def getSynset_Ids(self,
lemma,
lang,
targetLang=None,
pos=None,
source=None):
"""
get synsetids to request target url
lemma; word
lang: language
targetLang: target language that is often the same as lang; howerver rarely is not same.
pos: part of speech
source: wikipedia and so on
return synsetsids
"""
self.lang = lang
synset_url = self.make_url(function="getSynsetIds?",
lemma=lemma,
lang=self.lang,
pos=pos,
source=source)
synset_url = synset_url.format(lemma, self.lang, self.key)
synset_json = urllib.request.urlopen(synset_url).read()
synset_json = synset_json.decode(self.unicode, self.unicode_option)
synset_ids = [dict2obj(js) for js in json.loads(synset_json)]
self.synset_ids = synset_ids
return synset_ids
def getSenses(self,
lemma,
lang,
lemma_type="full",
force_compare=True,
useURL=False):
"""
get senses from synsets
lang: language
pos: part of speech
lemma_type: full, simple
force_compare: A == a. a big character is the same as a small character.
useURL: if it is True, it search from url. Unless, search from data has been already got
return: Senses
"""
if useURL:
lang = lang or self.lang
synset_url = self.make_url(function="getSenses?",
lemma=lemma,
lang=lang)
synset_json = urllib.request.urlopen(synset_url).read()
synset_json = synset_json.decode(self.unicode, self.unicode_option)
senses = [dict2obj(js) for js in json.loads(synset_json)]
else:
senses = []
if "full" == lemma_type:
for synset in self.synsets:
for sense in synset.senses:
if force_compare and \
sense.properties.fullLemma.lower() == lemma.lower() and \
sense.properties.language.lower() == lang.lower():
senses.append(sense)
elif sense.properties.fullLemma == lemma and \
sense.properties.language == lang:
senses.append(sense)
elif "simple" == lemma_type:
for synset in self.synsets:
for sense in synset.senses:
if force_compare and \
sense.properties.fullLemma.lower() == lemma.lower() and \
sense.properties.language.lower() == lang.lower():
senses.append(sense)
elif sense.properties.simpleLemma == lemma and \
sense.properties.language == lang:
senses.append(sense)
self.senses = senses
return senses
def getOutgoingEdges(self, synset_id):
"""
get edges that has relationships against other synsets to request target url
synset_id: a babelnet synsetid
return edges
"""
synset_url = self.make_url(function="getOutgoingEdges?", id=synset_id)
synset_url = synset_url.format(synset_id, self.key)
synset_json = urllib.request.urlopen(synset_url).read()
synset_json = synset_json.decode(self.unicode, self.unicode_option)
edges = [dict2obj(js) for js in json.loads(synset_json)]
self.edges = edges
return edges
def getWordNetId(self, synset_id):
"""
For your internal server only,
get WordNet id which is equal to BabelNet id
synset_id: a babelnet synsetid
return wordnet synsetid
"""
assert ("internal server only", self.key is None)
synset_url = self.make_url(function="getWordnetId?", id=synset_id)
synset_url = synset_url.format(synset_id, self.key)
synset_json = urllib.request.urlopen(synset_url).read()
synset_json = synset_json.decode(self.unicode, self.unicode_option)
wordnet_ids = [dict2obj(js) for js in json.loads(synset_json)]
return wordnet_ids
def getSynsets(self, synset_id, targetLang=None, change_lang=False):
"""
get synsets to request target url
synset_id: a babelnet synsetid
targetLang: target language that is often the same as lang; howerver rarely is not same.
change_lang: for translation or sub language
return synsets
"""
if change_lang and targetLang != None:
if type(targetLang) == list:
self.lang = targetLang[0]
else:
self.lang = targetLang
targetLang = targetLang or self.lang
synset_url = self.make_url(function="getSynset?",
id=synset_id,
targetLang=targetLang)
synset_json = urllib.request.urlopen(synset_url).read()
synset_json = synset_json.decode(self.unicode, self.unicode_option)
synsets = [dict2obj(json.loads(synset_json))]
self.synsets = synsets
return synsets
