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

titles

Browse source
This commit is contained in:
j 2012-11-08 20:34:26 +01:00
parent 91ebecccc3
commit c7a2ef21c7
1 changed files with 12 additions and 8 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

20
ox/web/imdb.py
View file

@ -301,7 +301,9 @@ class Imdb(SiteParser):
'working title', 'working title',
'reissue title', 'reissue title',
'IMAX version', 'IMAX version',
'alternative spelling' 'alternative spelling',
'informal short title',
'alternative transliteration'
): ):
if key in type: if key in type:
stop_word = True stop_word = True
@ -310,21 +312,23 @@ class Imdb(SiteParser):
types[type] = t[0] types[type] = t[0]
regexps = [ regexps = [
"^.+ \(imdb display title\) \(English title\)$", "^.+ \(imdb display title\) \(English title\)$",
"^International \(English title\)$",
"^International \(.+\) \(English title\)$",
"^.+ \(.+\) \(English title\)$",
"^USA$",
"^UK$",
"^USA \(imdb display title\)$", "^USA \(imdb display title\)$",
"^UK \(imdb display title\)$", "^UK \(imdb display title\)$",
"^International \(English title\)$",
"^International \(.+\) \(English title\)$",
"^USA$",
"^UK$",
] ]
if not filter(lambda c: c in ('USA', 'UK', 'Australia'), self.get('country', [])): if not filter(lambda c: c in ('USA', 'UK', 'Australia', 'New Zealand'), self.get('country', [])):
regexps.insert(2, "^.+ \(English title\)$") regexps.insert(5, "^[^(]+ \(English title\)$")
regexps.insert(6, "^.+ \(.+\) \(English title\)$")
regexps += [ regexps += [
"^USA \(.+\)$", "^USA \(.+\)$",
"^UK \(.+\)$", "^UK \(.+\)$",
"^Australia \(.+\)$", "^Australia \(.+\)$",
"(literal English title)",
"^International \(.+ title\)$", "^International \(.+ title\)$",
"^International \(.+\) \(.+ title\)$",
] ]
for regexp in regexps: for regexp in regexps:
for type in types: for type in types: