At 07:00 PM 15-01-02 +1100, TeamIBO wrote:

>A third option is to setup the join as you have it and then setup
>stored procedures to do the edit/insert/delete. Not sure whether this
>works for non-KeyRelation fields or not (I've never tried it).

This is the right option for the scenario where the user wants to edit the entire dataset on a hunt-and-pick basis, as if it all came from a single table. In this case, you would not set KeyRelation at all; and you would have parameterised dml stored procedures for all of the three UpdateSQL properties. This works beautfiully with IBO.

I strongly advise retaining the column names as parameter names throughout; and make certain that your output dataset contains only one instance of a join column name, i.e DON'T use select *, use a column list and output the join column(s) from only the leftmost table in the join. Use full SQL-92 JOIN syntax and proper identifiers on every column.

Avoid outer joins if possible because they may make your KeyLinks difficult to determine.

You will need to be especially careful about choosing your KeyLinks for this dataset. If you get it right, your refreshes will work like a dream.


regards,
Helen Borrie (TeamIBO Support)

** Please don't email your support questions privately **
Ask on the list and everyone benefits
Don't forget the IB Objects online FAQ - link from any page at www.ibobjects.com