corpus.py 文件源码

def is_part_of_speech(self, pos):
        if hasattr(self.resource, QUERY_ITEM_POS):
            current_token = tokens.COCAToken(pos, replace=False)
            rc_feature = getattr(self.resource, QUERY_ITEM_POS)
            _, table, _ = self.resource.split_resource_feature(rc_feature)
            S = "SELECT {} FROM {} WHERE {} {} '{}' LIMIT 1".format(
                getattr(self.resource, "{}_id".format(table)),
                getattr(self.resource, "{}_table".format(table)),
                getattr(self.resource, rc_feature),
                self.resource.get_operator(current_token),
                pos)
            engine = self.resource.get_engine()
            df = pd.read_sql(S.replace("%", "%%"), engine)
            engine.dispose()
            return len(df.index) > 0
        else:
            return False