Re: Opening view that uses a function - empty column

David J -

Thanks for the pointers.  I'm sure you can tell that these are (nearly) the first PG procedures that I have written.  I
didthe suggested refactoring. 

Could you elaborate a little on the "identity" problem?
There is only one set of schemas on the server against which I am testing.  I am using the same credentials to log in
viaPGAdmin (local and remote) and Java. 
The only non-default configuration I have for PG (that I can think of now) are a few entries in pg_hba.conf as follows:
host    all             all             127.0.0.1/32            md5
host    postgres        postgres        173.20.10.3/32          md5
hostssl    postgres    postgres    127.0.0.1/32        md5
hostssl postgres        postgres        173.20.10.3/32          md5

I appreciate all your help,
Brady

On Monday, July 30, 2012 2:22:04 PM UTC-6, "David Johnston" wrote:
> Couple of points - though they won't help the current question probably:
>
>
>
> >Brady's original
>
> ---------------my responses embedded
>
>
>
>
>
> > -----Original Message-----
>
> > From: pgsql-jdbc-owner@postgresql.org [mailto:pgsql-jdbc-
>
> > owner@postgresql.org] On Behalf Of Brady Mathis
>
> > Sent: Monday, July 30, 2012 1:38 PM
>
> > To: pgsql-jdbc@postgresql.org
>
> > Subject: Re: [JDBC] Opening view that uses a function - empty column
>
> >
>
> > Oh!  And, I made another observation while cracking away at this:  When I
>
> > connect to the DB using PGAdmin (1.14.3) from a remote workstation as
>
> > opposed to on the DB server itself, I see the same problem with the empty
>
> > external_id column when executing "select * from rhs.encounter;".
>
>
>
> ----------------Are you sure that when you connect to the server from
>
> "...the DB server itself" you are connecting to the same database as when
>
> you do so remotely?
>
>
>
> ----------------This really feels like an identity issue and not anything
>
> specifically related to the code you are showing us.  That said...see below.
>
>
>
> >
>
> > Could this be a problem with how I have implemented my functions?  There
>
> > are two  functions calls actually used to fill the column - I included the
>
> > complete text of the functions below.
>
> >
>
> > get_valid_xml - ensures that content from pg_largeobject is valid xml
>
> > extract_from_extended - uses xmlparse and xpath to get one specific value
>
> >
>
> > Thanks...again!
>
> >
>
> > /* Function to validate xml for xpath use in SQL query */ CREATE OR
>
> REPLACE
>
> > FUNCTION rhs.get_valid_xml(x text)
>
> >   RETURNS xml AS
>
> > $BODY$
>
> > BEGIN
>
> >     PERFORM XMLPARSE( DOCUMENT x );
>
> >         RETURN XMLPARSE( DOCUMENT x );
>
>
>
> ------------It is unnecessary to call XMLPARSE twice, especially since you
>
> are doing so on the "no exception" code path.
>
>
>
> >     EXCEPTION WHEN OTHERS THEN
>
> >         RETURN XMLPARSE( DOCUMENT '<?xml version="1.0"
>
> > encoding="UTF-8"?><fields></fields>' ); END; $BODY$
>
>
>
>
>
> ------------Whenever you are having difficulties you should avoid ignoring
>
> exceptions.  At worse perform a RAISE NOTICE when one occurs it you still
>
> want to send back an empty document.  However, raising the exception is also
>
> a valid and useful action.  If you are debugging then raising an exception
>
> (and reviewing it) is even more worthwhile.
>
>
>
>
>
> >   LANGUAGE plpgsql VOLATILE
>
> >   COST 100;
>
> > ALTER FUNCTION rhs.get_valid_xml(text) OWNER TO postgres;
>
> >
>
> >
>
> > /* Function to allow extract of fields from _extended */ CREATE OR REPLACE
>
> > FUNCTION rhs.extract_from_extended(field_name text, menu_data_id
>
> > bigint)
>
> >   RETURNS text AS
>
> > $BODY$
>
> >     select translate( xpath('//field[@name="'||$1||'" and
>
> > @type="String"]/text()', xmlparse(document ( rhs.get_valid_xml(
>
> > array_to_string( array(select lo.data from app.menu_data md,
>
> > pg_catalog.pg_largeobject lo where md.id = $2 and  md.xml01 = lo.loid),
>
> '') )
>
> > )))::text, '{}', ''); $BODY$
>
>
>
>
>
> ------------Here you call XMLParse/Document on the result of
>
> "get_valid_xml(...)";  this seems redundant since "get_valid_xml(...)"
>
> already does this.
>
>
>
>
>
> >   LANGUAGE sql VOLATILE
>
> >   COST 100;
>
> > ALTER FUNCTION rhs.extract_from_extended(text, bigint) OWNER TO
>
> > postgres;
>
>
>
>
>
>
>
>
>
>
>
> --
>
> Sent via pgsql-jdbc mailing list (pgsql-jdbc@postgresql.org)
>
> To make changes to your subscription:
>
> http://www.postgresql.org/mailpref/pgsql-jdbc