On Tue, Sep 05, 2006 at 02:48:45PM +0100, Gregory Stark wrote:
>
>
> So I'm thinking again about the problems with fixed length data types not
> having typmod available when they would need it. But I'm having trouble
> finding enough old posts to get a handle on exactly what the problem is.
Like Tom said, the problem is you don't have it. In the specific case
of type input functions, what typmod is the output? For type output
functions relying on a passed typmod is a security risk.
So you end up storing the typmod in the Datum itself, which brings you
right back to varlena.
> This would make a nice test of the new wiki. I would be indebted to whoever
> could summarize the root of the problem and explain exactly what circumstances
> the typmod is unavailable. I would summarize the responses and put them up on
> the wiki.
Well, the root of the problem depends on your perspective. If the
purpose behind all of this is to save disk space, perhaps the root of
the problem is that disk representation and memory representation are
intimately tied?
Have a nice day,
--
Martijn van Oosterhout <kleptog@svana.org> http://svana.org/kleptog/
> From each according to his ability. To each according to his ability to litigate.