返回另一个对象中列总和的成员函数答案

【问题标题】：Member function to return sum of column in another object返回另一个对象中列总和的成员函数
【发布时间】：2016-02-01 21:47:39
【问题描述】：

我目前已经在oracle中实现了下面的代码。

我正在努力创建一个返回村庄居民数量的成员函数。

我将如何做这件事并最终创建一个匿名块进行测试，然后最终针对它运行一个 select 语句？

房屋对象

create or replace type house as object (
 house_no number,
 postcode varchar2(10),
 no_of_residents number,
 member function pass(h house) return number
);
/

房屋对象主体

create or replace type body houses_p12214840 as
member function pass(h houses_p12214840) return number is
begin
    if (h.house_no=house_no) and (h.road_name=road_name) 
        and (h.postcode=postcode)
        and (h.no_of_residents=no_of_residents) 
        then 
            return 1;
        else 
            return 0;
    end if;
end;
end;
/

村阵式，多户型

create or replace type village_array is varray(50) of house;
/

我的村桌

create table village (
 villageid number,
 village_name varchar2(25),
 village village_array,
 PRIMARY KEY (villageid)
);
/

【问题讨论】：

标签： arrays object plsql sqlplus

【解决方案1】：

不清楚您想要的结果。

如果您只需要每个村庄中所有居民的总和，则不需要成员函数（“bool”[1/0] 也不需要）。

在 Column-Inline-Query 中使用 SUM 函数：

SELECT T.VILLAGEID,
       T.VILLAGE_NAME,
       (SELECT SUM(T2.NO_OF_RESIDENTS)
        FROM   TABLE(T.HOUSES) T2) SUM_RESIDENTS
FROM   VILLAGE T;

我使用以下示例进行了一些更改：

SET SERVEROUTPUT ON;
SET FEEDBACK OFF;
CLEAR;


CREATE TYPE HOUSE AS OBJECT
(
  HOUSE_NO        NUMBER,
  ROAD_NAME       VARCHAR2(255),
  POSTCODE        VARCHAR2(10),
  NO_OF_RESIDENTS NUMBER,
  MEMBER FUNCTION PASS(H HOUSE) RETURN NUMBER
);
/

CREATE OR REPLACE TYPE BODY HOUSE AS
  MEMBER FUNCTION PASS(H HOUSE) RETURN NUMBER IS
  BEGIN
    IF (H.HOUSE_NO = HOUSE_NO) AND (H.ROAD_NAME = ROAD_NAME) AND (H.POSTCODE = POSTCODE) AND (H.NO_OF_RESIDENTS = NO_OF_RESIDENTS) THEN
      RETURN 1;
    ELSE
      RETURN 0;
    END IF;
  END;
END;
/


CREATE OR REPLACE TYPE HOUSES_ARRAY IS VARRAY(50) OF HOUSE;
/

CREATE TABLE VILLAGE (
 VILLAGEID    NUMBER,
 VILLAGE_NAME VARCHAR2(25),
 HOUSES      HOUSES_ARRAY,
 PRIMARY KEY (VILLAGEID)
);


INSERT INTO village(VILLAGEID,VILLAGE_NAME,HOUSES) VALUES (1,'Village_1',HOUSES_ARRAY(HOUSE(1,'Road_1',12000,10),HOUSE(2,'Road_1',12000,2)));
INSERT INTO village(VILLAGEID,VILLAGE_NAME,HOUSES) VALUES (2,'Village_2',HOUSES_ARRAY(HOUSE(1,'Road_2',12010,5),HOUSE(2,'Road_2',12010,20)));


SELECT T.VILLAGEID,
       T.VILLAGE_NAME,
       (SELECT SUM(T2.NO_OF_RESIDENTS)
        FROM   TABLE(T.HOUSES) T2) SUM_RESIDENTS
FROM   VILLAGE T;


ROLLBACK;

DROP TABLE VILLAGE;
DROP TYPE HOUSES_ARRAY;
DROP TYPE HOUSE;

输出：

 VILLAGEID VILLAGE_NAME              SUM_RESIDENTS
---------- ------------------------- -------------
         1 Village_1                            12
         2 Village_2                            25

SQL>

或

SELECT T.VILLAGEID,
       T.VILLAGE_NAME,
       SUM(T2.NO_OF_RESIDENTS) SUM_RESIDENTS
FROM   VILLAGE T,
       TABLE(T.HOUSES) T2
GROUP  BY T.VILLAGEID,
          T.VILLAGE_NAME
;

注意过多的上下文更改 (SQLPLSQL)。
如果你能用 SQL 做，那就用 SQL 做。

【讨论】：