MySql 查询数据库中所有表名以及对比分布式库中字段和表的不同

查询数据库中所有表名
select table_name from information_schema.tables where table_schema='数据库名' and table_type='base table';
查询指定数据库中指定表的所有字段名column_name
select column_name from information_schema.columns where table_schema='数据库名' and table_name='表名';

#查看分布式系统中不同库所有相同表名的comment的不同的表

select * from
(
select TABLE_NAME, TABLE_COMMENT from information_schema.tables where table_schema='test1_1207_bak' and table_type='base table'
) a
INNER JOIN
(
select TABLE_NAME, TABLE_COMMENT from information_schema.tables where table_schema='test2_1207_bak' and table_type='base table'
) b on a.TABLE_NAME=b.TABLE_NAME
where a.TABLE_COMMENT!=b.TABLE_COMMENT;

#查看分布式库中同一表的同一字段类型,字段长度和comment等不同之处

select a.table_name,b.table_name,a.column_name,b.column_name,a.data_type,b.data_type,a.COLUMN_TYPE,b.COLUMN_TYPE,
a.CHARACTER_MAXIMUM_LENGTH,b.CHARACTER_MAXIMUM_LENGTH,a.NUMERIC_PRECISION,b.NUMERIC_PRECISION,
a.NUMERIC_SCALE,b.NUMERIC_SCALE,a.COLUMN_KEY,b.COLUMN_KEY,a.EXTRA,b.EXTRA,a.my_COLUMN_COMMENT,b.my_COLUMN_COMMENT from
(
select
table_name,column_name,data_type,COLUMN_TYPE,CHARACTER_MAXIMUM_LENGTH,NUMERIC_PRECISION,NUMERIC_SCALE,COLUMN_KEY,EXTRA,
CASE
WHEN COLUMN_COMMENT is null THEN ''
ELSE COLUMN_COMMENT
END as my_COLUMN_COMMENT
from information_schema.columns
where table_schema='test2_1207_bak'
) a
INNER JOIN
(
select
table_name,column_name,data_type,COLUMN_TYPE,CHARACTER_MAXIMUM_LENGTH,NUMERIC_PRECISION,NUMERIC_SCALE,COLUMN_KEY,EXTRA,
CASE
WHEN COLUMN_COMMENT is null THEN ''
ELSE COLUMN_COMMENT
END as my_COLUMN_COMMENT
from information_schema.columns
where table_schema='test1_1207_bak'
) b on a.table_name=b.table_name and a.column_name=b.column_name
where a.data_type!=b.data_type or a.my_COLUMN_COMMENT!=b.my_COLUMN_COMMENT;

#查询缺失表语句:

select table_name from information_schema.tables where table_schema='test1_1207_bak' and table_type='base table' and table_name not in(
select table_name from information_schema.tables where table_schema='test2_1207_bak' and table_type='base table'
);

#改用存储过程实现为

# 对比不同库中表区别存储过程
DROP PROCEDURE IF EXISTS p_find_dif_table;
delimiter $$
create procedure p_find_dif_table(database_name1 VARCHAR(100),database_name2 VARCHAR(100))
begin
-- 比较database_name1库表名与database_name2表名不同的表
select table_name as db1_table_name from information_schema.tables where table_schema=database_name1 and table_type='base table'
and table_name not in(
select table_name from information_schema.tables where table_schema=database_name2 and table_type='base table'
);
-- 比较database_name2库表名与database_name1表名不同的表
select table_name as db2_table_name from information_schema.tables where table_schema=database_name2 and table_type='base table'
and table_name not in(
select table_name from information_schema.tables where table_schema=database_name1 and table_type='base table'
);
end;
$$
delimiter ;
-- 调用存储过程
CALL p_find_dif_table('test1_1207_bak','test2_1207_bak');

#查询具体表缺失字段语句:

select column_name from information_schema.columns where table_schema='test2_1207_bak' and table_name='oi_auto'
and column_name not in (
select column_name from information_schema.columns where table_schema='test1_1207_bak' and table_name='oi_auto'
);

# 改用存储过程和游标实现,对比数据库缺失字段

drop TABLE if EXISTS `release_no_field`;
CREATE TABLE `release_no_field` (
`column_name` varchar(255),
`table_name` varchar(255),
`db_name` varchar(255)
) ENGINE=InnoDB DEFAULT CHARSET=utf8 COMMENT='正式库中缺失字段'; drop TABLE if EXISTS `beta_no_field`;
CREATE TABLE `beta_no_field` (
`column_name` varchar(255),
`table_name` varchar(255),
`db_name` varchar(255)
) ENGINE=InnoDB DEFAULT CHARSET=utf8 COMMENT='测试库中缺失字段'; # 对比不同库中缺失字段存储过程
DROP PROCEDURE IF EXISTS p_find_dif_column;
delimiter $$
create procedure p_find_dif_column(release_db VARCHAR(100),beta_db VARCHAR(100))
begin
DECLARE cur_table VARCHAR(100);
DECLARE no_more_record INT DEFAULT 0;
DECLARE cur_record CURSOR FOR
SELECT TABLE_NAME from information_schema.tables
where table_schema=release_db and table_type='base table' and TABLE_NAME not in('release_no_field','beta_no_field');
DECLARE CONTINUE HANDLER FOR NOT FOUND SET no_more_record = 1;
OPEN cur_record;
FETCH cur_record INTO cur_table; delete from release_no_field;
delete from beta_no_field; WHILE no_more_record != 1 DO
-- 正式库中缺失字段
INSERT INTO release_no_field(column_name,table_name,db_name) select column_name,table_name,release_db from information_schema.columns where table_schema=beta_db and table_name=cur_table
and column_name not in (
select column_name from information_schema.columns where table_schema=release_db and table_name=cur_table
);
-- 测试库中缺失字段
INSERT INTO beta_no_field(column_name,table_name,db_name) select column_name,table_name,beta_db from information_schema.columns where table_schema=release_db and table_name=cur_table
and column_name not in (
select column_name from information_schema.columns where table_schema=beta_db and table_name=cur_table
);
FETCH cur_record INTO cur_table;
END WHILE;
CLOSE cur_record; SELECT * from release_no_field;
SELECT * from beta_no_field;
end;
$$
delimiter ;

对比不同库同一表同一字段类型不一致

SELECT
hosp_release.table_name AS table_name,
hosp_release.column_name AS column_name,
hosp_release.column_type AS hosp_release_column_type,
hosp_beta.column_type AS hosp_beta_column_type
FROM
(
select table_name,column_name,column_type from information_schema.columns where table_schema='hosp2' and table_name in (
select TABLE_NAME from information_schema.tables where table_schema='hosp2' and table_type='base table'
)) hosp_release ,
(
select table_name,column_name,column_type from information_schema.columns where table_schema='hosp_shunde_0226' and table_name in (
select TABLE_NAME from information_schema.tables where table_schema='hosp_shunde_0226' and table_type='base table'
)) hosp_beta
WHERE hosp_release.table_name=hosp_beta.table_name and hosp_release.column_name=hosp_beta.column_name and hosp_release.column_type!=hosp_beta.column_type
上一篇:(转)【mysql元数据库】使用information_schema.tables查询数据库和数据表信息 ---数据记录大小统计


下一篇:企业2.0杀出一号种子选手 “Linkwedo”横空出世