From: perk AT iag DOT net (Bill Perkins)
To: MWODRICH AT eleceng DOT uct DOT ac DOT za, djgpp AT sun DOT soe DOT clarkson DOT edu
Subject: Re: Checking for disk in drive
Date: Sat, 06 May 1995 04:30:42 -0400
Lines: 34

> Recent replies indicate that one has to intercept the DOS critical 
> error handler etc. Why is all this neccessary in DJGPP ? 
> 
> I wrote a simple program in Turbo Pascal (flame war on the horizon?) 
> which lets you change drives to load stuff. Using the DOS routine 
> ChDir lets you change to, say, the A: drive and testing a variable 
> (DosError) tells you if an error occurred & what it was. There is no
> "abort/retry/fail" garbage involved, just an internal error number....
> 
> Considering how simple TPascal really is, and that it uses standard 
> DOS interrupt calls to do routines (like ChDir), can't DJGPP perform 
> the ChDir function in a similar (graceful) way ?
> 
> I realise this discussion list is not meant for arguments about DJGPP 
> vs. other languages, etc. but why use overkill and intercepts when 
> there is a simpler solution ?
> 
> (I do love DJGPP, despite this rant...)
> Mark Wodrich.
> -- 
>    _/_/_/_/_/      _/       Mark Wodrich
>   _/  _/  _/      _/        Electrical Engineering Student
>  _/      _/  _/  _/         University of Cape Town
> _/      _/_/_/_/_/          Cape Town, South Africa.
> 
>     "I'm gonna stick a quotation in here REAAAL soon ..."
> 

Just as a guess, as I haven't used TPascal in quite a few years (since I
learned C), TPascal probably revectors the critical error handler itself.
Borland's products, to date, have always been fine for PC's, albeit not
always portable to other platforms.
		Regards,
		Perk