek/film-videozone/141022-mv-ellis-cafecorsari',
48 'md5': '78a2b060a5083c4f055449a72477409d',
49 'info_dict': {
50 'id': '2126050',
51 'ext': 'flv',
52 'title': 'Bret Easton Ellis in Café Corsari',
53 'description': 'md5:f699986e823f32fd6036c1855a724ee9',
54 'timestamp': 1413967500.494,
55 'upload_date': '20141022',
56 'duration': 661,
57 },
58 'skip': 'HTTP Error 404: Not Found',
59 },
60 {
61 # YouTube video
62 'url': 'http://deredactie.be/cm/vrtnieuws/videozone/nieuws/cultuurenmedia/1.2622957',
63 'md5': 'b8b93da1df1cea6c8556255a796b7d61',
64 'info_dict': {
65 'id': 'Wji-BZ0oCwg',
66 'ext': 'mp4',
67 'title': 'ROGUE ONE: A STAR WARS STORY Official Teaser Trailer',
68 'description': 'md5:8e468944dce15567a786a67f74262583',
69 'uploader': 'Star Wars',
70 'uploader_id': 'starwars',
71 'upload_date': '20160407',
72 },
73 'add_ie': ['Youtube'],
74 },
75 {
76 'url': 'http://cobra.canvas.be/cm/cobra/videozone/rubriek/film-videozone/1.2377055',
77 'info_dict': {
78 'id': '2377055',
79 'ext': 'mp4',
80 'title': 'Cafe Derby',
81 'description': 'Lenny Van Wesemael debuteert met de langspeelfilm Café Derby. Een waar gebeurd maar ook verzonnen verhaal.',
82 'upload_date': '20150626',
83 'timestamp': 1435305240.769,
84 },
85 'params': {
86 # m3u8 download
87 'skip_download': True,
88 }
89 }
90 ]
91
92 def _real_extract(self, url):
93 video_id = self._match_id(url)
94
95 webpage = self._download_webpage(url, video_id)
96
97 video_id = self._search_regex(
98 r'data-video-id="([^"]+)_[^"]+"', webpage, 'video id', fatal=False)
99
100 src = self._search_regex(
101 r'data-video-src="([^"]+)"', webpage, 'video src', default=None)
102
103 video_type = self._search_regex(
104 r'data-video-type="([^"]+)"', webpage, 'video type', default=None)
105
106 if video_type == 'YouTubeVideo':
107 return self.url_result(src, 'Youtube')
108
109 formats = []
110
111 mobj = re.search(
112 r'data-video-iphone-server="(?P<server>[^"]+)"\s+data-video-iphone-path="(?P<path>[^"]+)"',
113 webpage)
114 if mobj:
115 formats.extend(self._extract_m3u8_formats(
116 '%s/%s' % (mobj.group('server'), mobj.group('path')),
117 video_id, 'mp4', m3u8_id='hls', fatal=False))
118
119 if src:
120 formats = self._extract_wowza_formats(src, video_id)
121 if 'data-video-geoblocking="true"' not in webpage:
122 for f in formats:
123 if f['url'].startswith('rtsp://'):
124 http_format = f.copy()
125 http_format.update({
126 'url': f['url'].replace('rtsp://', 'http://').replace('vod.', 'download.').replace('/_definst_/', '/').replace('mp4:', ''),
127 'format_id': f['format_id'].replace('rtsp', 'http'),
128 'protocol': 'http',
129 })
130 formats.append(http_format)
131
132 if not formats and 'data-video-geoblocking="true"' in webpage:
133 self.raise_geo_restricted('This video is only available in Belgium')
134
135 self._sort_formats(formats)
136
137 title = self._og_search_title(webpage)
138 description = self._og_search_description(webpage, default=None)
139 thumbnail = self._og_search_thumbnail(webpage)
140 timestamp = float_or_none(self._search_regex(
141 r'data-video-sitestat-pubdate="(\d+)"', webpage, 'timestamp', fatal=False), 1000)
142 duration = float_or_none(self._search_regex(
143 r'data-video-duration="(\d+)"', webpage, 'duration', fatal=False), 1000)
144
145 return {
146 'id': video_id,
147 'title': title,
148 'description': description,
149 'thumbnail': thumbnail,
150 'timestamp': timestamp,
151 'duration': duration,
152 'formats': formats,
153 }
|
