On Thu, Dec 6, 2012 at 1:36 PM, Tom Lane <tgl@sss.pgh.pa.us> wrote:
> I'm against putting I/O coercion semantics into tupconvert, period. Ever.
> If plpgsql wants that behavior rather than something more consistent
> with the rest of the system, it needs to implement it for itself.
I'm sure that can be done. I don't think anyone is objecting to that,
just trying to get useful behavior out of the system.
Are you going to commit a stripped-down version of the patch?
--
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company