转自:http://www.maomao365.com/?p=5710
摘要:
今天接到一个需求,有一张数据表,记录的是消费明细数据,
现在需要做一个累计结余,记录每次的数据结余合计,
下文将展示一种sql脚本的编写方式
实验环境:sqlserver 2008 R2
如下例所示:
detail表 记帐流水表 ========== 字段: qt_srMoney (money) 收入 qt_zcMoney (money) 支出 qt_date (datetime) 操作日期 qt_dkfs (nvarchar) 打款方式 ------------------ 需获取一个带结余的数据信息 ========== 根据期初表中的日期和金额,查询出记帐表中的结余。 帐面结余 =期初金额+收入-支出 实验环境:sql server 2008 R2 */ create table detail(qt_srMoney money, qt_zcMoney money, qt_date datetime, qt_dkfs nvarchar(100)) insert into detail values(100,0,‘2018-1-2 10:00:00‘,‘other‘) insert into detail values(0,10,‘2018-2-2 10:00:00‘,‘other‘) insert into detail values(0,20,‘2018-3-2 10:00:00‘,‘other‘) insert into detail values(30,0,‘2018-4-2 10:00:00‘,‘other‘) insert into detail values(10,60,‘2018-5-2 10:00:00‘,‘other‘) go declare @datestart datetime ---开始计算日期 set @datestart =‘2018-2-1‘ declare @qcomney money --期初 set @qcomney =100 select t1.qt_date,t1.qt_dkfs,t1.qt_srMoney,t1.qt_zcMoney,sum(t2.qt_srMoney-t2.qt_zcMoney)+@qcomney from detail t1,detail t2 where t1.qt_date>=t2.qt_date and t1.qt_date >=@datestart group by t1.qt_date,t1.qt_dkfs,t1.qt_srMoney,t1.qt_zcMoney order by t1.qt_date; go ---避免日期出现两条导致结余数据计算异常----- declare @datestart datetime ---开始计算日期 set @datestart =‘2018-2-1‘ declare @qcomney money --期初 set @qcomney =100 with cteName as ( select qt_srMoney,qt_zcMoney,qt_date,qt_dkfs, row_number() over ( order by qt_date) as keyID from detail where qt_date >=@datestart ) select t1.keyId,t1.qt_date,t1.qt_dkfs,t1.qt_srMoney,t1.qt_zcMoney,sum(t2.qt_srMoney-t2.qt_zcMoney)+@qcomney from cteName t1,cteName t2 where t1.keyId>=t2.keyId group by t1.qt_date,t1.qt_dkfs,t1.qt_srMoney,t1.qt_zcMoney,t1.keyId order by t1.keyId; go truncate table detail drop table detail
原文地址:https://www.cnblogs.com/lairui1232000/p/8964949.html
时间: 2024-11-09 01:46:31