[Sqlite] Sqlite的基本日常SQL操作语句汇总
原文地址：[Sqlite] Sqlite的基本日常SQL操作语句汇总 作者：mchdba
原博客地址：http://blog.itpub.net/26230597/abstract/1/
原作者：黄杉 (mchdba)
序言：
嵌入式数据库Sqlite的基本sql使用汇总，使用测试起来，与关系型数据库mysql在语法上有很多的相似之处，先准备测试数据：
CREATE TABLE COMPANY(ID INT NOT NULL, NAME VARCHAR(20),AGE INT,ADDRESS VARCHAR(20),SALARY DECIMAL(7,2));
INSERT INTO "COMPANY" VALUES(1,'Paul',32,'California',20000);
INSERT INTO "COMPANY" VALUES(2,'Allen',25,'Texas',15000);
INSERT INTO "COMPANY" VALUES(3,'Teddy',23,'Norway',20000);
INSERT INTO "COMPANY" VALUES(4,'Mark',25,'Rich-Mond',65000);
INSERT INTO "COMPANY" VALUES(5,'David',27,'Texas',85000);
INSERT INTO "COMPANY" VALUES(6,'Kim',22,'South-Hall',45000);
INSERT INTO "COMPANY" VALUES(7,'James',24,NULL,10000);
INSERT INTO "COMPANY" VALUES(8,'Xiaoteng',29,NULL,NULL);
1，分组统计排序
GROUP BY 进行分组统计数据，命令如下：
sqlite> SELECT NAME, SUM(SALARY) SALARY_SUM, COUNT(1) COUNT_NUM FROM COMPANY GROUP BY NAME;
ORDER BY 进行排序，命令如下：
sqlite> SELECT NAME, SUM(SALARY) SALARY_SUM, COUNT(1) COUNT_NUM FROM COMPANY GROUP BY NAME ORDER BY SALARY_SUM ASC;
HAVING 字句过滤数据记录，命令如下：
SELECT c.*,COUNT(1) COUNT_NUM FROM COMPANY c GROUP BY c.NAME HAVING (COUNT_NUM) > 1 ORDER BY COUNT_NUM ;
PS：在一个查询中，HAVING 子句必须放在 GROUP BY 子句之后，必须放在 ORDER BY 子句之前。下面是包含 HAVING 子句的 SELECT 语句的语法：
2，Limit分页统计语句
SQLite 的 LIMIT 子句用于限制由 SELECT 语句返回的数据数量。
第一页取值SQL： SELECT * FROM COMPANY ORDER BY ID LIMIT 0,3; 也可以从一个特定的偏移开始提取记录，从第四位开始提取 3 个记录，使用OFFSET关键字，SELECT * FROM COMPANY ORDER BY ID LIMIT 3 OFFSET 0；PS：首页从0开始取值。
第二页取值SQL： SELECT * FROM COMPANY ORDER BY ID LIMIT 3,3; 也可以从一个特定的偏移开始提取记录，从第四位开始提取 3 个记录，使用OFFSET关键字，SELECT * FROM COMPANY ORDER BY ID LIMIT 3 OFFSET 3；如下图所示：
3，Glob匹配字句
SQLite 的 GLOB 运算符是用来匹配通配符指定模式的文本值。如果搜索表达式与模式表达式匹配，GLOB 运算符将返回真（true），也就是 1。与 LIKE 运算符不同的是，GLOB 是大小写敏感的，对于下面的通配符，它遵循 UNIX 的语法。
星号 （*）
问号 （?）
星号（*）代表零个、一个或多个数字或字符。问号（?）代表一个单一的数字或字符。这些符号可以被组合使用。
下面一些实例演示了 带有 '*' 和 '?' 运算符的 GLOB 子句不同的地方：
下面是一个实例，它显示 COMPANY 表中 AGE 以 2 开头的所有记录，如下所示：
下面是一个实例，它显示 COMPANY 表中 ADDRESS 文本里包含一个连字符（-）的所有记录：
4，Distinct关键字过滤重复记录
SQLite 的 DISTINCT 关键字与 SELECT 语句一起使用，来消除所有重复的记录，并只获取唯一一次记录。
有可能出现一种情况，在一个表中有多个重复的记录。当提取这样的记录时，DISTINCT 关键字就显得特别有意义，它只获取唯一一次记录，而不是获取重复记录。
5，字符串连接操作
问题地址：http://bbs.csdn.net/topics/390886865
sqlite> CREATE TABLE t1(id int, name varchar(60));
sqlite> INSERT INTO "t1" VALUES(4,'[email protected]');
sqlite> select * from t1;
id name
---------- ----------
sqlite> update t1 set name=(id/2)||substr(name,instr(name,'@'),length(name)-instr(name,'@')+1) where id=4;
sqlite> select * from t1;
id name
---------- ----------
sqlite>
6，对Null值的处理
往表里面录入Null值
sqlite> INSERT INTO COMPANY(ID,NAME,AGE,ADDRESS,SALARY) VALUES(8,'Xiaoteng',29,NULL,18000);
sqlite>
修改某个字段为null值
sqlite> UPDATE COMPANY SET SALARY = NULL WHERE ID=8;
sqlite>
查询为null的记录
sqlite> SELECT * FROM COMPANY WHERE ADDRESS IS NULL;
ID NAME AGE ADDRESS SALARY
---------- ---------- ---------- ---------- ----------
7 James 24 10000
8 Xiaoteng 29
sqlite>
查询不为null的记录
sqlite> SELECT * FROM COMPANY WHERE ADDRESS IS NOT NULL;
ID NAME AGE ADDRESS SALARY
---------- ---------- ---------- ---------- ----------
1 Paul 32 California 20000
2 Allen 25 Texas 15000
3 Teddy 23 Norway 20000
4 Mark 25 Rich-Mond 65000
5 David 27 Texas 85000
6 Kim 22 South-Hall 45000
sqlite>
7，子查询
SELECt中的基本语法如下：
SELECT column_name [, column_name ]
FROM table1 [, table2 ]
WHERE column_name OPERATOR
(SELECT column_name [, column_name ]
FROM table1 [, table2 ]
[WHERE])
实例如下：
sqlite> SELECT * FROM COMPANY WHERE ID IN (SELECT ID FROM COMPANY WHERE SALARY > 45000);
INSERT语句中的子查询使用，基本语法：
INSERT INTO table_name [ (column1 [, column2 ]) ]
SELECT [ *|column1 [, column2 ]
FROM table1 [, table2 ]
[ WHERE VALUE OPERATOR ]
实例如下：
sqlite> INSERT INTO COMPANY_BKP
SELECT * FROM COMPANY
WHERE ID IN (SELECT ID
FROM COMPANY) ;
UPDATE语句中的子查询使用，基本语法如下：
UPDATE table
SET column_name = new_value
[ WHERE OPERATOR [ VALUE ]
(SELECT COLUMN_NAME
FROM TABLE_NAME)
[ WHERE) ]
实例如下：
sqlite> UPDATE COMPANY
SET SALARY = SALARY * 0.50
WHERE AGE IN (SELECT AGE FROM COMPANY_BKP
WHERE AGE >= 27 );
DELETE语句中的子查询使用，语法如下：
DELETE FROM TABLE_NAME
[ WHERE OPERATOR [ VALUE ]
(SELECT COLUMN_NAME
FROM TABLE_NAME)
[ WHERE) ]
实例如下：
sqlite> DELETE FROM COMPANY
WHERE AGE IN (SELECT AGE FROM COMPANY_BKP
WHERE AGE > 27 );
8，EXPLAIN分析
没有建立索引之前，分析都是表扫描：
sqlite> EXPLAIN SELECT * FROM COMPANY WHERE Salary < 20000;
addr opcode p1 p2 p3 p4 p5 comment
---------- ---------- ---------- ---------- ---------- ---------- ---------- ----------
0 Trace 0 0 0 00
1 Integer 20000 1 0 00
2 Goto 0 16 0 00
3 OpenRead 0 2 0 5 00
4 Rewind 0 14 0 00
5 Column 0 4 2 00
6 Ge 1 13 2 collseq(BI 6b
7 Column 0 0 4 00
8 Column 0 1 5 00
9 Column 0 2 6 00
10 Column 0 3 7 00
11 Column 0 4 8 00
12 ResultRow 4 5 0 00
13 Next 0 5 0 01
14 Close 0 0 0 00
15 Halt 0 0 0 00
16 Transactio 0 0 0 00
17 VerifyCook 0 1 0 00
18 TableLock 0 2 0 COMPANY 00
19 Goto 0 3 0 00
sqlite> EXPLAIN QUERY PLAN SELECT * FROM COMPANY WHERE Salary < 20000;
order from detail
---------- ---------- -------------
0 0 TABLE COMPANY
sqlite>
建立索引，再进行EXPLAIN分析查看结果，走了idx_sal索引扫描：
sqlite> CREATE INDEX idx_sal ON COMPANY(SALARY);
sqlite> EXPLAIN SELECT * FROM COMPANY WHERE Salary < 20000;
addr opcode p1 p2 p3 p4 p5 comment
---------- ---------- ---------- ---------- ---------- ---------- ---------- ----------
0 Trace 0 0 0 00
1 Integer 20000 1 0 00
2 Goto 0 25 0 00
3 OpenRead 0 2 0 5 00
4 OpenRead 1 3 0 keyinfo(1, 00
5 Affinity 2 0 0 cb 00
6 Rewind 1 22 2 0 00
7 SCopy 1 2 0 00
8 IsNull 2 22 0 00
9 Affinity 2 1 0 cb 00
10 IdxGE 1 22 2 1 00
11 Column 1 0 3 00
12 IsNull 3 21 0 00
13 IdxRowid 1 3 0 00
14 Seek 0 3 0 00
15 Column 0 0 4 00
16 Column 0 1 5 00
17 Column 0 2 6 00
18 Column 0 3 7 00
19 Column 1 0 8 00
20 ResultRow 4 5 0 00
21 Next 1 10 0 00
22 Close 0 0 0 00
23 Close 1 0 0 00
24 Halt 0 0 0 00
25 Transactio 0 0 0 00
26 VerifyCook 0 2 0 00
27 TableLock 0 2 0 COMPANY 00
28 Goto 0 3 0 00
sqlite> EXPLAIN QUERY PLAN SELECT * FROM COMPANY WHERE Salary < 20000;
order from detail
---------- ---------- --------------------------------
0 0 TABLE COMPANY WITH INDEX idx_sal
sqlite>
9，删除重复数据但是一条记录
录入测试数据
sqlite> .dump
PRAGMA foreign_keys=OFF;
BEGIN TRANSACTION;
CREATE TABLE COMPANY(ID INT NOT NULL, NAME VARCHAR(20),AGE INT,ADDRESS VARCHAR(20),SALARY DECIMAL(7,2));
INSERT INTO "COMPANY" VALUES(2,'Allen',25,'Texas',15000);
INSERT INTO "COMPANY" VALUES(3,'Teddy',23,'Norway',20000);
INSERT INTO "COMPANY" VALUES(4,'Mark',25,'Rich-Mond',65000);
INSERT INTO "COMPANY" VALUES(5,'David',27,'Texas',85000);
INSERT INTO "COMPANY" VALUES(6,'Kim',22,'South-Hall',45000);
INSERT INTO "COMPANY" VALUES(7,'James',24,'Houston',10000);
INSERT INTO "COMPANY" VALUES(7,'James',28,'Houston',20000);
INSERT INTO "COMPANY" VALUES(4,'Mark',29,'Rich-Mond',95000);
COMMIT;
sqlite>
查看重复记录数
sqlite> select * from company order by name;
ID NAME AGE ADDRESS SALARY
---------- ---------- ---------- ---------- ----------
2 Allen 25 Texas 15000
5 David 27 Texas 85000
7 James 24 Houston 10000
7 James 28 Houston 20000
6 Kim 22 South-Hall 45000
4 Mark 25 Rich-Mond 65000
4 Mark 29 Rich-Mond 95000
3 Teddy 23 Norway 20000
sqlite>
通过rowid来删除重复记录
sqlite> DELETE FROM COMPANY WHERE rowid NOT IN(SELECT MAX(rowid) rowid FROM COMPANY GROUP BY NAME);
sqlite>
再查看最新的数据记录，已经删除了重复NAME的记录
sqlite> select * from company;
ID NAME AGE ADDRESS SALARY
---------- ---------- ---------- ---------- ----------
2 Allen 25 Texas 15000
3 Teddy 23 Norway 20000
5 David 27 Texas 85000
6 Kim 22 South-Hall 45000
7 James 28 Houston 20000
4 Mark 29 Rich-Mond 95000
sqlite>