>>>>> "Tom" == Tom Lane <tgl@sss.pgh.pa.us> writes:
[...]
Tom> so that a finite value should never map to INT[64]_MIN, making it
Tom> safe to do as you suggest. I agree that distinguishing +Inf from
Tom> NaN is probably more useful than distinguishing it from the very
Tom> largest class of finite values, so will do it as you suggest.
Tom> Thanks!
It would make sense to make sure there's a test case in which at least
one value of all three of: a finite value much greater than 10^332, a
+Inf, and a NaN were all present in the same sort, if there isn't one
already.
--
Andrew (irc:RhodiumToad)
