]> Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/vrt.py
Merge pull request #1 from e7appew/python3
[youtubedl] / youtube_dl / extractor / vrt.py
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 determine_ext,
9 float_or_none,
10 )
11
12
13 class VRTIE(InfoExtractor):
14 _VALID_URL = r'https?://(?:deredactie|sporza|cobra(?:\.canvas)?)\.be/cm/(?:[^/]+/)+(?P<id>[^/]+)/*'
15 _TESTS = [
16 # deredactie.be
17 {
18 'url': 'http://deredactie.be/cm/vrtnieuws/videozone/programmas/journaal/EP_141025_JOL',
19 'md5': '4cebde1eb60a53782d4f3992cbd46ec8',
20 'info_dict': {
21 'id': '2129880',
22 'ext': 'flv',
23 'title': 'Het journaal L - 25/10/14',
24 'description': None,
25 'timestamp': 1414271750.949,
26 'upload_date': '20141025',
27 'duration': 929,
28 },
29 'skip': 'HTTP Error 404: Not Found',
30 },
31 # sporza.be
32 {
33 'url': 'http://sporza.be/cm/sporza/videozone/programmas/extratime/EP_141020_Extra_time',
34 'md5': '11f53088da9bf8e7cfc42456697953ff',
35 'info_dict': {
36 'id': '2124639',
37 'ext': 'flv',
38 'title': 'Bekijk Extra Time van 20 oktober',
39 'description': 'md5:83ac5415a4f1816c6a93f8138aef2426',
40 'timestamp': 1413835980.560,
41 'upload_date': '20141020',
42 'duration': 3238,
43 },
44 'skip': 'HTTP Error 404: Not Found',
45 },
46 # cobra.be
47 {
48 'url': 'http://cobra.be/cm/cobra/videozone/rubriek/film-videozone/141022-mv-ellis-cafecorsari',
49 'md5': '78a2b060a5083c4f055449a72477409d',
50 'info_dict': {
51 'id': '2126050',
52 'ext': 'flv',
53 'title': 'Bret Easton Ellis in Café Corsari',
54 'description': 'md5:f699986e823f32fd6036c1855a724ee9',
55 'timestamp': 1413967500.494,
56 'upload_date': '20141022',
57 'duration': 661,
58 },
59 'skip': 'HTTP Error 404: Not Found',
60 },
61 {
62 # YouTube video
63 'url': 'http://deredactie.be/cm/vrtnieuws/videozone/nieuws/cultuurenmedia/1.2622957',
64 'md5': 'b8b93da1df1cea6c8556255a796b7d61',
65 'info_dict': {
66 'id': 'Wji-BZ0oCwg',
67 'ext': 'mp4',
68 'title': 'ROGUE ONE: A STAR WARS STORY Official Teaser Trailer',
69 'description': 'md5:8e468944dce15567a786a67f74262583',
70 'uploader': 'Star Wars',
71 'uploader_id': 'starwars',
72 'upload_date': '20160407',
73 },
74 'add_ie': ['Youtube'],
75 },
76 {
77 'url': 'http://cobra.canvas.be/cm/cobra/videozone/rubriek/film-videozone/1.2377055',
78 'md5': '',
79 'info_dict': {
80 'id': '2377055',
81 'ext': 'mp4',
82 'title': 'Cafe Derby',
83 'description': 'Lenny Van Wesemael debuteert met de langspeelfilm Café Derby. Een waar gebeurd maar ook verzonnen verhaal.',
84 'upload_date': '20150626',
85 'timestamp': 1435305240.769,
86 },
87 'params': {
88 # m3u8 download
89 'skip_download': True,
90 }
91 }
92 ]
93
94 def _real_extract(self, url):
95 video_id = self._match_id(url)
96
97 webpage = self._download_webpage(url, video_id)
98
99 video_id = self._search_regex(
100 r'data-video-id="([^"]+)_[^"]+"', webpage, 'video id', fatal=False)
101
102 src = self._search_regex(
103 r'data-video-src="([^"]+)"', webpage, 'video src', default=None)
104
105 video_type = self._search_regex(
106 r'data-video-type="([^"]+)"', webpage, 'video type', default=None)
107
108 if video_type == 'YouTubeVideo':
109 return self.url_result(src, 'Youtube')
110
111 formats = []
112
113 mobj = re.search(
114 r'data-video-iphone-server="(?P<server>[^"]+)"\s+data-video-iphone-path="(?P<path>[^"]+)"',
115 webpage)
116 if mobj:
117 formats.extend(self._extract_m3u8_formats(
118 '%s/%s' % (mobj.group('server'), mobj.group('path')),
119 video_id, 'mp4', m3u8_id='hls', fatal=False))
120
121 if src:
122 if determine_ext(src) == 'm3u8':
123 formats.extend(self._extract_m3u8_formats(
124 src, video_id, 'mp4', entry_protocol='m3u8_native',
125 m3u8_id='hls', fatal=False))
126 formats.extend(self._extract_f4m_formats(
127 src.replace('playlist.m3u8', 'manifest.f4m'),
128 video_id, f4m_id='hds', fatal=False))
129 if 'data-video-geoblocking="true"' not in webpage:
130 rtmp_formats = self._extract_smil_formats(
131 src.replace('playlist.m3u8', 'jwplayer.smil'),
132 video_id, fatal=False)
133 formats.extend(rtmp_formats)
134 for rtmp_format in rtmp_formats:
135 rtmp_format_c = rtmp_format.copy()
136 rtmp_format_c['url'] = '%s/%s' % (rtmp_format['url'], rtmp_format['play_path'])
137 del rtmp_format_c['play_path']
138 del rtmp_format_c['ext']
139 http_format = rtmp_format_c.copy()
140 http_format.update({
141 'url': rtmp_format_c['url'].replace('rtmp://', 'http://').replace('vod.', 'download.').replace('/_definst_/', '/').replace('mp4:', ''),
142 'format_id': rtmp_format['format_id'].replace('rtmp', 'http'),
143 'protocol': 'http',
144 })
145 rtsp_format = rtmp_format_c.copy()
146 rtsp_format.update({
147 'url': rtsp_format['url'].replace('rtmp://', 'rtsp://'),
148 'format_id': rtmp_format['format_id'].replace('rtmp', 'rtsp'),
149 'protocol': 'rtsp',
150 })
151 formats.extend([http_format, rtsp_format])
152 else:
153 formats.extend(self._extract_f4m_formats(
154 '%s/manifest.f4m' % src, video_id, f4m_id='hds', fatal=False))
155
156 if not formats and 'data-video-geoblocking="true"' in webpage:
157 self.raise_geo_restricted('This video is only available in Belgium')
158
159 self._sort_formats(formats)
160
161 title = self._og_search_title(webpage)
162 description = self._og_search_description(webpage, default=None)
163 thumbnail = self._og_search_thumbnail(webpage)
164 timestamp = float_or_none(self._search_regex(
165 r'data-video-sitestat-pubdate="(\d+)"', webpage, 'timestamp', fatal=False), 1000)
166 duration = float_or_none(self._search_regex(
167 r'data-video-duration="(\d+)"', webpage, 'duration', fatal=False), 1000)
168
169 return {
170 'id': video_id,
171 'title': title,
172 'description': description,
173 'thumbnail': thumbnail,
174 'timestamp': timestamp,
175 'duration': duration,
176 'formats': formats,
177 }