fix parsing of URLs

This commit is contained in:
Romain Bignon 2012-01-31 18:08:09 +01:00
commit ce91c52912
3 changed files with 6 additions and 4 deletions

View file

@ -47,6 +47,7 @@ class InaBackend(BaseBackend, ICapVideo):
return self.browser.iter_search_results(pattern)
def fill_video(self, video, fields):
return video
if fields != ['thumbnail']:
# if we don't want only the thumbnail, we probably want also every fields
with self.browser:

View file

@ -31,7 +31,7 @@ __all__ = ['InaBrowser']
class InaBrowser(BaseBrowser):
DOMAIN = 'ina.fr'
PAGES = {'http://boutique\.ina\.fr/video/.+\.html': BoutiqueVideoPage,
PAGES = {'http://boutique\.ina\.fr/(video|audio)/.+\.html': BoutiqueVideoPage,
'http://www\.ina\.fr/.+\.html': VideoPage,
'http://boutique\.ina\.fr/recherche/.+': SearchPage,
}

View file

@ -56,8 +56,9 @@ class BaseVideoPage(BasePage):
def get_url(self):
qs = parse_qs(self.document.getroot().cssselect('param[name="flashvars"]')[0].attrib['value'])
url = 'http://mp4.ina.fr/lecture/lire/id_notice/%s/token_notice/%s' % (qs['id_notice'][0], qs['token_notice'][0])
return url
s = self.browser.readurl('http://boutique.ina.fr/player/infovideo/id_notice/%s' % qs['id_notice'][0])
s = s[s.find('<Media>')+7:s.find('</Media>')]
return '%s/pkey/%s' % (s, qs['pkey'][0])
def parse_date_and_duration(self, text):
duration_regexp = re.compile('(.* - )?(.+) - ((.+)h)?((.+)min)?(.+)s')
@ -103,7 +104,7 @@ class VideoPage(BaseVideoPage):
class BoutiqueVideoPage(BaseVideoPage):
URL_REGEXP = re.compile('http://boutique.ina.fr/video/(.+).html')
URL_REGEXP = re.compile('http://boutique.ina.fr/(audio|video)/(.+).html')
def create_id(self, id):
return u'boutique.%s' % id