Hi Jess,

 

I am facing the same problem right now. While, I do not know of any OOTB ways to do this, I initially implemented an all rule solution.

In my case, I had a lookup table that based on a age or gender got a rate or some value.

 

Using some Excel Trickery, I got the 2 dimensional table into  one dimensional columns of the type:

Age Gender   Value

Age and Gender became conditions and Value became the Action.

 

In your case, you could form a column with a dummy LHS(1=1) and keep all the columns as Action Columns.

In each Action Column you could build up your Fact.

 

For Example you could do the following:

 

Fact{ int a; int b;

}

Condition                        Action                Action

F: Fact

                                        f.set($1);            f.set($1)

Eval(true)                        Col A                  Col B

Dummy Value                 <values for a>    <values for b>

 

If you generate a rule base out of this, it should load your tables straight as facts.

 

However, do note, that I did some rather Q&D benchmarks for this and soon realized that the rulebase size and memory usage was getting really large. (Especially the PermGen Size).

 

You may want to have a look at the memory usage, in case you need to load a large number of tables. Due, to the above(and other reasons), I just wrote a custom parser using JXL (The same Excel Library Drools uses for Parsing Excels). It is really neat and would make a simple solution.

 

 

In, case you get another solution do post it to the list.

 

Cheers,

Vikrant