Re: Oid registry

From: Peter Eisentraut <peter_e(at)gmx(dot)net>
To: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
Cc: Andrew Dunstan <andrew(at)dunslane(dot)net>, PostgreSQL-development <pgsql-hackers(at)postgresql(dot)org>
Subject: Re: Oid registry
Date: 2012-09-25 21:53:04
Message-ID: 506227C0.5080605@gmx.net
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

On 9/24/12 11:39 PM, Tom Lane wrote:
> My recollection of the PGCon discussion is that people wanted to allow
> client-side code to hard-wire type OIDs for add-on types, in more or
> less the same way that things like JDBC know that "25" is "text".

If I write a custom uint type and want to explain to JDBC that it is
number-like, I don't think it's very attractive to have to register a
hard-wired OID for it. There would not only be the burden of
maintaining this OID database, JDBC and every driver would on the other
hand have to maintain its own database of all custom types in the world.

I would rather imagine a system where the type communicates its
properties to the client using some sort of label system, like "number",
"string", "pair of number" (point), "pair of string" (hstore). Or
something with more or less detail. Maybe something like that already
exists (maybe somewhere in thrift, protocol buffers, etc.?).

In response to

Browse pgsql-hackers by date

  From Date Subject
Next Message Peter Eisentraut 2012-09-25 21:54:28 Re: Oid registry
Previous Message Peter Eisentraut 2012-09-25 21:45:14 Re: Oid registry