sql server prevent duplicate combination of two columns

I won't claim that this is the most elegant way of handling it but the code works.

(I declared the table to have a column called strategy with definition char(50))

declare @start datetime
declare @strat char(50)

set @start = '20120101'
set @strat = 'strategy1'

if not exists (select 1 from taba where StartDate = @start and Strategy = @strat)
   insert into mytable values (@start, @strat)

Select allOpen in new window

From a non code perspective, if the columns really are supposed to be unique, why not create a unique index on the fields?  Of course, then you'd have to trap a duplicate key condition on the insert.

You can create a unique key on the column combination i.e. StartDate, Strategy. This will not allow any duplicates at the backend. But you need to handle the error due to unique key in your program. We can help more here if we need your real requirement. I think you might be needing an extra data table here. It would be easier if you explain more on your requirement.

create unique index on those columns in that you wish to be unique values.

Thanks, I guess I was asking how do I create a unique index on those columns using entity framework code first

Thanks!

I am currently inserting one record at a time, and I do a query to see if the row already exists.  In other areas of my code, I create a tmp datatable object (C#) then use sql bulk copy for the insert.

So I want to change from
if(checkToSeeIfFileExists(row))
  insertFile(row)

into
datatable.add(row)
datatable.add(row)
datatable.add(row)


sqlBulkCopy(datatable)

if I use constrains, how would I be able to use this with bulk copy?