>>
>>Hi Todd... can you give me an example of #1 in your reply?
>
>Setup what I call a conversion table. This has several advantages that can be discussed in depth if you want.
>
>Structure as follows
>
>ncode (a numeric field length determined by number of codes max expected)
>name (a character field to store name of corresponding code)
>
>index on ncode as primary index (makes ncode a primary key)
>index on name
>
>when adding new code use a prime number table, instead of adding a new code as old code plus 1, is old code and skip to next record in prime number table then return value.
>
>Result is all ncodes are prime numbers.
>
>Now when searching to see like your example with the charvalue $ myfield, you do the following.
>
>Use a udf to convert the charvalue to a number based on the conv table
>so if charvalue was 'Red shoes female' and red was a 7 and shoes were a 11 and female were a 3 then red shoes female is 7*11*3 or 241. Now to pick all records that have red or shoes or female stored in them, issue a sql select where mod(241,myfield)=0 which is optimizable.
>
>HTH

Hi Todd...
I need some code if you do not mind.
I need some code that will SHOW a FORM if it detects the FORM is already
in memory. thanks