Yeah, a better error message would probably be a good idea.
> (Should there be an option to non-fatally use the 'medium' packsize on
> 32-bit systems when the 'large' is requested, or is that just delaying
> the problem of running out of address space so that a lot of time is
> wasted?)
Well, on a 32-bit system, the current architecture can only handle a
2G datafile. There are ways of increasing that using relative instead
of absolute seeking, but that would require quite the change in how
seeking happens. I suppose the best idea would be a better way of
detecting that 'Q' isn't available (using eval{}) and throwing a
warning if 'large' is selected and degrading to 'medium'.
And, now that I think about it, doing the same thing when opening a
file on 32-bit systems that was created with 'large' (throwing a fatal
error) would be a good idea. Thanks for bringing this up.
--
Thanks,
Rob Kinyon
Thanks,
Rob
--
Thanks,
Rob Kinyon
QM - Does this problem still exist?
Rob