I have a full text catalog with two tables in it.
tableA has 4 columns (a1, a2, a3, a4) of wich 3 are indexed in the catalog, a2,a3,a4. a1 is the primary key.
t
I'm not positive that I understood what you were trying to do. I interpreted your question as you want to return all items in Table A that matched the search term. Furthermore you wanted to sum the rank from the item in TableA plus the matching items in TableB.
The best way I can think to do this is to use a table variable with 3 queries.
DECLARE @Results Table (a1 Int UNIQUE, Rank Int) --Insert into @Results all matching items from TableA INSERT INTO @Results (a1, Rank) ( SELECT TableA.a1, FT.Rank FROM TableA INNER JOIN FreeTextTable(TableA, *, 'search term') FT ON TableA.A1 = FT.[Key] ) --Update all of the ranks in @Results with a sum of current value plus the sum of --all sub items (in TableB) UPDATE @Results SET Rank = RS.Rank + FT.Rank FROM @Results RS INNER JOIN TableB ON RS.A1 = TableB.b2 INNER JOIN FreeTextTable(TableB, *, 'search term') FT ON TableB.b1 = FT.[Key] --Now insert into @Results any items that has a match in TableB but not in TableA --This query may/may not be desired based on your business rules. INSERT INTO @Results (SkillKeyId, Rank) ( SELECT TableB.b2, Sum(FT.Rank) FROM TableB INNER JOIN FreeTextTable(TableB, *, 'search term') FT ON TableB.b1 = FT.[key] LEFT JOIN @Results RS ON RS.a1 = TableB.b2 WHERE RS.a1 IS NULL GROUP BY TableB.b2 ) --All that's left is to return the results SELECT TableA.*, RS.Rank AS Total_Rank FROM TableA INNER JOIN @Results RS ON TableA.a1 = RS.a1 ORDER BY RS.Rank DESC
This isn't as elegant as using one query, but it should be easy to follow and allows you to decide whether or not to include records in the 3rd query.