Re: Tablefunc crosstab error messages

From: Noah Misch <noah(at)leadboat(dot)com>
To: Mali Akmanalp <mali(at)akmanalp(dot)com>
Cc: pgsql-hackers(at)postgresql(dot)org
Subject: Re: Tablefunc crosstab error messages
Date: 2012-10-03 16:21:20
Message-ID: 20121003162120.GA23518@tornado.leadboat.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

On Mon, Aug 27, 2012 at 08:07:02PM -0400, Mali Akmanalp wrote:
> Returning the type information to the caller seems like a pain
> but compatCrosstabTupleDescs already has instances in it where it fails
> with an error message, so I propose we just do that and tell the user the
> expected and actual types here, instead of returning false here and
> throwing a generic error message in the caller.
>
> What do you think? Let me know so I can write up a patch for you guys.

That would prove helpful, +1 from me.

> Also, just curious, is the reason the query passed into crosstab is a
> string (rather than an SQL expression) that it's just easier to implement
> that way?

Yes. Accepting the query as proper SQL syntax would require modifying the
backend grammar. Extensions cannot do that, but they can define functions
accepting a string and run it as a SQL command.

In response to

Browse pgsql-hackers by date

  From Date Subject
Next Message Tom Lane 2012-10-03 16:26:42 Re: do we EXEC_BACKEND on Mac OS X?
Previous Message Boszormenyi Zoltan 2012-10-03 16:15:58 Re: [PATCH] Make pg_basebackup configure and start standby [Review]