Re: PATCH: complist with long display lines

[Peter Stephenson <pws@csr.com>]

Bart Schaefer wrote:
> How did you ever track this down?

That's three evenings which will never come again.

Once I'd found that array indexing for mgtab was off, which took a
certain amount of digging since it's referenced in some odd ways, it
wasn't so hard to find out how the limits were being set and how the
assumption in the line limit was being violated.

As I indicated, whether the assumption is present at all and only the
places where I've applied the fix is another matter to be determined...

> It may be worthwhile (or at least satisfying to paranoid types) to add
> the following patch, which I've had sitting around uncommitted because
> it failed to resolve the addressing issues that you seem to finally
> have identified.  This patch prevents the infinite loop alluded to in
> 21842, by catching cases where an array index starts out negative and
> decrements from there.  It might be unnecessary now, but:

I don't think I've done anything which would prevent this infinite loop.
mcols comes from the global columns which comes from the environment, so
if that's negative there are problems.  So I think this would be useful.

-- 
Peter Stephenson <pws@csr.com>                  Software Engineer
CSR PLC, Churchill House, Cambridge Business Park, Cowley Road
Cambridge, CB4 0WZ, UK                          Tel: +44 (0)1223 692070


To access the latest news from CSR copy this link into a web browser:  http://www.csr.com/email_sig.php