HiveSQL实战——大数据开发面试高频SQL题

news2024/11/16 16:01:03

查询每个区域的男女用户数

0 问题描述

每个区域内男生、女生分别有多少个

1 数据准备

use wxthive;
create table t1_stu_table
(
    id  int,
    name   string,
    class string,
    sex  string
);
insert overwrite table  t1_stu_table
values
     (4,'张文华','二区','男'),
     (3,'李思雨','一区','女'),
     (1,'王小凤','一区','女'),
     (7,'李智瑞','三区','男'),
     (6,'徐文杰','二区','男'),
     (8,'徐雨秋','三区','男'),
     (5,'张青云','二区','女'),
     (9,'孙皓然','三区','男'),
     (10,'李春山','三区','男'),
     (2,'刘诗迪','一区','女');

2 数据分析

select
       sex ,
       count(case when class = '一区' then sex else null end ) as 1area,
       count(case when class = '二区' then sex else null end ) as 2area,
       count(case when class = '三区' then sex else null end ) as 3area
from
     t1_stu_table
group by sex;

3 小结

略

==========================*****==========================

每个季度绩效得分大于70分的员工

0 问题描述

计算每个季度绩效得分都大于70分的员工

1 数据准备

create table t2_score_info_table
(
    id  int,
    name   string,
    subject string,
    score  int
);

insert overwrite table  t2_score_info_table
values
 (1,'王小凤','第一季度',88),
 (1,'王小凤','第二季度',55),
 (1,'王小凤','第三季度',72),
 (3,'徐雨秋','第一季度',92),
 (3,'徐雨秋','第二季度',77),
 (3,'徐雨秋','第三季度',93),
 (2,'张文华','第一季度',70),
 (2,'张文华','第二季度',77),
 (2,'张文华','第三季度',91);

2 数据分析

select
       id,
       name,
       min(score) as min_score
from
     t2_score_info_table
group by
         id,
         name
having min_score > 70;

ps：只要保证每个季度的每位员工的最小绩效得分大于70分，就可以说明这位员工该季度的绩效得分都大于70分。

3 小结

==========================*****==========================

行列互换

0 问题描述

把下表所示的纵向存储数据的方式改成横向存储数据的方式（行转列）

1 数据准备

create table t3_row_col_table
(
    year_num  int,
    month_num   int,
    sales int
);

insert overwrite table  t3_row_col_table
values
(2019,1,100),
(2019,2,200),
(2019,3,300),
(2019,4,400),
(2020,1,200),
(2020,2,400),
(2020,3,600),
(2020,4,800);

2 数据分析

select
     year_num,
     sum(case when month_num= 1 then sales end ) as m1,
     sum(case when month_num= 2 then sales end ) as m2,
     sum(case when month_num= 3 then sales end ) as m3,
     sum(case when month_num= 4 then sales end ) as m4
from t3_row_col_table
group by year_num

3 小结

==========================*****==========================

计算用户留存情况（*）

0 问题描述

计算用户的次日留存数、三日留存数、七日留存数

1 数据准备

create table t4_user_login
(
    uid   int,
    login_time string
);


insert overwrite table  t4_user_login
values
(1,'2021-4-21 6:00'),
(1,'2021-4-24 10:00'),
(1,'2021-4-25 19:00'),
(2,'2021-4-22 10:00'),
(2,'2021-4-28 9:00'),
(2,'2021-4-29 14:00'),
(3,'2021-4-27 8:00'),
(3,'2021-4-28 10:00');

2 数据分析

select
       count(case when day_value = 1 then uid  else null end ) as cnt1,
       count(case when day_value = 3 then uid  else null end ) as cnt2,
       count(case when day_value = 7 then uid  else null end ) as cnt3
from
     (select
           uid,
           max(date_format(login_time,'yyyy-MM-dd')),
           min(date_format(login_time,'yyyy-MM-dd')),
           datediff(max(date_format(login_time,'yyyy-MM-dd')), min(date_format(login_time,'yyyy-MM-dd'))) as day_value
     from t4_user_login
     group by uid
     ) t1;

3 小结

==========================*****==========================

筛选最受欢迎的课程

0 问题描述

筛选最受欢迎的课程course

1 数据准备

create table t5_course_table
(
    uid   int,
    name string,
    grade string,
    course string
);

insert overwrite table  t5_course_table
values
(1,'王小凤','一年级','心理学'),
(2,'刘诗迪','二年级','心理学'),
(3,'李思雨','三年级','社会学'),
(4,'张文华','一年级','心理学'),
(5,'张青云','二年级','心理学'),
(6,'徐文杰','三年级','计算机'),
(7,'李智瑞','一年级','心理学'),
(8,'徐雨秋','二年级','计算机'),
(9,'孙皓然','三年级','社会学'),
(10,'李春山','一年级','社会学');

2 数据分析

-- 解法一：
select
      course,
      count(course) as course_count
from t5_course_table
group by  course
order by course_count desc 
limit 1;
     
-- 解法二：
select
      course,
      course_count
from 
    (
     select
          course,
          count(course) as course_count,
          row_number()over(order by count(course) desc) as rn
     from t5_course_table
     group by  course
    )t1
where rn = 1;

3 小结

==========================*****==========================

筛选出每个年级最受欢迎的三门课程

0 问题描述

筛选出每个年级最受欢迎的三门课程

1 数据准备

create table t5_course_table
(
    uid   int,
    name string,
    grade string,
    course string
);

insert overwrite table  t5_course_table
values
(1,'王小凤','一年级','心理学'),
(2,'刘诗迪','二年级','心理学'),
(3,'李思雨','三年级','社会学'),
(4,'张文华','一年级','心理学'),
(5,'张青云','二年级','心理学'),
(6,'徐文杰','三年级','计算机'),
(7,'李智瑞','一年级','心理学'),
(8,'徐雨秋','二年级','计算机'),
(9,'孙皓然','三年级','社会学'),
(10,'李春山','一年级','社会学');

2 数据分析

select
      grade,
      course
from
  (
    select
          grade,
          course,
          cnt,
          row_number() over(partition by grade order by cnt desc) as rn
    from
        (
         select
              grade,
              course,
              count(uid) as cnt
         from t5_course_table
         group by  grade, course
        )t1
  )t2
where rn <= 3;