Sql 如何在Postgres中为同一列的总和创建两列？

我有一个包含以下列的表：

Type | Date | Amount

当类型具有类似“E”的值时，我使用SUM函数在“Amount”列中添加值（如SUM1）。 我的问题是：

select 
  to_char("Date", 'Mon') as mon,
  extract(year from "Date") as yyyy,
  sum("Amount") as SUM1 
from "Transactions"
where "Type" LIKE 'E%' and "Date" between '01/01/2015' and '08/31/2015'
group by 1,2
order by yyyy, mon;

这将产生如下行：

mon | yyyy | SUM1

mon | yyyy | SUM1 | SUM2

然而，我想做的是还显示另一个sum列（SUM2），显示从“Amount”中添加的值的总和，其中“Type”类似于“p”，因此显示如下行：

mon | yyyy | SUM1

mon | yyyy | SUM1 | SUM2

---

您可以使用case表达式作为sum函数的参数，以选择按如下方式求和的内容：

select 
    to_char("Date", 'Mon') as mon,
    extract(year from "Date") as yyyy,
    sum(case when "Type" LIKE 'E%' then "Amount" else 0 end) as SUM1,
    sum(case when "Type" LIKE 'P%' then "Amount" else 0 end) as SUM2
from "Transactions"
where "Date" between '01/01/2015' and '08/31/2015'
  and "Type" like 'E%' or "Type" like 'P%'
group by 1,2
order by yyyy, mon;

---

取决于您的PostgreSQL版本。对于旧款，您可以使用

CASE

进行条件求和：

select to_char("Date", 'Mon') as mon,
extract(year from "Date") as yyyy,
sum(CASE WHEN "Type" LIKE 'E%' then "Amount" ELSE 0 END) as SUM1,
sum(CASE WHEN "Type" LIKE 'P%' then "Amount" ELSE 0 END) as SUM2
from "Transactions"
where "Date" between '01/01/2015' and '08/31/2015'
group by 1,2
order by yyyy, mon;

或者，如果您使用的是PostgreSQL 9.4或更高版本，则可以使用更好的

过滤器

语法：

select to_char("Date", 'Mon') as mon,
extract(year from "Date") as yyyy,
sum("Amount") FILTER (WHERE "Type" LIKE 'E%') as SUM1,
sum("Amount") FILTER (WHERE "Type" LIKE 'P%') as SUM2
from "Transactions"
where "Date" between '01/01/2015' and '08/31/2015'
group by 1,2
order by yyyy, mon;