Re: BUG #14545: Unable to retrieve parameter mode 'VARIADIC'

From: "David G(dot) Johnston" <david(dot)g(dot)johnston(at)gmail(dot)com>
To: Prajval Poojari <ppoojari(at)erwin(dot)com>
Cc: "pgsql-bugs(at)postgresql(dot)org" <pgsql-bugs(at)postgresql(dot)org>
Subject: Re: BUG #14545: Unable to retrieve parameter mode 'VARIADIC'
Date: 2017-02-14 22:43:48
Message-ID: CAKFQuwaxvr1p7chJd2BGPpkYg0VzL4LVxnpEctLSsMTSaYgvOQ@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-bugs

On Tue, Feb 14, 2017 at 3:15 PM, Prajval Poojari <ppoojari(at)erwin(dot)com> wrote:

> Please let me know your comments
>

I don't have access to an authoritative definition for what the allowed
values for information_schema.parameters.parameter_mode are.​

The fact that our "VARIADIC" mode maps to "IN" on the information schema is
not accidental. The question is whether it is out-of-date. Tom Lane
committed that particular change back in 2008 (d89737d3) though the file
header suggests it conforms to the ISO/IEC 9075-11:2011 specification.

Is there a reason you don't just use the PostgreSQL catalogs? Are you
aware of other implementations of information schema that have this level
of detail?

Changing the value from "IN" to "VARIADIC" would be a breaking change for
systems that only care about the "IN" property of the argument. Even
matching the standard on this topic might not be warranted given 9 years of
the present behavior. Without even that benefit the hurdle to change this
is extremely high - especially when pg_proc is authoritative from our
internal perspective.

David J.

In response to

Responses

Browse pgsql-bugs by date

  From Date Subject
Next Message Prajval Poojari 2017-02-14 22:58:45 Re: BUG #14545: Unable to retrieve parameter mode 'VARIADIC'
Previous Message Prajval Poojari 2017-02-14 22:15:33 Re: BUG #14545: Unable to retrieve parameter mode 'VARIADIC'