Path: utzoo!attcan!uunet!husc6!cmcl2!nrl-cmf!ames!pasteur!ucbvax!hplabs!hpda!hpcuhb!hpindda!wag
From: wag@hpindda.HP.COM (William Gilliam)
Newsgroups: comp.sys.ibm.pc
Subject: Re: dBase III+ bug?
Message-ID: <4330106@hpindda.HP.COM>
Date: 3 Sep 88 18:56:17 GMT
References: <16834@vax5.CCS.CORNELL.EDU>
Organization: HP Information Networks, Cupertino, CA
Lines: 16

I have been in similar situations where Dbase will simply hang.  Once, when
I called Ashton-Tate to describe the problem and explain what I was doing in
my procedure files, I was sure that they were going to tell me that it was
a bug in *their* code because it took them days to call me back.  But, when
they did call, they said that either (1) I had somehow arranged my index
files into a knot where they all ended up pointing to each other's addresses,
(2) I was exceeding the program capacity, or (3) my copy of Dbase was
corrupted.  I didn't accept any of those reasons (except for the possibility
of a bad version of Dbase), and after checking my code further, I found the
culprit:  One of my index files depended on a field that, in certain cases 
that I neglected, would have a null value.  Dbase seems to frown on things
like that, but I hear that Dbase IV contains routines to handle such things.
We shall see...


WG