[rtbf] improve extraction(fixes #9267)

This commit is contained in:
remitamine 2016-04-21 22:52:49 +01:00
parent 9260cf1d97
commit 7c36ea7d54

View File

@ -4,12 +4,18 @@ from __future__ import unicode_literals
from .common import InfoExtractor from .common import InfoExtractor
from ..utils import ( from ..utils import (
int_or_none, int_or_none,
unescapeHTML, ExtractorError,
) )
class RTBFIE(InfoExtractor): class RTBFIE(InfoExtractor):
_VALID_URL = r'https?://(?:www\.)?rtbf\.be/(?:video/[^?]+\?.*\bid=|ouftivi/(?:[^/]+/)*[^?]+\?.*\bvideoId=)(?P<id>\d+)' _VALID_URL = r'''(?x)
https?://(?:www\.)?rtbf\.be/
(?:
video/[^?]+\?.*\bid=|
ouftivi/(?:[^/]+/)*[^?]+\?.*\bvideoId=|
auvio/[^/]+\?.*id=
)(?P<id>\d+)'''
_TESTS = [{ _TESTS = [{
'url': 'https://www.rtbf.be/video/detail_les-diables-au-coeur-episode-2?id=1921274', 'url': 'https://www.rtbf.be/video/detail_les-diables-au-coeur-episode-2?id=1921274',
'md5': '799f334ddf2c0a582ba80c44655be570', 'md5': '799f334ddf2c0a582ba80c44655be570',
@ -17,7 +23,11 @@ class RTBFIE(InfoExtractor):
'id': '1921274', 'id': '1921274',
'ext': 'mp4', 'ext': 'mp4',
'title': 'Les Diables au coeur (épisode 2)', 'title': 'Les Diables au coeur (épisode 2)',
'description': 'Football - Diables Rouges',
'duration': 3099, 'duration': 3099,
'upload_date': '20140425',
'timestamp': 1398456336,
'uploader': 'rtbfsport',
} }
}, { }, {
# geo restricted # geo restricted
@ -26,45 +36,63 @@ class RTBFIE(InfoExtractor):
}, { }, {
'url': 'http://www.rtbf.be/ouftivi/niouzz?videoId=2055858', 'url': 'http://www.rtbf.be/ouftivi/niouzz?videoId=2055858',
'only_matching': True, 'only_matching': True,
}, {
'url': 'http://www.rtbf.be/auvio/detail_jeudi-en-prime-siegfried-bracke?id=2102996',
'only_matching': True,
}] }]
_IMAGE_HOST = 'http://ds1.ds.static.rtbf.be'
_PROVIDERS = {
'YOUTUBE': 'Youtube',
'DAILYMOTION': 'Dailymotion',
'VIMEO': 'Vimeo',
}
_QUALITIES = [ _QUALITIES = [
('mobile', 'mobile'), ('mobile', 'SD'),
('web', 'SD'), ('web', 'MD'),
('url', 'MD'),
('high', 'HD'), ('high', 'HD'),
] ]
def _real_extract(self, url): def _real_extract(self, url):
video_id = self._match_id(url) video_id = self._match_id(url)
data = self._download_json(
'http://www.rtbf.be/api/media/video?method=getVideoDetail&args[]=%s' % video_id, video_id)
webpage = self._download_webpage( error = data.get('error')
'http://www.rtbf.be/video/embed?id=%s' % video_id, video_id) if error:
raise ExtractorError('%s said: %s' % (self.IE_NAME, error), expected=True)
data = self._parse_json( data = data['data']
unescapeHTML(self._search_regex(
r'data-media="([^"]+)"', webpage, 'data video')), provider = data.get('provider')
video_id) if provider in self._PROVIDERS:
return self.url_result(data['url'], self._PROVIDERS[provider])
if data.get('provider').lower() == 'youtube':
video_url = data.get('downloadUrl') or data.get('url')
return self.url_result(video_url, 'Youtube')
formats = [] formats = []
for key, format_id in self._QUALITIES: for key, format_id in self._QUALITIES:
format_url = data['sources'].get(key) format_url = data.get(key + 'Url')
if format_url: if format_url:
formats.append({ formats.append({
'format_id': format_id, 'format_id': format_id,
'url': format_url, 'url': format_url,
}) })
thumbnails = []
for thumbnail_id, thumbnail_url in data.get('thumbnail', {}).items():
if thumbnail_id != 'default':
thumbnails.append({
'url': self._IMAGE_HOST + thumbnail_url,
'id': thumbnail_id,
})
return { return {
'id': video_id, 'id': video_id,
'formats': formats, 'formats': formats,
'title': data['title'], 'title': data['title'],
'description': data.get('description') or data.get('subtitle'), 'description': data.get('description') or data.get('subtitle'),
'thumbnail': data.get('thumbnail'), 'thumbnails': thumbnails,
'duration': data.get('duration') or data.get('realDuration'), 'duration': data.get('duration') or data.get('realDuration'),
'timestamp': int_or_none(data.get('created')), 'timestamp': int_or_none(data.get('created')),
'view_count': int_or_none(data.get('viewCount')), 'view_count': int_or_none(data.get('viewCount')),
'uploader': data.get('channel'),
'tags': data.get('tags'),
} }