Date: Tue, 11 Jul 1995 18:22:34 +0200 To: "A.Appleyard" From: Anton Helm Subject: Re: Request for feature in DJGPP V2 Cc: djgpp AT sun DOT soe DOT clarkson DOT edu At 16:53 1995-07-11 BST, A.Appleyard wrote: >No need to alter the compiler! In your program, wherever logic reaches a place >which it never should, put in text to print out an appropriate `bug found' >error message. E.g. these bug-found statements in a text editor that I wrote:- > >if(n!=_keyarray) MOAN("BUG: this is not a keyarray"); return (*k)[N];} ... >pr(CW,"BUG: %sbound to a %s",cw,f.n>0?"string":keysort[-f.n]); ... >pr(CW,f.s?"BUG: %sbound to empty string":"%sis unbound",cw); ... >Display="BUG: this key sequence is not complete"; return; ... >pr(CW,"BUG: %sbound to integer %1d",cw,f.i); Display=CW; return;} ... > ... etc etc ?????????????????? Maybe I'm a little tired, but I don't understand what you mean. I *DO* have my own error handling functions that produce nice little messages on discovering something that might be wrong. The problems are bugs that cannot be predicted and didn't occur during testing. I just had such a bug that appeared for the first time after 2 years. I used this program at least 5 times a day. Tony *************************************************************************** Dipl.-Ing. Anton HELM * Private: * Institut fuer Nachrichtentechnik * * und Hochfrequenztechnik * Anton HELM * Guszhausstr. 25/389 * Gratian-Marx Str. 7/27 * A-1040 Wien * A-1110 Wien * AUSTRIA * AUSTRIA * ****************************************** PHONE : +43-1-58801-3520 OOOOOOO O O O O O OOOOO OO O FAX : +43-1-5870583 O O O O O O O O O O email: Anton DOT Helm AT nt DOT tuwien DOT ac DOT at O O O OOO O O O O OOO O O O tony AT nt DOT tuwien DOT ac DOT at O O O O O O O O O O O ********************************** O OOOO OO OO O OOOOO O OO