[ina] Simplify

Download the feed with ‘_download_xml’ to make the extraction easier
This commit is contained in:
Jaime Marquínez Ferrándiz 2014-02-04 23:15:04 +01:00
parent 50451f2a18
commit de563c9da0
1 changed files with 19 additions and 22 deletions

View File

@ -1,39 +1,36 @@
# encoding: utf-8
from __future__ import unicode_literals
import re
from .common import InfoExtractor
class InaIE(InfoExtractor):
"""Information Extractor for Ina.fr"""
_VALID_URL = r'(?:http://)?(?:www\.)?ina\.fr/video/(?P<id>I?[A-F0-9]+)/.*'
_VALID_URL = r'http://(?:www\.)?ina\.fr/video/(?P<id>I?[A-F0-9]+)/.*'
_TEST = {
u'url': u'http://www.ina.fr/video/I12055569/francois-hollande-je-crois-que-c-est-clair-video.html',
u'file': u'I12055569.mp4',
u'md5': u'a667021bf2b41f8dc6049479d9bb38a3',
u'info_dict': {
u"title": u"Fran\u00e7ois Hollande \"Je crois que c'est clair\""
'url': 'http://www.ina.fr/video/I12055569/francois-hollande-je-crois-que-c-est-clair-video.html',
'md5': 'a667021bf2b41f8dc6049479d9bb38a3',
'info_dict': {
'id': 'I12055569',
'ext': 'mp4',
'title': 'François Hollande "Je crois que c\'est clair"',
}
}
def _real_extract(self,url):
def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url)
video_id = mobj.group('id')
mrss_url='http://player.ina.fr/notices/%s.mrss' % video_id
video_extension = 'mp4'
webpage = self._download_webpage(mrss_url, video_id)
mrss_url = 'http://player.ina.fr/notices/%s.mrss' % video_id
info_doc = self._download_xml(mrss_url, video_id)
self.report_extraction(video_id)
video_url = self._html_search_regex(r'<media:player url="(?P<mp4url>http://mp4.ina.fr/[^"]+\.mp4)',
webpage, u'video URL')
video_url = info_doc.find('.//{http://search.yahoo.com/mrss/}player').attrib['url']
video_title = self._search_regex(r'<title><!\[CDATA\[(?P<titre>.*?)]]></title>',
webpage, u'title')
return [{
'id': video_id,
'url': video_url,
'ext': video_extension,
'title': video_title,
}]
return {
'id': video_id,
'url': video_url,
'title': info_doc.find('.//title').text,
}