Hi Przemek,
> > One part which might be much for me is the memo handling, > > I'd appreciate your help in this area. > > OK, I'll look at it but we should not have any bigger problems here. > Just simply memo file should also be temporary and maybe we should > block some operations like memo packing. Thank you. > > > hb_rddCreateTableTemp() > > > we should have: > > > if( szAlias == NULL ) > > > szAlias = ""; > > I'd vote for RT error. > > OK but only if user does not pass alias at all. If he passes explicitly "" > as alias then we should allow empty alias just like in all other > open/create > functions to not force allocating unnecessary temporary alias. Hm, few minutes later I've found it's better to allow empty alias. It could be useful in practice. So that's what I did. > > xhb RDDs shipped as binaries, but I've never tried them > > with Harbour and it's difficult to guess whether they are > > working even without these modifications. > > I guess you are talking about SQLRDD and RMDBF* RDDs. > They cannot work just like any other binary RDDs without recompilation > with Harbour header files due to differences between both compilers. Yes, so we don't lose anything here then. Anyhow I'm trying to avoid the DBOPENINFO modification (plus your suggested fixes). Will redo the tests and if okay, commit. Brgds, Viktor
_______________________________________________ Harbour mailing list [email protected] http://lists.harbour-project.org/mailman/listinfo/harbour
