>Actually, the > RI generator of VFP 3.0 doesn't work well when a table has a composite > primary key, dunno if 5.0 does it right. > > > >Ciao, > > > >J. Luis > > don't think it does... > it doesn't (btw RI code generated by xCase does take care of that - fox VFP3.0 and 5.0) Arnon