鲨臂的题目们

最新推荐文章于 2025-12-15 21:10:21 发布

原创最新推荐文章于 2025-12-15 21:10:21 发布 · 166 阅读

0 ·

CC 4.0 BY-SA版权

文章标签：

#数据库 #sql #mysql

1.如何在20s内完成两个表各100w条数据的快速比对。

CREATE DATABASE test;
use test;
CREATE TABLE student (
	id int,
	name varchar(255),
	address varchar(255),
	city varchar(255)
);

CREATE TABLE teacher (
	id int,
	name varchar(255),
	address varchar(255),
	city varchar(255)
);

INSERT INTO student VALUES (1, '刘一', '郑十家隔壁', '河南');
INSERT INTO student VALUES (2, '陈二',  '李四家隔壁', '安徽');
INSERT INTO student VALUES (3, '张三',  '白娘子家隔壁', '杭州');
INSERT INTO student VALUES (4, '李四',  '许仙家隔壁', '杭州');
INSERT INTO student VALUES (5, '王五',  '李四家隔壁', '杭州');
INSERT INTO student VALUES (6, '赵六',  '赵六家隔壁', '杭州');
INSERT INTO student VALUES (7, '孙七',  '张三家隔壁', '杭州');
INSERT INTO student VALUES (8, '周八',  '雷峰塔附近', '杭州');
INSERT INTO student VALUES (9, '吴九',  '孙七家隔壁', '杭州');
INSERT INTO student VALUES (10, '郑十',  '周八家隔壁', '杭州');

INSERT INTO teacher VALUES (1, '刘一', '郑十家隔壁', '河南');
INSERT INTO teacher VALUES (2, '陈二',  '李四家隔壁', '安徽');
INSERT INTO teacher VALUES (3, '张三',  '白娘子家隔壁', '杭州');
INSERT INTO teacher VALUES (4, '李四',  '许仙家隔壁', '杭州');
INSERT INTO teacher VALUES (5, '王五',  '李四家隔壁', '杭州');
INSERT INTO teacher VALUES (6, '赵六',  '赵六家隔壁', '杭州');
INSERT INTO teacher VALUES (7, '孙七',  '张三家隔壁', '杭州');
INSERT INTO teacher VALUES (8, '周八',  '雷峰塔附近', '杭州');
INSERT INTO teacher VALUES (9, '吴九1',  '孙七家隔壁', '杭州');

-- select md5(concat_ws('|',name,address)) as record_key from teacher;
-- SELECT * FROM student;
-- select * from teacher;
--  select   l.record_key,r.record_key
select count(distinct l.record_key) as left_table_num, 
    count(r.record_key) as right_table_num, 
    count(distinct case when l.record_key= r.record_key then l.record_key END) as left_right_equal_num 
from 
    (select md5(concat_ws('|',name,address)) as record_key from student) l
left join 
    (select md5(concat_ws('|',name,address)) as record_key from teacher) r
    on l.record_key = r.record_key 
union all
select count(distinct l.record_key) , 
    count(r.record_key) , 
    count(distinct case when l.record_key= r.record_key then l.record_key END)  
from 
    (select md5(concat_ws('|',name,address)) as record_key from teacher) r
left join 
    (select md5(concat_ws('|',name,address)) as record_key from student) l
    on r.record_key = l.record_key  
where l.record_key is NULL

left join + right join +union all 模拟 full outer join

左表数量，右表中匹配的数量，左右表相等的数量。

未匹配的记录， right 中独有的记录，未匹配的记录