From:	sandmann AT clio DOT rice DOT edu (Charles Sandmann)
Message-Id:	<10010021344.AA12079@clio.rice.edu>
Subject:	Re: Beta Testers Needed
To:	eliz AT is DOT elta DOT co DOT il (Eli Zaretskii)
Date:	Mon, 2 Oct 2000 08:44:56 -0500 (CDT)
Cc:	djgpp-workers AT delorie DOT com
In-Reply-To:	<Pine.SUN.3.91.1001002072512.27106C-100000@is> from "Eli Zaretskii" at Oct 02, 2000 07:32:38 AM
X-Mailer:	ELM [version 2.5 PL2]
Mime-Version:	1.0
Reply-To:	djgpp-workers AT delorie DOT com
Errors-To:	nobody AT delorie DOT com
X-Mailing-List:	djgpp-workers AT delorie DOT com
X-Unsubscribes-To:	listserv AT delorie DOT com

> On Sun, 1 Oct 2000, Charles Sandmann wrote:
> > Originally system() was failing, but I wasn't checking the return status.
> > I changed the path, then system() worked.  Then I changed the path to a
> > simple failure example, and system() failed.  It continued to fail when 
> > I put the path back to just c:\DOS
> > 
> > I changed the system("command") to i = system("command.com") so I could 
> > check the return status, and it started working again.
> 
> Thanks.  I have a few more questions (if you remember the answers ;-):
> 
> During this episode, was COMMAND.COM the only program you tried to 
> invoke?  If not, what were the other programs?

Yes.  It was hard coded as "COMMAND".  Just looking for a prompt...

> Also, I understand that the failed cases had one invalid directory as 
> part of PATH, but otherwise PATH was correct, right?  In particular, the 
> directory where COMMAND.COM resides was always part of the PATH, right?

Actually the failed cases had the path from your autoexec.bat that you
sent me to diagnose PMODE hangs under REDIR.  The *ONLY* direcory which was
valid (and like the 4th in the list 8) was C:\DOS

> "C:\DOS" is where you have COMMAND.COM on that machine, yes?

Yes.  But to make life more complicated, it was booted from a floppy.
I can't remember if COMSPEC had been reset to the hard disk or not.
I believe I was trying to run the image from floppy.

> Finally, the invalid directory that was part of PATH, was it pointing to 
> an existing drive or to an invalid drive?

Actually, there was an existing drive, then non-existant drive, existant
drive, etc...

> Btw, the ``normal'' failure mode of `system' is to print "Bad command or
> file name", the message which comes from COMMAND.COM when `system' gives
> up and passes the command line to it.  Is that what you saw?

I had a printf("Calling system") and printf("Return from system") around the
call and those were the only things I saw - nothing in between.

- Raw text -