Re: 64-bit API for large objects

On Sat, Sep 24, 2005 at 12:13:11PM -0400, Tom Lane wrote:
> David Fetter <david(at)fetter(dot)org> writes:
> > On Fri, Sep 23, 2005 at 05:40:09PM -0400, Tom Lane wrote:
> >> For that matter, we can't even guarantee that they work at all: not
> >> all platforms even *have* int64 types.
>
> > What platforms that PG supports don't have int64 arithmetic?
>
> We claim to build with any ANSI C compiler, and there is no requirement
> for a 64-bit type in ANSI C.
>
> The historical project policy is that we should still build without
> such a type, and everything should still work except that the effective
> bounds of bigint data correspond to int32 instead of int64 limits.
> I see no reason to back off that policy. It's not very much harder
> to do it right.

So what happens if you attempt to put a value greater than 2^32 into a
bigint on a non-int64 platform?

I would argue that by default we should not allow users to even create
bigints on any platform where bigint = int. And if the default is
overridden, we should still throw a warning.
--
Jim C. Nasby, Sr. Engineering Consultant jnasby(at)pervasive(dot)com
Pervasive Software http://pervasive.com work: 512-231-6117
vcard: http://jim.nasby.net/pervasive.vcf cell: 512-569-9461