bug report: 5.0.14 all platforms, category/subject UNKNOWN

[Kenneth Brody]

Quoting John Esak (Thu, 2 Jun 2005 11:18:41 -0400):
[...quoting Nancy Palmquist...]
> > I have to say that I have seen this very behavior, but have made some
> > code changes and got stuff to work.  I just never had the time to fuss
> > around until I figured out the exact circumstances.
> >
> > My hat is off to you John for your attack of this problem.
[...]
> Thanks,
> Boy, it did take us some time... me for about an hour day before
> yesterday and Tony and I another hour or so yesterday. In any case...
> Ken rsponded to me privately that this VERY bug was reported after the
> release of 5.0.14 and therefore is ALREADY FIXED for the next version!
[...]

FYI - We haven't verified that this is the same issue, but it certainly
sounds like it.

The basic issue comes down to having DECLARE GLOBALs in the auto and
input tables, and then CALLing a table which DECLARE GLOBALs more
variables, such that the new declarations causes the total number of
global variables to cross a multiple of 32 boundary.  (ie: between
the auto and input, you have 50 DECLARE GLOBALs, and the CALLed table
has 20 new DECLARE GLOBALs, causing you to cross the boundary at 64.)

--
KenBrody at BestWeb dot net        spamtrap: <g8ymh8uf001 at sneakemail.com>
http://www.hvcomputer.com
http://www.fileProPlus.com