SQL，力扣题目1635，Hopper 公司查询 I

一、力扣链接

二、题目描述

表: Drivers

+-------------+---------+
| Column Name | Type    |
+-------------+---------+
| driver_id   | int     |
| join_date   | date    |
+-------------+---------+
driver_id 是该表的主键(具有唯一值的列)。
该表的每一行均包含驾驶员的ID以及他们加入Hopper公司的日期。

表: Rides

+--------------+---------+
| Column Name  | Type    |
+--------------+---------+
| ride_id      | int     |
| user_id      | int     |
| requested_at | date    |
+--------------+---------+
ride_id 是该表的主键(具有唯一值的列)。
该表的每一行均包含行程ID(ride_id)，用户ID(user_id)以及该行程的日期(requested_at)。
该表中可能有一些不被接受的乘车请求。

表: AcceptedRides

+---------------+---------+
| Column Name   | Type    |
+---------------+---------+
| ride_id       | int     |
| driver_id     | int     |
| ride_distance | int     |
| ride_duration | int     |
+---------------+---------+
ride_id 是该表的主键(具有唯一值的列)。
该表的每一行都包含已接受的行程信息。
表中的行程信息都在“Rides”表中存在。

编写解决方案以报告 2020 年每个月的以下统计信息：

截至某月底，当前在Hopper公司工作的驾驶员数量（active_drivers）。
该月接受的乘车次数（accepted_rides）。

返回按month 升序排列的结果表，其中month 是月份的数字（一月是1，二月是2，依此类推）。

三、目标拆解

四、建表语句

Create table If Not Exists Drivers (driver_id int, join_date date)
Create table If Not Exists Rides (ride_id int, user_id int, requested_at date)
Create table If Not Exists AcceptedRides (ride_id int, driver_id int, ride_distance int, ride_duration int)
Truncate table Drivers
insert into Drivers (driver_id, join_date) values ('10', '2019-12-10')
insert into Drivers (driver_id, join_date) values ('8', '2020-1-13')
insert into Drivers (driver_id, join_date) values ('5', '2020-2-16')
insert into Drivers (driver_id, join_date) values ('7', '2020-3-8')
insert into Drivers (driver_id, join_date) values ('4', '2020-5-17')
insert into Drivers (driver_id, join_date) values ('1', '2020-10-24')
insert into Drivers (driver_id, join_date) values ('6', '2021-1-5')
Truncate table Rides
insert into Rides (ride_id, user_id, requested_at) values ('6', '75', '2019-12-9')
insert into Rides (ride_id, user_id, requested_at) values ('1', '54', '2020-2-9')
insert into Rides (ride_id, user_id, requested_at) values ('10', '63', '2020-3-4')
insert into Rides (ride_id, user_id, requested_at) values ('19', '39', '2020-4-6')
insert into Rides (ride_id, user_id, requested_at) values ('3', '41', '2020-6-3')
insert into Rides (ride_id, user_id, requested_at) values ('13', '52', '2020-6-22')
insert into Rides (ride_id, user_id, requested_at) values ('7', '69', '2020-7-16')
insert into Rides (ride_id, user_id, requested_at) values ('17', '70', '2020-8-25')
insert into Rides (ride_id, user_id, requested_at) values ('20', '81', '2020-11-2')
insert into Rides (ride_id, user_id, requested_at) values ('5', '57', '2020-11-9')
insert into Rides (ride_id, user_id, requested_at) values ('2', '42', '2020-12-9')
insert into Rides (ride_id, user_id, requested_at) values ('11', '68', '2021-1-11')
insert into Rides (ride_id, user_id, requested_at) values ('15', '32', '2021-1-17')
insert into Rides (ride_id, user_id, requested_at) values ('12', '11', '2021-1-19')
insert into Rides (ride_id, user_id, requested_at) values ('14', '18', '2021-1-27')
Truncate table AcceptedRides
insert into AcceptedRides (ride_id, driver_id, ride_distance, ride_duration) values ('10', '10', '63', '38')
insert into AcceptedRides (ride_id, driver_id, ride_distance, ride_duration) values ('13', '10', '73', '96')
insert into AcceptedRides (ride_id, driver_id, ride_distance, ride_duration) values ('7', '8', '100', '28')
insert into AcceptedRides (ride_id, driver_id, ride_distance, ride_duration) values ('17', '7', '119', '68')
insert into AcceptedRides (ride_id, driver_id, ride_distance, ride_duration) values ('20', '1', '121', '92')
insert into AcceptedRides (ride_id, driver_id, ride_distance, ride_duration) values ('5', '7', '42', '101')
insert into AcceptedRides (ride_id, driver_id, ride_distance, ride_duration) values ('2', '4', '6', '38')
insert into AcceptedRides (ride_id, driver_id, ride_distance, ride_duration) values ('11', '8', '37', '43')
insert into AcceptedRides (ride_id, driver_id, ride_distance, ride_duration) values ('15', '8', '108', '82')
insert into AcceptedRides (ride_id, driver_id, ride_distance, ride_duration) values ('12', '8', '38', '34')
insert into AcceptedRides (ride_id, driver_id, ride_distance, ride_duration) values ('14', '1', '90', '74')

五、过程分析

1、计算每月活跃的驾驶员数

2、计算每月接受的乘车次数

3、把1和2连接在一起

六、代码实现

with recursive t1(n) as(
    select 1
    union all
    select n + 1 from t1 where n + 1 <= 12
)   -- 12个月份
-- 卡点t2
,t2 as(
    select n, join_date, month(join_date) m, count(driver_id) over(order by n) + (select count(driver_id) from drivers where join_date < '2020-01-01') sum_cnt   -- 加上2020年之前的驾驶员数作累加
    from t1 left join Drivers d on t1.n = month(d.join_date) and year(join_date) = '2020'
-- 过滤2020年的数据并与当年所有的月份相连，根据图示理出思路
)
,t3 as(
    select n, count(ride_id) cnt from t1 left join
    (select requested_at, a.ride_id
    from Rides r left join AcceptedRides a on r.ride_id = a.ride_id
    where year(requested_at) = '2020') b
    on t1.n = month(b.requested_at)
    group by n
)
select t2.n month, max(sum_cnt) active_drivers, max(cnt) accepted_rides   -- 如果月内多于一天出现新增驾驶员/新增接受次数
from t2 join t3 on t2.n = t3.n
group by month;

七、结果验证

八、小结

1、CTE表达式 + 递归 + 子查询

2、日期函数year()、month()，聚合函数count()、max()