]> Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/radiocanada.py
Fix extraction from youtube.
[youtubedl] / youtube_dl / extractor / radiocanada.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 xpath_text,
9 find_xpath_attr,
10 determine_ext,
11 int_or_none,
12 unified_strdate,
13 xpath_element,
14 ExtractorError,
15 determine_protocol,
16 unsmuggle_url,
17 )
18
19
20 class RadioCanadaIE(InfoExtractor):
21 IE_NAME = 'radiocanada'
22 _VALID_URL = r'(?:radiocanada:|https?://ici\.radio-canada\.ca/widgets/mediaconsole/)(?P<app_code>[^:/]+)[:/](?P<id>[0-9]+)'
23 _TESTS = [
24 {
25 'url': 'http://ici.radio-canada.ca/widgets/mediaconsole/medianet/7184272',
26 'info_dict': {
27 'id': '7184272',
28 'ext': 'mp4',
29 'title': 'Le parcours du tireur capté sur vidéo',
30 'description': 'Images des caméras de surveillance fournies par la GRC montrant le parcours du tireur d\'Ottawa',
31 'upload_date': '20141023',
32 },
33 'params': {
34 # m3u8 download
35 'skip_download': True,
36 }
37 },
38 {
39 # empty Title
40 'url': 'http://ici.radio-canada.ca/widgets/mediaconsole/medianet/7754998/',
41 'info_dict': {
42 'id': '7754998',
43 'ext': 'mp4',
44 'title': 'letelejournal22h',
45 'description': 'INTEGRALE WEB 22H-TJ',
46 'upload_date': '20170720',
47 },
48 'params': {
49 # m3u8 download
50 'skip_download': True,
51 },
52 }
53 ]
54
55 def _real_extract(self, url):
56 url, smuggled_data = unsmuggle_url(url, {})
57 app_code, video_id = re.match(self._VALID_URL, url).groups()
58
59 metadata = self._download_xml(
60 'http://api.radio-canada.ca/metaMedia/v1/index.ashx',
61 video_id, note='Downloading metadata XML', query={
62 'appCode': app_code,
63 'idMedia': video_id,
64 })
65
66 def get_meta(name):
67 el = find_xpath_attr(metadata, './/Meta', 'name', name)
68 return el.text if el is not None else None
69
70 if get_meta('protectionType'):
71 raise ExtractorError('This video is DRM protected.', expected=True)
72
73 device_types = ['ipad']
74 if not smuggled_data:
75 device_types.append('flash')
76 device_types.append('android')
77
78 formats = []
79 error = None
80 # TODO: extract f4m formats
81 # f4m formats can be extracted using flashhd device_type but they produce unplayable file
82 for device_type in device_types:
83 validation_url = 'http://api.radio-canada.ca/validationMedia/v1/Validation.ashx'
84 query = {
85 'appCode': app_code,
86 'idMedia': video_id,
87 'connectionType': 'broadband',
88 'multibitrate': 'true',
89 'deviceType': device_type,
90 }
91 if smuggled_data:
92 validation_url = 'https://services.radio-canada.ca/media/validation/v2/'
93 query.update(smuggled_data)
94 else:
95 query.update({
96 # paysJ391wsHjbOJwvCs26toz and bypasslock are used to bypass geo-restriction
97 'paysJ391wsHjbOJwvCs26toz': 'CA',
98 'bypasslock': 'NZt5K62gRqfc',
99 })
100 v_data = self._download_xml(validation_url, video_id, note='Downloading %s XML' % device_type, query=query, fatal=False)
101 v_url = xpath_text(v_data, 'url')
102 if not v_url:
103 continue
104 if v_url == 'null':
105 error = xpath_text(v_data, 'message')
106 continue
107 ext = determine_ext(v_url)
108 if ext == 'm3u8':
109 formats.extend(self._extract_m3u8_formats(
110 v_url, video_id, 'mp4', m3u8_id='hls', fatal=False))
111 elif ext == 'f4m':
112 formats.extend(self._extract_f4m_formats(
113 v_url, video_id, f4m_id='hds', fatal=False))
114 else:
115 ext = determine_ext(v_url)
116 bitrates = xpath_element(v_data, 'bitrates')
117 for url_e in bitrates.findall('url'):
118 tbr = int_or_none(url_e.get('bitrate'))
119 if not tbr:
120 continue
121 f_url = re.sub(r'\d+\.%s' % ext, '%d.%s' % (tbr, ext), v_url)
122 protocol = determine_protocol({'url': f_url})
123 f = {
124 'format_id': '%s-%d' % (protocol, tbr),
125 'url': f_url,
126 'ext': 'flv' if protocol == 'rtmp' else ext,
127 'protocol': protocol,
128 'width': int_or_none(url_e.get('width')),
129 'height': int_or_none(url_e.get('height')),
130 'tbr': tbr,
131 }
132 mobj = re.match(r'(?P<url>rtmp://[^/]+/[^/]+)/(?P<playpath>[^?]+)(?P<auth>\?.+)', f_url)
133 if mobj:
134 f.update({
135 'url': mobj.group('url') + mobj.group('auth'),
136 'play_path': mobj.group('playpath'),
137 })
138 formats.append(f)
139 if protocol == 'rtsp':
140 base_url = self._search_regex(
141 r'rtsp://([^?]+)', f_url, 'base url', default=None)
142 if base_url:
143 base_url = 'http://' + base_url
144 formats.extend(self._extract_m3u8_formats(
145 base_url + '/playlist.m3u8', video_id, 'mp4',
146 'm3u8_native', m3u8_id='hls', fatal=False))
147 formats.extend(self._extract_f4m_formats(
148 base_url + '/manifest.f4m', video_id,
149 f4m_id='hds', fatal=False))
150 if not formats and error:
151 raise ExtractorError(
152 '%s said: %s' % (self.IE_NAME, error), expected=True)
153 self._sort_formats(formats)
154
155 subtitles = {}
156 closed_caption_url = get_meta('closedCaption') or get_meta('closedCaptionHTML5')
157 if closed_caption_url:
158 subtitles['fr'] = [{
159 'url': closed_caption_url,
160 'ext': determine_ext(closed_caption_url, 'vtt'),
161 }]
162
163 return {
164 'id': video_id,
165 'title': get_meta('Title') or get_meta('AV-nomEmission'),
166 'description': get_meta('Description') or get_meta('ShortDescription'),
167 'thumbnail': get_meta('imageHR') or get_meta('imageMR') or get_meta('imageBR'),
168 'duration': int_or_none(get_meta('length')),
169 'series': get_meta('Emission'),
170 'season_number': int_or_none('SrcSaison'),
171 'episode_number': int_or_none('SrcEpisode'),
172 'upload_date': unified_strdate(get_meta('Date')),
173 'subtitles': subtitles,
174 'formats': formats,
175 }
176
177
178 class RadioCanadaAudioVideoIE(InfoExtractor):
179 'radiocanada:audiovideo'
180 _VALID_URL = r'https?://ici\.radio-canada\.ca/audio-video/media-(?P<id>[0-9]+)'
181 _TEST = {
182 'url': 'http://ici.radio-canada.ca/audio-video/media-7527184/barack-obama-au-vietnam',
183 'info_dict': {
184 'id': '7527184',
185 'ext': 'mp4',
186 'title': 'Barack Obama au Vietnam',
187 'description': 'Les États-Unis lèvent l\'embargo sur la vente d\'armes qui datait de la guerre du Vietnam',
188 'upload_date': '20160523',
189 },
190 'params': {
191 # m3u8 download
192 'skip_download': True,
193 },
194 }
195
196 def _real_extract(self, url):
197 return self.url_result('radiocanada:medianet:%s' % self._match_id(url))