Joining Datasets

[news]

I have to join datasets, what is the quickest way?

eg

ds1 = [1,2,3,4,5]
ds2=[4,5,8,9,20]

jds = [4,5];

ie the data common to ds1 and ds2 is inserted into jds..


I know I can loop ds1 and test but this will be slow, and I am wondering if
it is better to create a stored procedure which checks, or insert into temp
tables and then check with an SQL query, eg a join


I am using two datasets as an example but it can be several.

 

[Richard Welch]

Select X from ds1 inner join ds2 where ds1.X = ds2.X

X = field name

 

[Uri Dor]

if they're sorted, then looping (in parallel) would be the fastest thing.
I believe even sorting and looping would be faster than inserting into a
DB and using SQL.
the main question here is why do you need to do this join in the
datasets and not originally in the DB where the data came from?

 

[Pat A]

If the data is coming from the database,

select from d1
intersect
select from d2