Unless you specify, dataset queries are executed in parallel. The only time when this is not true is when there are dependencies between queries used for parameters. If you don't want them to execute in parallel, you can set the transaction flag on the data source. In this case, they will execute in the order they were defined. I can't guarantee that this will always be this way, but this is how it works in the RS2000 and RS2005 releases.
Hi,
Can you explain more about the transaction flag?
I have 4 datasets and i want to create one temp sql table for all of them.
i want that one of the four will executed first, build the temp table and only then all the rest tree datasets will executed.
Is it doable?
Thanks,
Yuval,
Yuvalch,
If you are using a temporary table like that, how will you handle concurrency issues when multiiple users run the report at the same time? I only ask this because I was thinking about going down this path, but couldn't find an elegant way to handle the concurrency issue.
R
One solution is to include a UserID column in the temp table, which you can get as a global variable. HTH.
Hello:
Can someone please explain where I would find the transaction flag on the data source. I have a situation where I need a dataset to execute out of the normal order.
Best regards
the flag is called "Use single transaction" and its in the data source
regarding the concurrency issues i use a temp table with a unique name: the name contain a unique number that generated by the rdl code:
shared function RndNum() as Integer
Randomize()
return CInt(Int((10000 - 1 + 1) * Rnd() + 1))
end function
