BIRT Dynamic Columns

Question

What I am currently looking to do is use our existing UI to select a number of columns from various tables (yes, multiple tables) and pass them into a BIRT report as parameters. From there, I am planning on building a query that will dynamically replace the columns into the query and pull the results automatically. I'll have to hide columns with no value passed to them as well. I also expect I'll have to setup the query to be a little heavy handed and already know all of the logical connections in the database (e.g. connect the proper tables, etc).

My question is this the best way to manage a dynamic column/table in a dataset? or is there a better way to manage this method? I'v seen some online information about the "ad-hoc" BIRT report designer that lets non-programmers create reports, but I am not looking for other people to actually build the report, just generate one using an existing template with interchangeable columns.

Answer 1

I think the easiest way is to firstly build a report with all columns that you need.

Then apply some logic on the visibility of the column. You can use the parameters there as well. Select your table, select a column, open the properties window and take a look at the visibility tab there. Just add some logic that results in a true or false.

If you are using a crosstab to display the information, you could use the filter options to include or exclude columns.

Yes, this will load data that is not used, but you need to build realy big reports for performance becomming a real issue.

If you try to add this logic in the actual dataset, you have to make the query and fetch script dynamic and then you still have the problem with the visualisation of the columns. I think you'll end up using the visibility script anyway (to show/hide the colums on the report), so might just start from there and have a working report fast.