Phil, Excellent, many thanks!
Ron -------------------------------------------------- From: "Phil Krylov" <p...@newstar.rinet.ru> Sent: Saturday, January 31, 2009 5:31 PM To: "Ron Pinkas" <ron.pin...@xharbour.com> Cc: <xharbour-developers@lists.sourceforge.net> Subject: Re: [xHarbour-developers] 2009-01-29 11:57 UTC-0430 Ron Pinkas <ron/at/xharbour.com> > Hi, > > On Sat, Jan 31, 2009 at 12:06 AM, Ron Pinkas <ron.pin...@xharbour.com> > wrote: >> Phil, I do have a question about class ERROR() implementation. Do you >> know >> why is it implemented as SIMULATED datas? The problem is that an Error >> Object can't be processed by functions such as ValToPrg(), and I can't >> think >> of any reason what it was implemented that way. Array access is just as >> simple with true datas. Do you have any reason we should not correct it? > > I think that 1) no C level API allowed to add datas at that moment. 2) > using methods allows us to validate values assigned, e.g. check if > they have needed type; in :genCode assignment we automatically set > :description. > Anyhow, I see the reason to use actual datas. I changed it, although > leaving intact most assign methods. > > -- Ph. > ------------------------------------------------------------------------------ This SF.net email is sponsored by: SourcForge Community SourceForge wants to tell your story. http://p.sf.net/sfu/sf-spreadtheword _______________________________________________ xHarbour-developers mailing list xHarbour-developers@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/xharbour-developers