Nested Decision Table Issue

[Former Member]

Hi All,

I have a below scenario where in I will have a input like

- Japan  Tomato  200.

Below is how my decision table setup looks like. The 'Item' condition has nested decision tables.

Based on the input i want the the BRF PLus to evalute Rule2 and further evaluate nested Decisoin table 'Vegie' to check if tomato belongs to Veggie and return result as 'WASI'.

The issue that i am facing is, when i try to activate the Main decision table, i get the error message that the 'Item' column must have a boolean result.

Is it possible using my approach?

Now i am guessing how to achieve this:

        -> without boolegan

        -> and With Boolegan

Main Descision Table
	Country	Item	Quantity	Approver (Result)
Rule1	Paris	Fruits	>1000	SAM
Rule2	Japan	Vegie	>100	Wasi
Rule3	Spain	DryFruits	>500	Mike
Sub Decision Table - Fruits
Item	Item Group (Result)
Apple	Fruits
Banana	Fruits
grapes	Fruits
Sub Decision Table - Veggie
Item	Item Group (Result)
tomato	Veggie
potato	Veggie
onion	Veggie
Sub Decision Table - DryFruits
Item	Item Group (Result)
dates	DryFruits
almond	DryFruits
cashew	DryFruits

[carsten_ziegler · Accepted solution]

You nest the decision table not into the cell but into the column.

In your main decision table there will be a column ITEM GROUP.

The column ITEM GROUP is a decision table with all lines of your nested tables above.

Then you can use the result of the nested table in your main table like any other field/column.

[Former Member · Accepted solution]

Thanks Carsten for reply,

Yes, I think  thats what I have done.  but while activation i get the error message that the 'Item' column must have a boolean result..

Master Decision table column 'Item' and Nested decision table column 'Item Group' is the same.

[Former Member · Accepted solution]

Wow. All this time I've been using BRFplus and I didn't know you could nest expressions at the decision table column level (as opposed to the cell level). Now I fell silly. This could have simplified some of the decision tables I've created in the past. Thanks Carsten!

Jameela

When Carsten suggests nesting a decision table into a column of the main decision table he means that you should:

1. Create a new decision table that consolidates all the entries from your three sub decision tables into one. (The three separate decision tables are no longer needed and can be deleted)

2. Add a new column in the main decision table via the "Insert Column -> From Expression" option and then select your new decision table.

Once that is done, you will be able to enter values in each cell of the new column that will be compared to the result of new decision table.

Regards

Glen

[carsten_ziegler · Accepted solution]

This does not make any sense to me. A column never enforces a specific result type.

Can you activate the inner decision table?

Then empty all cells of the respective column of your main decision table. Then please activate the main decision table.

[carsten_ziegler · Accepted solution]

You are not silly at all. There are many good answers from you in this forum.

Technichally it is very simple. A data object as well as an expression is just an ID for the using object (e.g. the main decision table). During processing the ID needs to be resolved into a value. That value can be the value of a data object or the result of an expression. Once the value is there it can be used such as in comparisons of cell values.

The beauty of the architecture is that an expression does not need to know what an ID is, as long as its type fits. Expressions have results that have types such as context data objects.

[Former Member · Accepted solution]

Thanks Carsten and Glen for the help.

[kutjohn]

Hi Jameela,

Did you find the solution???

Regards JP

[Former Member]

Yes John.. reply by Glenn is the solution.

[Former Member]

Hi

Rather than using decision tables for the nested expressions, you could use value ranges instead. If the Item matches one of the values in the value range then a boolean True will be returned to that cell in the decision table.

Regards

Glen