Re: Wrong results from in_range() tests with infinite offset

I wrote:
> When the current row's value is +infinity, actual computation of
> base - offset would yield NaN, making it a bit unclear whether
> we should consider -infinity to be in-range.  It seems to me that
> we should, as that gives more natural-looking results in the test
> cases, so that's how the patch does it.

Actually, after staring at those results awhile longer, I decided
they were wrong.  The results shown here seem actually sane ---
for instance, -Infinity shouldn't "infinitely precede" itself,
I think.  (Maybe if you got solipsistic enough you could argue
that that is valid, but it seems pretty bogus.)