Question

我有类似

的sql

select
name,col2,col3,date
from
table1 join on few tables

结果是

name col2 col3 date
a    a    a    datetime1
a    a    a    datetime1
b    b    b    datetime2
b    b    b    datetime3

并且我不知道该怎么做，但我需要将日期列替换为列显示每个名称/行的工作日数

我虽然关于子查询，如

,(select COUNT(distinct DATENAME(dw, date) NOT IN ('Saturday', 'Sunday')) where name = '...' from ... where ...) as WorkingDays

但我需要帮助才能让它正常工作，谢谢。

Answer 1

也许你想要这个：

SELECT Name, C2,  
       WorkingDays = Sum(CASE WHEN Datename(DW, [date])IN( 'Saturday', 'Sunday' ) 
                         THEN 0 ELSE 1 END) 
                       OVER (PARTITION BY Name) 
FROM dbo.data

以下是样本数据的sql-fiddle：http://sqlfiddle.com/#!3/1279f/14/0

Answer 2

如果我的理解没有错，你只需要一个其他列的组合并计算＆＃39; date＆＃39;

   select
        name,col2,col3,count(date) as no_of_days
        from
        table1 join on few tables
        where DATENAME(dw, date) NOT IN ('Saturday', 'Sunday')
        group by name,col2,col3

子查询中的mssql distinct count

2 个答案: