0x2620/python-ox
2
1
Fork 0
Code Issues Pull requests Releases Wiki Activity

fix only one connection, parse connection description too

Browse source
This commit is contained in:
j 2012-09-29 18:13:58 +02:00
parent da09714910
commit e0dd4d53b1
1 changed files with 8 additions and 4 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

12
ox/web/imdb.py
View file

@ -381,7 +381,7 @@ class Imdb(SiteParser):
if 'connections' in self: if 'connections' in self:
cc={} cc={}
if len(self['connections']) == 2 and isinstance(self['connections'][0], basestring): if len(self['connections']) == 3 and isinstance(self['connections'][0], basestring):
self['connections'] = [self['connections']] self['connections'] = [self['connections']]
for rel, data, _ in self['connections']: for rel, data, _ in self['connections']:
#cc[unicode(rel)] = re.compile('<a href="/title/tt(\d{7})/">(.*?)</a>').findall(data) #cc[unicode(rel)] = re.compile('<a href="/title/tt(\d{7})/">(.*?)</a>').findall(data)
@ -389,11 +389,15 @@ class Imdb(SiteParser):
title = c[1] title = c[1]
if title.startswith('"') and title.endswith('"'): if title.startswith('"') and title.endswith('"'):
title = title[1:-1] title = title[1:-1]
return { r = {
'id': c[0], 'id': c[0],
'title': title 'title': title,
} }
cc[unicode(rel)] = map(get_conn, re.compile('<a href="/title/tt(\d{7})/">(.*?)</a>').findall(data)) description = c[2].split('<br />')
if len(description) == 2:
r['description'] = description[-1].strip()
return r
cc[unicode(rel)] = map(get_conn, re.compile('<a href="/title/tt(\d{7})/">(.*?)</a>(.*?)<\/div', re.DOTALL).findall(data))
self['connections'] = cc self['connections'] = cc