Re: sqlstate 02000 while declaring cursor/freeing statement - Mailing list pgsql-admin

From Michael Fuhr
Subject Re: sqlstate 02000 while declaring cursor/freeing statement
Date
Msg-id 20050830151320.GA56850@winnie.fuhr.org
Whole thread Raw
In response to Re: sqlstate 02000 while declaring cursor/freeing statement  ("Thomas F. O'Connell" <tfo@sitening.com>)
Responses Re: sqlstate 02000 while declaring cursor/freeing statement
List pgsql-admin
On Tue, Aug 30, 2005 at 09:01:36AM -0500, Thomas F. O'Connell wrote:
> On Aug 29, 2005, at 2:19 PM, andy rost wrote:
> >
> >    $declare loop1 cursor with hold for
> >    select distinct ( tabname )
> >    from meta ;
> >
> >results in sqlca.sqlcode = 100 and sqlca.sqlstate = '02000'
>
> [snip]
>
> I don't see anything wrong with your cursor declaration, so this just
> looks like it's indicating that no rows were returned.

See my response to the same thread in pgsql-sql about what's
happening:

http://archives.postgresql.org/pgsql-sql/2005-08/msg00274.php

The ECPG code generator doesn't issue the DECLARE statement until
the cursor is opened, so checking sqlcode/sqlstate immediately after
DECLARE but before OPEN will get the error code values from a
previous command.

Here's an excerpt from the DECLARE documentation:

  The PostgreSQL  server does not implement an OPEN statement for
  cursors; a cursor is considered to be open when it is declared.
  However, ECPG, the embedded SQL preprocessor for PostgreSQL,
  supports the standard SQL cursor conventions, including those
  involving DECLARE and OPEN statements.

Since declaring a cursor in PostgreSQL also opens it, ECPG apparently
defers doing anything with the cursor until it finds the OPEN
statement (i.e., EXEC SQL DECLARE generates no code, and EXEC SQL
OPEN generates a DECLARE).

--
Michael Fuhr

pgsql-admin by date:

Previous
From: Michael Fuhr
Date:
Subject: Re: plpgsql question
Next
From: "Joel Fradkin"
Date:
Subject: connections running out