miglioria title_unify per ricerca

This commit is contained in:
Enrico
2021-02-18 21:45:38 +01:00
committed by mac12m99
parent 54b56e2556
commit e40bdc341a

View File

@@ -748,24 +748,20 @@ def title_unify(title):
import unicodedata
u_title = ''
sep_count = 0
for c in unicodedata.normalize('NFD', title if PY3 else title.decode('utf-8', 'ignore')):
for c in unicodedata.normalize('NFD', title):
cat = unicodedata.category(c)
if cat != 'Mn':
if cat == 'Pd':
c_new = '-'
elif cat in ['Ll', 'Lu'] or c in [':']:
elif cat in ['Ll', 'Lu'] or c == ':':
c_new = c
else:
c_new = ' '
if c_new in ['-', ':']:
sep_count += 1
u_title += c_new
if sep_count == 1: # subtitle, split but only if there's one, it might be part of title
if (u_title.count(':') + u_title.count('-')) == 1:
# subtitle, split but only if there's one, it might be part of title
spl = u_title.replace(':', '-').split('-')
if len(spl[0]) > 5:
u_title = spl[0]
else:
u_title = spl[1]
return u_title.strip()
u_title = spl[0] if len(spl[0]) > 5 else spl[1]
return u_title.strip()