TMDB: miglioria riconoscimento
This commit is contained in:
mac12m99
@@ -120,8 +120,8 @@ def peliculas(item):
|
||||
item.title += support.typo(item.lang2, '_ [] color kod')
|
||||
if item.args == 'novita':
|
||||
item.title = item.title
|
||||
if 'wp-content' in item.thumbnail and not item.infoLabels['year']:
|
||||
item.infoLabels['year'] = item.thumbnail.split('/')[5]
|
||||
# if 'wp-content' in item.thumbnail and not item.infoLabels['year']:
|
||||
# item.infoLabels['year'] = item.thumbnail.split('/')[5]
|
||||
return item
|
||||
return locals()
|
||||
|
||||
|
||||
@@ -478,3 +478,27 @@ def get_md5(cadena):
|
||||
devuelve = binascii.hexlify(md5.new(cadena).digest())
|
||||
|
||||
return devuelve
|
||||
|
||||
|
||||
def title_unify(title):
|
||||
import unicodedata
|
||||
|
||||
u_title = ''
|
||||
if type(title) == str: title = u'' + title
|
||||
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 == ':':
|
||||
c_new = c
|
||||
else:
|
||||
c_new = ' '
|
||||
u_title += c_new
|
||||
|
||||
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('-')
|
||||
u_title = spl[0] if len(spl[0]) > 5 else spl[1]
|
||||
|
||||
return u_title.strip()
|
||||
|
||||
50
core/tmdb.py
50
core/tmdb.py
@@ -411,51 +411,24 @@ def set_infoLabels_item(item, seekTmdb=True, idioma_busqueda=def_lang, lock=None
|
||||
__leer_datos(otmdb)
|
||||
return len(item.infoLabels)
|
||||
|
||||
# title might contain - or : --> try to search only second title
|
||||
def splitTitle():
|
||||
if '-' in item.fulltitle:
|
||||
item.infoLabels['tvshowtitle'] = item.fulltitle.split('-')[1]
|
||||
item.infoLabels['title'] = item.infoLabels['tvshowtitle']
|
||||
elif ':' in item.fulltitle:
|
||||
item.infoLabels['tvshowtitle'] = item.fulltitle.split(':')[1]
|
||||
item.infoLabels['title'] = item.infoLabels['tvshowtitle']
|
||||
else:
|
||||
return False
|
||||
return True
|
||||
def unify():
|
||||
new_title = scrapertools.title_unify(item.fulltitle)
|
||||
if new_title != item.fulltitle:
|
||||
item.infoLabels['tvshowtitle'] = scrapertools.title_unify(item.infoLabels['tvshowtitle'])
|
||||
item.infoLabels['title'] = scrapertools.title_unify(item.infoLabels['title'])
|
||||
item.fulltitle = new_title
|
||||
return True
|
||||
# We check what type of content it is...
|
||||
if item.contentType == 'movie':
|
||||
tipo_busqueda = 'movie'
|
||||
elif item.contentType == 'undefined': # don't know
|
||||
def detect():
|
||||
# try movie first
|
||||
results = search(otmdb_global, 'movie')
|
||||
if results:
|
||||
item.contentType = 'movie'
|
||||
infoMovie = item.infoLabels
|
||||
if infoMovie['title'] == item.fulltitle: # exact match -> it's probably correct
|
||||
return results
|
||||
|
||||
# try tvshow then
|
||||
item.infoLabels = {'tvshowtitle': item.infoLabels['tvshowtitle']} # reset infolabels
|
||||
results = search(otmdb_global, 'tv')
|
||||
if results:
|
||||
item.contentType = 'tvshow'
|
||||
else:
|
||||
item.infoLabels = infoMovie
|
||||
|
||||
return results
|
||||
|
||||
results = detect()
|
||||
if not results:
|
||||
if splitTitle():
|
||||
results = detect()
|
||||
return results
|
||||
tipo_busqueda = 'multi'
|
||||
else:
|
||||
tipo_busqueda = 'tv'
|
||||
|
||||
ret = search(otmdb_global, tipo_busqueda)
|
||||
if not ret:
|
||||
if splitTitle():
|
||||
if not ret: # try with unified title
|
||||
if unify():
|
||||
ret = search(otmdb_global, tipo_busqueda)
|
||||
return ret
|
||||
# Search in tmdb is deactivated or has not given result
|
||||
@@ -1028,7 +1001,7 @@ class Tmdb(object):
|
||||
# We sort result based on fuzzy match to detect most similar
|
||||
if len(results) > 1:
|
||||
from lib.fuzzy_match import algorithims
|
||||
results.sort(key=lambda r: algorithims.trigram(text_simple, r['title'] if self.busqueda_tipo == 'movie' else r['name']), reverse=True)
|
||||
results.sort(key=lambda r: algorithims.trigram(text_simple, r.get('name', '') if self.busqueda_tipo == 'tv' else r.get('title', '')), reverse=True)
|
||||
|
||||
# We return the number of results of this page
|
||||
self.results = results
|
||||
@@ -1043,7 +1016,6 @@ class Tmdb(object):
|
||||
logger.error(msg)
|
||||
return 0
|
||||
|
||||
|
||||
def __discover(self, index_results=0):
|
||||
self.result = ResultDictDefault()
|
||||
results = []
|
||||
|
||||
@@ -485,7 +485,7 @@ class SearchWindow(xbmcgui.WindowXML):
|
||||
if self.item.mode in ['all', 'search']:
|
||||
if self.item.type:
|
||||
self.item.mode = self.item.type
|
||||
self.item.text = title_unify(self.item.text)
|
||||
self.item.text = scrapertools.title_unify(self.item.text)
|
||||
self.thread = Thread(target=self.search)
|
||||
self.thread.start()
|
||||
elif self.item.mode in ['movie', 'tvshow', 'person_']:
|
||||
@@ -766,27 +766,3 @@ class SearchWindow(xbmcgui.WindowXML):
|
||||
server.window = True
|
||||
server.globalsearch = True
|
||||
return run(server)
|
||||
|
||||
|
||||
def title_unify(title):
|
||||
import unicodedata
|
||||
|
||||
u_title = ''
|
||||
if type(title) == str: title = u'' + title
|
||||
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 == ':':
|
||||
c_new = c
|
||||
else:
|
||||
c_new = ' '
|
||||
u_title += c_new
|
||||
|
||||
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('-')
|
||||
u_title = spl[0] if len(spl[0]) > 5 else spl[1]
|
||||
|
||||
return u_title.strip()
|
||||
|
||||
Reference in New Issue
Block a user