Sender: tgl@mythos.jpl.nasa.gov
Message-ID: <3618195E.6FE14EF6@alumni.caltech.edu>
Date: Mon, 05 Oct 1998 00:57:02 +0000
From: "Thomas G. Lockhart" <lockhart@alumni.caltech.edu>
Organization: Caltech/JPL
MIME-Version: 1.0
To: Tom Lane <tgl@sss.pgh.pa.us>
CC: pgsql-hackers@postgreSQL.org, pgsql-interfaces@postgreSQL.org
Subject: Re: [HACKERS] TCL_ARRAYS code in libpgtcl is pretty seriously broken
References: <19449.907539856@sss.pgh.pa.us>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

> 1. Blithely assumes that any data value beginning with '{' and ending
> with '}' must represent an array value.  Should have some more robust
> way of discovering whether a column is array type.  (In fairness, this
> might require a FE/BE protocol change, unless arrayness can be
> determined from the tuple descriptors provided by the backend, ie,
> field type OID, size, and attmod.  Anybody know a way to do that?)
> Comments?

Postgres seems to use a convention that a type name which starts with an
underscore is the array type for the corresponding non-underscore,
non-array type. Also, the typelem field in pg_type is non-zero for array
types. 

This isn't a definitive answer and there may be another way to discover
array-ness but it's where I would look. Not sure if you'd be happy
having to do a select on pg_type for every query unless you're doing it
already...

                    - Tom