Re: Clarification of certain SQLSTATE class - Mailing list pgsql-hackers

From Tom Lane
Subject Re: Clarification of certain SQLSTATE class
Date
Msg-id 28998.1359072708@sss.pgh.pa.us
Whole thread Raw
In response to Clarification of certain SQLSTATE class  (gabrielle <gorthx@gmail.com>)
List pgsql-hackers
gabrielle <gorthx@gmail.com> writes:
> I'm working with some DB2 users, converting them to Pg, and I'm a bit
> confused about a certain class of SQLSTATE codes, specifically 02xxx "No
> data" (http://www.postgresql.org/docs/9.2/static/errcodes-appendix.html)

> As an example:  I enable %e in log_line_prefix so I can see the SQLSTATE
> values.  I run an UPDATE against a non-existent row, and find that my
> SQLSTATE value is 00000, indicating success.  I understand that this could
> be considered a 'success' because the query didn't actually throw an error;
>  but, based on the spec, I expected to see a SQLSTATE of 02000.

> I thought that Pg always generated an appropriate SQLSTATE code, and that
> it was just up to $client code to pick up that value, but that doesn't seem
> to be the case.  Is this part of the SQL standard we don't implement?

Yup.
        regards, tom lane



pgsql-hackers by date:

Previous
From: Dimitri Fontaine
Date:
Subject: Re: .gitignore additions
Next
From: Andres Freund
Date:
Subject: Re: autovacuum not prioritising for-wraparound tables