I needed to write a query for sum the values with the previous row value in current row of a column. So that I have written a single query after a long googled. Here I am trying to give you what I have done.
I have a table with three fields Id, Name and Mark. I have values of all fields like as follows.
Id Name Mark
--- ----- -----
1 aaaa 10
2 bbbb 20
3 cccc 30
Now I wants to get the results set of that table like as
Id Name Mark
--- ----- -----
1 aaaa 10
2 bbbb 30
3 cccc 60
So I need a single select query to do this performance. For that I have written a query using cross join.
Code Snippets for Table and Query
CREATE TABLE [dbo].[Marks]( [Id] [bigint] NOT NULL, [Name] [varchar](50) COLLATE SQL_Latin1_General_CP1_CI_AS NOT NULL, [Mark] [bigint] NOT NULL ) ON [PRIMARY]
select * from Marks
select a.Id, a.Name, sum(b.Mark) as Mark
from Marks a cross join Marks b
where b.Id <= a.Id group by a.Id, a.Name
Hence we have done the fetching the result set for sum of the previous rows in sql server.
For more ...
visit here
...S.VinothkumaR.
1 comment:
Sir,I am gunalan i am new to .net i want some tips to prepare .net
Post a Comment