SQL 按连续出现的值分组

Question

我有以下 table 2 列

ID     |     Dept 

1      |    A    
2      |    A
3      |    B
4      |    B
5      |    B
6      |    A

我想做一个计数，使输出看起来像下面的 table。

Dept   |     Count

A      |    2    
B      |    3
A      |    1

提前感谢您的帮助！

Answer 1

运行此查询：

SELECT Dept, count(*) FROM table_name group By Dept

Answer 2

您不能使用 SQL 执行此操作。 Count 计算不同的项目，因此在您的情况下 count 会给您 A 和 B.

的计数

您只能 count/group 按 table 中的值，不能按行顺序。如果您无论如何都不使用 order by，则无法保证 SQL 中的顺序。

Answer 3

从您的示例来看，您似乎想要计算每个部门的连续记录。

您可以通过组合行号和订购 ID 来完成此操作。

create table tblDept (
    id int not null, 
    dept varchar(50)
);

insert into tblDept values (1, 'A');
insert into tblDept values (2, 'A');
insert into tblDept values (3, 'B');
insert into tblDept values (4, 'B');
insert into tblDept values (5, 'B');
insert into tblDept values (6, 'A');

with orderedDepts as (
  select
    dept,
    id,
    row_number() over (partition by dept order by id) - 
      row_number() over (order by id) as rn
  from tblDept
)
select
  dept,
  count(*) as num
from orderedDepts
group by
  dept,
  rn
 order by
   max(id)

给出输出：

+------+-----+
| DEPT | NUM |
+------+-----+
| A    |   2 |
| B    |   3 |
| A    |   1 |
+------+-----+

SQL Fiddle

Answer 4

与迈克尔的略有不同，结果相同：

with cte1 as (
  select   id,
           dept,
           row_number() over (partition by dept order by id) - 
             row_number() over (order by id) group_num
  from     test),
cte2 as (
  select   dept,
           group_num,
           count(*) c_star,
           max(id) max_id
  from     cte1
  group by dept,
           group_num)
select   dept,
         c_star
from     cte2
order by max_id;

http://sqlfiddle.com/#!4/ff747/1

SQL 按连续出现的值分组

SQL Grouping by sequential occurrences of a value

sql

database

oracle

gaps-and-islands