Statistics of functions commonly used in sql online programming on Niuke.com (continuously updated)

This article is just to record some sql functions and precautions used when brushing sql test questions. The knowledge points are relatively fragmented. If the judge wants to learn more about its usage, you may have to go to Baidu by yourself;

 #sql中关键字出现的顺序是：
select/from/where/group by/having/order by/limit

1.distinct 

distinct # 得到不重复的值

2. limit 

#注意：查询到的记录默认从0开始编号，而sql字符串默认从1开始编号；要注意两者的区别
select * from user limit i,n; # 从索引i处开始查找，返回n条查询结果
select * from user limit n; # 返回n条查询结果

3. between and 

# 查找年龄在[20,23]之间的设备号，性别，年龄
select device_id,gender,age from user_profile where age between 20 and 23; 

4. Inequality sign 

# 不等号
select device_id,gender,age,university from user_profile where age <> "";
select device_id,gender,age,university from user_profile where university!="复旦大学";

5. Judgment representation of attribute value as string

# 对某个属性值的字符串判断的错误写法
select device_id,gender,age,university,gpa from user_profile where gpa >3.5 and gender is "male";
# 正确写法
select device_id,gender,age,university,gpa from user_profile where gpa >3.5 and gender = "male";

6. 或  or

# 或
select device_id,gender,age,university,gpa 
from user_profile 
where university="北京大学" or gpa > 3.7

7. in 

# in
select device_id,gender,age,university,gpa from user_profile where university in ("北京大学","复旦大学","山东大学");

8. Mixed use of operators

# 操作符的混合使用
select device_id,gender,age,university,gpa from user_profile
where (gpa > 3.5 and university = "山东大学") or (gpa > 3.8 and university = "复旦大学");

 9. like % _

# like
# %:0个或多个字符
# _:1个字符
select device_id,age,university from user_profile where university like '%北京%';

 10. max

# max
select max(gpa) from user_profile where university = "复旦大学";

11. count avg 

# count avg
select count(gender),avg(gpa) from user_profile where gender="male";

 12. group by a,b,c... and the correct notation

# group by a,b,c,...  可以多个属性值为一组，来进行分组，一组对应多个符合条件的值
# 注意：没有这种写法：group by (a,b,c,...)
select gender,university,count(id),avg(active_days_within_30),avg(question_cnt) from user_profile group by gender,university;
# 下面这种写法是不对的
select gender,university,count(id),avg(active_days_within_30),avg(question_cnt) from user_profile group by (gender,university);

13. having 

# 分组后选取符合条件的值用having
# 取出平均发贴数低于5的学校或平均回帖数小于20的学校
select university,avg(question_cnt) as avg_question_cnt,avg(answer_cnt) as avg_answer_cnt 
from user_profile 
group by university 
having avg_question_cnt < 5 or avg_answer_cnt < 20;
# 错误写法1  既然分组了，对分组后得到的值按要求取舍，只能用having，不能用where
select university,avg(question_cnt),avg(answer_cnt) from user_profile where avg(question_cnt) < 5 or avg(answer_cnt) < 20 group by university;
# 错误写法2   把聚合函数得到的值作为新的列名，再根据新的列名查找
select university,avg(question_cnt),avg(answer_cnt) from user_profile group by university having avg(question_cnt) < 5 or avg(answer_cnt) < 20;

14. order by 

# order by 在 group by 之后
# asc 升序(默认) desc 降序
select university,avg(question_cnt) as avg_question_cnt 
from user_profile 
group by university
order by avg_question_cnt;

15. A left join B on 条件 

# A left join B on 条件 ：将属性值与A表相匹配的B表插入到A表上
select question_practice_detail.device_id,question_practice_detail.question_id,question_practice_detail.result 
from question_practice_detail left join user_profile 
on  question_practice_detail.device_id = user_profile.device_id
where user_profile.university = "浙江大学"
order by question_practice_detail.question_id;

16. Keep n digits after the decimal point

# 保留小数点后n为
round(x,n)

 17. Count the number of rows in the data table

# 统计表中数据的行数
count(*)

18. The difference between union all and union

# 将2个sql语句查询到的结果(A,B)给合并起来，结果不去重，保留重复的行
A union all B
# 比如
select device_id,gender,age,gpa from user_profile where university = "山东大学"
union all
select device_id,gender,age,gpa from user_profile where gender = "male";

# union 结果去重，不保留重复的行

19. if statement 

# sql if语句
# if(condition,A,B) 如果符合条件就取值A，否则取值B
select 
    if(age >= 25,"25岁及以上","25岁以下") as age_cut,count(*) 
    from user_profile group by age_cut;

20. case statement 

# case 语句
select device_id,gender,
    case
        when age < 20 then "20岁以下"
        when age between 20 and 24 then "20-24岁"
        when age >= 25 then "25岁及以上"
        else "其他"
    end
    as age_cut
from user_profile;

21. Date functions 

# 日期
# day(date) 日期中当月的第几天
# month(date) 日期中的月份
# year(date) 日期中的年份
select day(date) as day,count(*) as question_cnt 
from question_practice_detail 
where month(date) = 8 
group by day;

22. Date difference 

# datediff(date2,date1) 日期差

23. substring_index(str,delim,count)  

# substring_index(str,delim,count) 根据分隔符在字符串中的位置，来返回分隔符最左边或最右边的所有内容
# str 源字符串 delim 分隔符 count 第几个分隔符
#count > 0:从左往右数，第count个分割符的左边的所有内容
#count < 0:从右往左数，第coun个分隔符的右边的所有内容
############################
# profile:180cm,75kg,27,male
select substring_index(profile,',',-1) as gender,count(device_id) as number 
from user_submit
group by gender;

24. substring(str,start,len) 

# 截取字符串
substring(str,start,len) 字符串的索引编号默认从1开始 从字符串str的start位置开始，截取长度为len的字符串

25. The combination of sum and if 

# sum 与 if 的结合使用
sum(if(result = "right",1,0))

26.  count(column)   count(a,if(b_condition,true,null))   count(a_condition or null)

# count(column) 具有自动忽略空值的功能，即对column中值为空的不计数
# count(a,if(b_condition,true,null)) if中的b不能与a相同，否则会报错；这个公式根据b的值来统计a的数目
# 注意：count 与 if相结合使用的时候，空值只能用null表示，不能用0
# count(a_condition or null) 根据a的自身条件来统计 
################################################
# 现在运营想要了解2021年8月份所有练习过题目的总用户数和练习过题目的总次数
select count(distinct device_id,if(result is not null,TRUE,null)) did_cnt,
count(result is not null or null) question_cnt
from question_practice_detail
where year(date) = '2021' and month(date) = '8';

 27. group_concat()

# group_concat()  将分组后指定的属性值用分隔符进行连接 分隔符默认为","
# 分割符指定的时候需要加关键字 separator
select university,group_concat(distinct age separator ",") age_concat
from user_profile
group by university;

28.  concat(str1,str2,....)

# 字符串拼接，str1 + str2
concat(str1,str2,...)

29. Return in specified order

# 按"下旬，中旬，上旬"的顺序进行排序
order by locate(substr(dt_range,9,2),"下旬,中旬,上旬")

30. date_format(date,format) date format

# date_format(date,format) 日期的格式化
date_format(event_date,"%Y年%m月")

31. Character Length Byte Length

# 字符长度 char_length(str) 汉字、字母、数字都算一个字符
# 字节长度 length(str) utf8，汉字3个字节；gbk汉字2个字节
#######################################
# 注意：不能用nick_len作为where判断的条件
select device_id,nick_name,char_length(nick_name) nick_len
from user_submit
where char_length(nick_name) > 4
order by device_id desc;

32. upper(str) lower(str) string case conversion

upper(str) 转大写字母
lower(str) 转小写字母

33. trim(str) remove spaces on both sides

trim(str) 去左右两边的空格

34. reverse(str) reverse the string

# 反置字符串
reverse(str)

35. with as usage

# with as 用于提前将连接操作从主函数中分离出来，以使逻辑更加清晰
# 2种写法：
# with name as (...)
# with as (...) name
with q as (select device_id,date_format(event_date,"%Y-%m") new_date,count(*) cnt
        from question_practice_detail
        group by device_id,date_format(event_date,"%Y-%m")
        having count(*) > 1)
select distinct q1.device_id
from q q1 left join q q2 on q1.device_id = q2.device_id
# 多个别名
with name1 as (...),
    name2 as (...),
    name3 as (...),
    ...
#注意：以下这种写法不对
with name1 as (...)
with name2 as (...)
with name3 as (...)

36. left join、right join、inner join

reference blog

A left join B on condition  #以A表为基准，用B的信息进行填补
A right join B on condition  #以B表为基准，用A的信息进行填补
A inner join B on condition  #以condition为基准，取AB两表共有的
# 当多个表使用inner join的时候，要尽量把属性列多的那个表放在最左边

37. Insert statement insert into table_name values(,,,)

insert into table_name (columns1,columns2,...) values (v1,v2,...)

insert into table_name set c = v1,c = v2,... 

insert into table_name select * from table_name2

reference

# 插入语句 空值用null 自增值auto_increment用default填补
insert into table_name values (,,,) #插入的值的顺序要严格按照定义时的顺序一致
# 指定列名来插入值
insert into table_name (column_name1,column_name2,column_name3,...) values (v1,v2,v3,...)
# 或
insert into table_name set column_name1 = v1,column_name2 = v2,column_name3 = v3,....
# 将查询到的数据插入到一个表中
insert into table_name select * from table_name2 # table_name为要插入的表，table_name2为要查询的表，需保证查询出来的值要与table_name中预定义的属性值的顺序一致
# 若要将select查询到的指定数值插入
insert into new_table (column1,column2,...) select (column1,column2,...) from table where..

 38. date_add(date, interval num type) and date_sub(date, interval num type) date interval

reference

# 求日期间隔
date_add(date,interval 3 day) # 在date1的基础上，增加3天
# day是间隔单位，还可取值second,minute,year,month,hour,week等
date_sub(date,interval 3 day) # 在date1的基础上，减去3天

39. insert into/insert ignore into/replace into when inserting duplicate key values

reference

# 具体用法与37类似，这里只讲操作的大致原理
# 插入一行数据，若主键已存在，则会报错，否则，直接插入
insert into 
# 插入一行数据，若主键已存在，则新的数据替代原来的数据(先删除再插入)，否则，直接插入
replace into
# 插入一行数据，若主键已存在，则不插入，否则，直接插入
insert ignore into

40. update table_name set column1 = v1, column2 = v2... where update an attribute value in the table

# 更新属性值
update table_name set column1 = v1,column2 = v2... where ...

41. delete from table_name where... Delete the specified record in the table

delete from table_name where ...

42. timestampdiff(type,start,end) Find the time difference, the unit is type

# type取值与38类似
# end-start的单位为type的一个时间差
timestampdiff(type,start,end)

 43. ifnull(exp,a) 

# exp不为null，则返回exp的值，否则返回b；
# exp可以为一个条件表达式，也可以为一个字符串或数字等等
ifnull(exp,b)

44. truncate table table_name deletes records and resets the auto-increment key value

# 删除表中的记录并重置自增主键的索引
truncate table table_name
# 或
truncate table_name

45. Create a new table

reference

create table if not exists table_name(
        column_name type # 字段名，类型
        primary key # 主键
        foreign key # 外键
        auto_increment #自增
        comment "name" # 列注释
        default value # 默认值
        unique # 唯一约束，同列不允许有相同的值;解决表中多个字段需要唯一性约束的问题；允许多个空值
        not null #非空,默认为空
) default charset = "" # 编码形式

 46. ​​Modify/delete/add field names in the table

reference

alter table table_name add column column_name type # 增加字段名
alter table table_name drop column column_name type # 删除字段名
alter table table_name modify column column_name type # 修改字段类型
alter table table_name change column column_name new_column_name type # 修改字段名
# 在指定的字段名后添加一个新的字段，假如新字段名为new_column 指定的字段为specified_column
alter table table_name add new_column type after specified_column
# 注意：不管修改什么，type均不能少，否则会报错

47. Modify table name

alter table old_table_name rename to new_table_name

48. Delete table

drop table table_name

49. Create Index

# 创建索引的基本语法
create index_mode index index_name on table_name(column_name)
# index_mode:空值 普通索引，索引值可重复
#            unique 唯一索引，索引值不可重复，可有多个空值
#            fulltext 全文索引，在给定的列中存储有关重要的词及位置信息，一种特殊类型的基于标记的功能性索引             

50. Delete index

# method 1
drop index index_name on table_name;
# method 2
alter table table_name drop index index_name;

51. Find the number of days in the current month, how many days are there in the current month

last_day(date) #当月有几天

52. any_value()

MySQL提供了any_value()函数来抑制ONLY_FULL_GROUP_BY值被拒绝。
所以只需要在非group by的列上加any_value()就可以了

53. The correct combination of union and order by

# 当order by出现在union的子句中,order by会失效
# 比如：以下得到的结果并不是上表与下标都排序好，再连接到一起的；它们虽然连接到了一起，虽然各排各的，最后再连接，但并不是排序好的
# 因为order by 出现在 union的子句中，排序会失效
(select exam_id tid,count(distinct uid) uv,count(*) pv 
from exam_record
group by exam_id
order by uv desc,pv desc)
union
(select question_id tid,count(distinct uid) uv,count(*) pv
from practice_record
group by question_id
order by uv desc,pv desc)
###############################################
# 要想使order by与union一起使用时，使得order by生效，那么要把order by放在union的子句的子句中
# a,b两个表名不可少
select * from
(select exam_id tid,count(distinct uid) uv,count(*) pv 
from exam_record
group by exam_id
order by uv desc,pv desc)a
union
select  * from
(select question_id tid,count(distinct uid) uv,count(*) pv
from practice_record
group by question_id
order by uv desc,pv desc)b
###############################################
##### 综上 #####
## 如果想要对未union前两个sql语句的查询结果进行排序，分别单独排序需要的数据，查出以后再使用union连接
select * from
( select * from t1  order by 字段 )t1 -- 一定要对表重新命名，否则报错 
union select * from
( select * from t2  order by 字段 )t2

54. When connecting left and right, consider duplicate rows

# 左连接时，当右表连接到左表，当左表的一个值对应右表的多个值的时候；
# 要考虑到左表出现重复行的情况 
# 当右表多个值对应左表多个值时，要考虑到连接后的左右表出现重复行的情况
# 还要考虑每个记录表中的提交时间可能会重复的情况