def postprocess_data(self, data):
result = {}
for item in ('names refs', 'titles refs', 'characters refs'):
result[item] = {}
for k, v in data.get(item, []):
k = k.strip()
v = v.strip()
if not (k and v):
continue
if not v.endswith('/'): continue
imdbID = analyze_imdbid(v)
if item == 'names refs':
obj = Person(personID=imdbID,
name=k,
accessSystem=self._as,
modFunct=self._modFunct)
elif item == 'titles refs':
obj = Movie(movieID=imdbID,
title=k,
accessSystem=self._as,
modFunct=self._modFunct)
else:
obj = Character(characterID=imdbID,
name=k,
accessSystem=self._as,
modFunct=self._modFunct)
# XXX: companies aren't handled: are they ever found in text,
# as links to their page?
result[item][k] = obj
return result
def _add_ref(self, kind):
"""Add a reference entry to the names and titles dictionaries."""
if kind == 'tt':
if self._titleRefCID and self._titleCN:
if not self._titlesRefs.has_key(self._titleCN):
try:
movie = Movie(movieID=str(self._titleRefCID),
title=self._titleCN,
accessSystem=self._as,
modFunct=self._modFunct)
self._titlesRefs[self._titleCN] = movie
except IMDbParserError:
pass
self._titleRefCID = u''
self._titleCN = u''
self._inTTRef = 0
self._inLinkTTRef = 0
elif kind == 'nm' and self._nameRefCID and self._nameCN:
# XXX: 'Neo' and 'Keanu Reeves' are two separated
# entry in the dictionary. Check the ID value instead
# of the key?
if not self._namesRefs.has_key(self._nameCN):
try:
person = Person(name=self._nameCN,
personID=str(self._nameRefCID),
accessSystem=self._as,
modFunct=self._modFunct)
self._namesRefs[self._nameCN] = person
except IMDbParserError:
pass
self._nameRefCID = u''
self._nameCN = u''
self._inNMRef = 0
elif kind == 'ch' and self._characterRefCID and self._characterCN:
if not self._charactersRefs.has_key(self._characterCN):
try:
character = Character(name=self._characterCN,
characterID=str(
self._characterRefCID),
accessSystem='http')
self._charactersRefs[self._characterCN] = character
except IMDbParserError:
pass
self._characterRefCID = u''
self._characterCN = u''
self._inCHRef = 0