Re: Dealing with empty aggregates?

["Bryn M. Reeves" <bmr at redhat dot com>]

On 05/19/2011 11:25 PM, Josh Stone wrote:
> We've debated empty-@sum behavior in the past, and IMO it makes perfect
> sense to let @sum(EMPTY) = 0.  Frank has agreed to let this change, with
> appropriate --compatible fallback behavior.  If there are no other
> objections, then I'll go ahead and relax this.

This would certainly suit my use case. I knocked together a quick and dirty
"nfsdtop" for a user a while back and since it worked so well I thought I'd take
another look and convert it to aggregates etc.

The reporting script would definitely be a lot simpler without the need to
special-case empty data sets.

Cheers.
Bryn.