def predict(self, data):
try:
if "news" in data.lower() or "latest" in data.lower():
# News query
source, query = self._query_extractor.get_news_tokens(data)
response = (_ga() if "guardian" in source else
_nyt()).get_news(query)
if len(response) <= 0:
return {
"phrase": "Sorry, no relevant results were returned."
}, 500
i, done = 0, media_aggregator.shorten_news(response[0])
while (not done) and ((i + 1) < len(response)):
i += 1
done = shorten_news(response[i])
else:
# Knowledge query
done = get_gkg(
self._query_extractor.get_knowledge_tokens(data))
ret_val = {"urls": done}
if not done:
ret_val["phrase"] = "Sorry, no valid results were returned."
return ret_val, done
except:
return {"phrase": "Sorry, something unexpected happened."}, False
def predict(self, data):
s=_qe.QueryExtractor()
_classes, _datapropertiesValues, _dataProperties = InputsForQuery(data.lower())
try:
if _classes != "" and _datapropertiesValues == "" and _dataProperties == "":
print("Information Question")
print(_classes)
# Query Classes
runQuery = SparqlQueriesClasses()
Queryresponse = runQuery.search(_classes.capitalize())
response = TemplateGeneratorInformation(_classes, Queryresponse)
done = response
if len(response) <= 0:
response = "Sorry, no relevant results were returned."
i, done = 0, response
while (not done) and ((i + 1) < len(response)):
i += 1
done = response
elif _classes != "" and _datapropertiesValues != "" and _dataProperties == "":
print("List Question")
print(_datapropertiesValues)
print(_classes)
# Query Names
runQuery = SparqlQueriesNames()
Queryresponse = runQuery.search(_classes.capitalize(),_datapropertiesValues.capitalize())
response = TemplateGeneratorList(_classes, _datapropertiesValues, Queryresponse)
if len(response) <= 0:
response = "Sorry, no relevant results were returned."
i, done = 0, response
while (not done) and ((i + 1) < len(response)):
i += 1
done = response
elif _dataProperties != "" and _datapropertiesValues != "" and _classes == "":
print("Specific Question")
print(_datapropertiesValues)
print(_dataProperties)
print("museums")
runQuery = SparqlQueriesSpecific()
Queryresponse = runQuery.search("Museums", _dataProperties.capitalize(), _datapropertiesValues.capitalize())
response = TemplateGeneratorSpecific(_dataProperties, _datapropertiesValues, Queryresponse)
if len(response) <= 0:
response = "Sorry, no relevant results were returned."
i, done = 0, response
while (not done) and ((i + 1) < len(response)):
i += 1
done = response
else:
# Knowledge query
done = get_gkg(self._query_extractor.get_knowledge_tokens(data))
ret_val = {"urls": done}
if not done:
ret_val["phrase"] = "Sorry, no valid results were returned."
return ret_val, done
except Exception as e:
return {"phrase": "Sorry, something unexpected happened.", "original_exception": e.message}, False
def predict(self, data):
try:
if "news" in data.lower() or "latest" in data.lower():
# News query
source, query = self._query_extractor.get_news_tokens(data)
response = (_ga() if "guardian" in source else _nyt()).get_news(query)
if len(response) <= 0:
return {"phrase": "Sorry, no relevant results were returned."}, 500
i, done = 0, media_aggregator.shorten_news(response[0])
while (not done) and ((i + 1) < len(response)):
i += 1
done = shorten_news(response[i])
else:
# Knowledge query
done = get_gkg(self._query_extractor.get_knowledge_tokens(data))
ret_val = {"urls": done}
if not done:
ret_val["phrase"] = "Sorry, no valid results were returned."
return ret_val, done
except Exception, e:
return {"phrase": "Sorry, something unexpected happened.", "original_exception": e.message}, False