>I have three .dbf named dcountry1, dstate1, ddistrict1
>
>1.
>table: dcountry1.dbf
>fields: country
>index: country candidate
>
>2.
>table: dstate1.dbf
>fields: state, country
>index: country+state candidate
> country regular
>
>3.
>table: ddistrict1.dbf
>fields: district, state, country
>index: country+state+district candidate
> country+state regular
>
>Set Relation:
>1. country(dcountry1.dbf) into country(dstate1.dbf)
>2. country+state(dstate1.dbf) into country+state(ddistrict1.dbf)
>
>At the time of prepare RI I am getting a message like following:
>
>Cascade updates that are based on expression-based index keys, may not behave as expected. The system has identified expression-based keys in the following relationship;
>
>dstate1 + ddistrict1
>
>Select Ok to continue with RI code generation. (Note: You can override the generated __ri_update_*code by creating stored procedures of the same name, and including those stored procedures AFTER the end of the generated RI code.
>
>
>I would like to know how to solve the problem. How to set correct RI.
>
>Regards,

Isn't 3rd table is all you need? The other 2 just looks like redundant data.
Cetin