Shell脚本中执行sql语句操作mysql的5种方法

对于自动化运维，诸如备份恢复之类的，DBA经常需要将SQL语句封装到shell脚本。本文描述了在Linux环境下mysql数据库中，shell脚本下调用sql语句的几种方法，供大家参考。对于脚本输出的结果美化，需要进一步完善和调整。以下为具体的示例及其方法。

1、将SQL语句直接嵌入到shell脚本文件中

代码如下:
--演示环境 
[root@SZDB ~]# more /etc/issue 
CentOS release 5.9 (Final) 
Kernel /r on an /m 
 
root@localhost[(none)]> show variables like 'version'; 
+---------------+------------+ 
| Variable_name | Value      | 
+---------------+------------+ 
| version       | 5.6.12-log | 
+---------------+------------+ 
 
[root@SZDB ~]# more shell_call_sql1.sh  
#!/bin/bash 
# Define log 
TIMESTAMP=`date +%Y%m%d%H%M%S` 
LOG=call_sql_${TIMESTAMP}.log 
echo "Start execute sql statement at `date`." >>${LOG} 
 
# execute sql stat 
mysql -uroot -p123456 -e " 
tee /tmp/temp.log 
drop database if exists tempdb; 
create database tempdb; 
use tempdb 
create table if not exists tb_tmp(id smallint,val varchar(20)); 
insert into tb_tmp values (1,'jack'),(2,'robin'),(3,'mark'); 
select * from tb_tmp; 
notee 
quit" 
 
echo -e "/n">>${LOG} 
echo "below is output result.">>${LOG} 
cat /tmp/temp.log>>${LOG} 
echo "script executed successful.">>${LOG} 
exit; 
 
[root@SZDB ~]# ./shell_call_sql1.sh  
Logging to file '/tmp/temp.log' 
+------+-------+ 
| id   | val   | 
+------+-------+ 
|    1 | jack  | 
|    2 | robin | 
|    3 | mark  | 
+------+-------+ 
Outfile disabled. 

2、命令行调用单独的SQL文件

代码如下:
[root@SZDB ~]# more temp.sql  
tee /tmp/temp.log 
drop database if exists tempdb; 
create database tempdb; 
use tempdb 
create table if not exists tb_tmp(id smallint,val varchar(20)); 
insert into tb_tmp values (1,'jack'),(2,'robin'),(3,'mark'); 
select * from tb_tmp; 
notee 
 
[root@SZDB ~]# mysql -uroot -p123456 -e "source /root/temp.sql" 
Logging to file '/tmp/temp.log' 
+------+-------+ 
| id   | val   | 
+------+-------+ 
|    1 | jack  |