SQL进阶系列之9用SQL处理数列

写在前面

关系模型的数据结构里，并没有顺序的概念，但SQL处理有序集合也有坚实的理论基础

生成连续编号

--生成连续编号
CREATE TABLE Digits
 (digit INTEGER PRIMARY KEY); 

INSERT INTO Digits VALUES (0);
INSERT INTO Digits VALUES (1);
INSERT INTO Digits VALUES (2);
INSERT INTO Digits VALUES (3);
INSERT INTO Digits VALUES (4);
INSERT INTO Digits VALUES (5);
INSERT INTO Digits VALUES (6);
INSERT INTO Digits VALUES (7);
INSERT INTO Digits VALUES (8);
INSERT INTO Digits VALUES (9);

-- 创建00-99的序列
SELECT (D1.digit + D2.digit * 10) AS seq
FROM Digits AS D1 CROSS JOIN Digits AS D2
ORDER BY seq;

-- 创建1到542的序列
SELECT (D1.digit + D2.digit * 10 + D3.digit * 100) AS seq
FROM Digits AS D1,Digits AS D2,Digits AS D3
WHERE (D1.digit + D2.digit * 10 + D3.digit * 100) BETWEEN 1 AND 542
ORDER BY seq;

CREATE VIEW Sequence (seq) AS
SELECT D1.digit + (D2.digit * 10 ) + (D3.digit * 100)
FROM Digits D1,Digits D2,Digits D3;

SELECT seq FROM Sequence WHERE seq BETWEEN 1 AND 100 ORDER BY seq;

求全部的缺失编号

-- EXCEPT版 
SELECT seq FROM Sequence WHERE seq BETWEEN 1 AND 12
EXCEPT (SELECT * FROM SeqTbl);
--  NOT IN版
SELECT seq FROM Sequence WHERE seq BETWEEN 1 AND 12 AND seq
NOT IN (SELECT * FROM SeqTbl);

-- 动态地指定连续编号范围的SQL语句
SELECT seq FROM Sequence WHERE seq BETWEEN (SELECT MIN(seq) FROM SeqTbl) AND (SELECT MAX(seq) FROM SeqTbl) EXCEPT (SELECT * FROM SeqTbl);

三个人能坐得下吗？

--三个人能坐得下吗？
CREATE TABLE Seats
 ( seat   INTEGER NOT NULL  PRIMARY KEY,
   status CHAR(6) NOT NULL
     CHECK (status IN ('未预订', '已预订')) ); 

INSERT INTO Seats VALUES (1,  '已预订');
INSERT INTO Seats VALUES (2,  '已预订');
INSERT INTO Seats VALUES (3,  '未预订');
INSERT INTO Seats VALUES (4,  '未预订');
INSERT INTO Seats VALUES (5,  '未预订');
INSERT INTO Seats VALUES (6,  '已预订');
INSERT INTO Seats VALUES (7,  '未预订');
INSERT INTO Seats VALUES (8,  '未预订');
INSERT INTO Seats VALUES (9,  '未预订');
INSERT INTO Seats VALUES (10,  '未预订');
INSERT INTO Seats VALUES (11,  '未预订');
INSERT INTO Seats VALUES (12,  '已预订');
INSERT INTO Seats VALUES (13,  '已预订');
INSERT INTO Seats VALUES (14,  '未预订');
INSERT INTO Seats VALUES (15,  '未预订');

-- 找出需要的空位(1):不考虑座位的换排
SELECT S1.seat AS start_seat,'~',S2.seat AS end_seat
FROM Seats S1,Seats S2
WHERE S2.seat = S1.seat + 2 AND NOT EXISTS 
(SELECT * FROM Seats S3 WHERE S3.seat BETWEEN S1.seat AND S2.seat AND S3.status <> '未预订');

--考虑座位的折返
CREATE TABLE Seats2
 ( seat   INTEGER NOT NULL  PRIMARY KEY,
   row_id CHAR(1) NOT NULL,
   status CHAR(6) NOT NULL
     CHECK (status IN ('未预订', '已预订')) ); 

INSERT INTO Seats2 VALUES (1, 'A', '已预订');
INSERT INTO Seats2 VALUES (2, 'A', '已预订');
INSERT INTO Seats2 VALUES (3, 'A', '未预订');
INSERT INTO Seats2 VALUES (4, 'A', '未预订');
INSERT INTO Seats2 VALUES (5, 'A', '未预订');
INSERT INTO Seats2 VALUES (6, 'B', '已预订');
INSERT INTO Seats2 VALUES (7, 'B', '已预订');
INSERT INTO Seats2 VALUES (8, 'B', '未预订');
INSERT INTO Seats2 VALUES (9, 'B', '未预订');
INSERT INTO Seats2 VALUES (10,'B', '未预订');
INSERT INTO Seats2 VALUES (11,'C', '未预订');
INSERT INTO Seats2 VALUES (12,'C', '未预订');
INSERT INTO Seats2 VALUES (13,'C', '未预订');
INSERT INTO Seats2 VALUES (14,'C', '已预订');
INSERT INTO Seats2 VALUES (15,'C', '未预订');

-- 找出需要的空位(2):考虑座位的换排
SELECT S1.seat AS start_seat,'~',S2.seat AS end_seat
FROM Seats2 S1,Seats2 S2
WHERE S2.seat = S1.seat + 2 AND NOT EXISTS 
(SELECT * FROM Seats2 S3 WHERE S3.seat BETWEEN S1.seat AND S2.seat AND (S3.status <> '未预订' OR S3.row_id <> S1.row_id));

最多能坐下多少人

--最多能坐下多少人？
CREATE TABLE Seats3
 ( seat   INTEGER NOT NULL  PRIMARY KEY,
   status CHAR(6) NOT NULL
     CHECK (status IN ('未预订', '已预订')) ); 

INSERT INTO Seats3 VALUES (1,  '已预订');
INSERT INTO Seats3 VALUES (2,  '未预订');
INSERT INTO Seats3 VALUES (3,  '未预订');
INSERT INTO Seats3 VALUES (4,  '未预订');
INSERT INTO Seats3 VALUES (5,  '未预订');
INSERT INTO Seats3 VALUES (6,  '已预订');
INSERT INTO Seats3 VALUES (7,  '未预订');
INSERT INTO Seats3 VALUES (8,  '已预订');
INSERT INTO Seats3 VALUES (9,  '未预订');
INSERT INTO Seats3 VALUES (10, '未预订');

-- 先生成存储了所有序列的视图
CREATE VIEW Sequences(start_seat,end_seat,seat_cnt) AS 
SELECT S1.seat AS start_seat,S2.seat AS end_seat,S2.seat-S1.seat + 1 AS seat_cnt
FROM Seats3 S1,Seats3 S2
WHERE S1.seat < S2.seat
AND NOT EXISTS (SELECT * FROM Seats3 S3 WHERE (S3.seat BETWEEN S1.seat AND S2.seat AND S3.status<> '未预订') OR (S3.seat = S2.seat + 1 AND S3.status = '未预订') OR (S3.seat = S1.seat - 1 AND S3.status = '未预订'));

-- 取出最长的序列
SELECT start_seat,end_seat,seat_cnt FROM 
Sequences WHERE seat_cnt = (SELECT MAX(seat_cnt) FROM Sequences);

单调递增和单调递减

--单调递增和单调递减
CREATE TABLE MyStock
 (deal_date  DATE PRIMARY KEY,
  price      INTEGER ); 

INSERT INTO MyStock VALUES ('2007-01-06', 1000);
INSERT INTO MyStock VALUES ('2007-01-08', 1050);
INSERT INTO MyStock VALUES ('2007-01-09', 1050);
INSERT INTO MyStock VALUES ('2007-01-12', 900);
INSERT INTO MyStock VALUES ('2007-01-13', 880);
INSERT INTO MyStock VALUES ('2007-01-14', 870);
INSERT INTO MyStock VALUES ('2007-01-16', 920);
INSERT INTO MyStock VALUES ('2007-01-17', 1000);

-- 生成起点到终点的组合
SELECT My1.deal_date,My2.deal_date
FROM MyStock AS My1,MyStock AS My2
WHERE My1.deal_date < My2.deal_date
AND NOT EXISTS
(SELECT * FROM MyStock AS My3,MyStock AS My4
WHERE My3.deal_date BETWEEN My1.deal_date AND My2.deal_date
AND My3.deal_date BETWEEN My1.deal_date AND My2.deal_date 
AND My3.deal_date < My4.deal_date
AND My3.price >= My4.price);

--排除掉子集，只取最长的时间区间
SELECT MIN(start_date) AS start_date,          /* 最大限度地向前延伸起点 */
       end_date
  FROM  (SELECT S1.deal_date AS start_date,
                MAX(S2.deal_date) AS end_date  /* 最大限度地向后延伸终点 */
           FROM MyStock S1, MyStock S2
          WHERE S1.deal_date < S2.deal_date
            AND NOT EXISTS
             (SELECT *
                FROM MyStock S3, MyStock S4
               WHERE S3.deal_date BETWEEN S1.deal_date AND S2.deal_date
                 AND S4.deal_date BETWEEN S1.deal_date AND S2.deal_date
                 AND S3.deal_date < S4.deal_date
                 AND S3.price >= S4.price)
         GROUP BY S1.deal_date) TMP
GROUP BY end_date
ORDER BY start_date;

小结

• SQL处理数据的方法有两种
• 第一种把数据看成忽略了顺序的集合
• 第二种把数据看成有序的集合，此时的基本方法如下：
  • 首先自连接生成起点到终点的集合
  • 其次在子查询中描述内部的各个元素之间必须满足的关系
• 要在SQL中表达全称量化时，需要将全称量化命题转化为存在量化命题的否定形式，并使用NOT EXISTS谓词。

练习题

/* 练习题1-9-1：求所有的缺失编号——NOT EXISTS和外连接 
   NOT EXISTS版  */
SELECT seq
  FROM Sequence N
 WHERE seq BETWEEN 1 AND 12
   AND NOT EXISTS
        (SELECT *
           FROM SeqTbl S
          WHERE N.seq = S.seq );

/* 练习题1-9-1：求所有的缺失编号——NOT EXISTS和外连接 
   NOT EXISTS版  */
SELECT N.seq
  FROM Sequence N LEFT OUTER JOIN SeqTbl S
    ON N.seq = S.seq
 WHERE N.seq BETWEEN 1 AND 12
   AND S.seq IS NULL;

/* 练习题1-9-2：求序列——面向集合的思想 */
SELECT S1.seat AS start_seat, '～' , S2.seat AS end_seat
  FROM Seats S1, Seats S2, Seats S3
 WHERE S2.seat = S1.seat + (:head_cnt -1)
   AND S3.seat BETWEEN S1.seat AND S2.seat
 GROUP BY S1.seat, S2.seat
HAVING COUNT(*) = SUM(CASE WHEN S3.status = '未预订' THEN 1 ELSE 0 END);

/* 坐位有换排时 */
SELECT S1.seat AS start_seat, ' ～ ' , S2.seat AS end_seat
  FROM Seats2 S1, Seats2 S2, Seats2 S3
 WHERE S2.seat = S1.seat + (:head_cnt -1)
   AND S3.seat BETWEEN S1.seat AND S2.seat
 GROUP BY S1.seat, S2.seat
HAVING COUNT(*) = SUM(CASE WHEN S3.status = '未预订'
                            AND S3.row_id = S1.row_id THEN 1 ELSE 0 END);

/* 练习题1-9-3：求所有的序列——面向集合的思想 */
SELECT S1.seat AS start_seat,
       S2.seat AS end_seat,
       S2.seat - S1.seat + 1 AS seat_cnt
  FROM Seats3 S1, Seats3 S2, Seats3 S3
 WHERE S1.seat <= S2.seat /* 第一步：生成起点和终点的组合 */
   AND S3.seat BETWEEN S1.seat - 1 AND S2.seat + 1
 GROUP BY S1.seat, S2.seat
HAVING COUNT(*) = SUM(CASE WHEN S3.seat BETWEEN S1.seat AND S2.seat
                            AND S3.status = '未预订' THEN 1 /* 条件1 */
                           WHEN S3.seat = S2.seat + 1 AND S3.status = '已预订' THEN 1 /* 条件2 */
                           WHEN S3.seat = S1.seat - 1 AND S3.status = '已预订' THEN 1 /* 条件3 */
                           ELSE 0 END);