# or POSITION('java.lang.StringBuilder' in AcceptedAnswer);''')
# return session.execute(
# '''select * from stackoverflow.body_html_text_clean_java
# where match(stackoverflow.body_html_text_clean_java.Title,
# stackoverflow.body_html_text_clean_java.Question,
# stackoverflow.body_html_text_clean_java.AcceptedAnswer) against('java.lang.StringBuilder');''')
except Exception:
traceback.print_exc()
return None
@staticmethod
def add_all_data(session, data):
session.add_all(data)
session.commit()
@staticmethod
def delete_by_id(session, Id):
session.delete(session.query(PostHtmlText).filter_by(Id=Id).first())
session.commit()
if __name__ == "__main__":
engine = EngineFactory.create_engine_by_schema_name("stackoverflow")
metadata = MetaData(bind=engine)
# delete all table
# Base.metadata.drop_all(bind=engine)
# create the table
Base.metadata.create_all(bind=engine)
sentence_id_list.append(each.sentence_id)
sentence_list.append(sentence_text)
paragraph_id_list.append(paragraph_id)
data_map.setdefault("merge_np_id", merge_np_id)
data_map.setdefault("noun_phrase", noun_phrase)
data_map.setdefault("sentence_id_list", sentence_id_list)
data_map.setdefault("sentence_list", sentence_list)
data_map.setdefault("paragraph_id_list", paragraph_id_list)
return data_map
def pack_all_to_json():
result = []
noun_phase_merge_clean_list = session.query(NounPhraseMergeClean).all()
if noun_phase_merge_clean_list is not None:
for noun_phase_merge_clean in noun_phase_merge_clean_list:
print "current np_id is ", noun_phase_merge_clean.id
print "current noun phrase is ", noun_phase_merge_clean.text
data_map = get_sentence_list_and_relation_id_by_merge_np_id(
noun_phase_merge_clean.id)
if data_map is not None and len(data_map) > 0:
result.append(data_map)
with open('np_data.json', 'w') as outfile:
json.dump(result, outfile)
if __name__ == "__main__":
engine = EngineFactory.create_engine_by_schema_name('codehub', echo=False)
session = EngineFactory.create_session(engine=engine, autocommit=False)
pack_all_to_json()
