From: | Andreas Pflug <pgadmin(at)pse-consulting(dot)de> |
---|---|
To: | Donald Fraser <demolish(at)cwgsy(dot)net> |
Cc: | "[pgADMIN]" <pgadmin-support(at)postgresql(dot)org> |
Subject: | Re: pgAdmin III - bug with reverse engineered SQL |
Date: | 2003-07-31 12:36:19 |
Message-ID: | 3F290D43.4080200@pse-consulting.de |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgadmin-support |
Donald Fraser wrote:
> pgAdmin III July 16th Build
> PostgreSQL 7.3.3
>
> The reverse engineered SQL for indexes that use a function do not work.
> For example I create an index with the following command:
> CREATE UNIQUE INDEX tbl_security_fullname_key
> ON tbl_security
> USING btree (get_securityname_4idx(s_umbname, s_name, s_classname, id));
> pgAdmin III produces:
> CREATE UNIQUE INDEX tbl_security_fullname_key
> ON public.tbl_security
> USING btree (public.get_securityname_4idx(s_umbname::citext,
> s_name::citext, s_classname::citext, id::int4) citext_ops);
> The problem exists with the appended data type on the column names as
> the parameters of the function.
> If I try to execute pgAdmin's version I get the following error message:
> ERROR: parser: parse error at or near "::" at character 220.
>
>
Donald,
that citext_ops is the operator class, which is correct according to the
7.3.3 doc. Additionally, the parse error is reported at "::". Please try
to trace this down, I don't have a 7.3 server running any more. Please
attach a "select pg_get_viewdef(OID_of_the_Index)" output, so we can see
what pgsql likes to see.
Regards,
Andreas
From | Date | Subject | |
---|---|---|---|
Next Message | Donald Fraser | 2003-07-31 13:27:54 | Re: pgAdmin III - bug with reverse engineered SQL with function indexes |
Previous Message | Donald Fraser | 2003-07-31 11:33:25 | pgAdmin III - bug with reverse engineered SQL with function indexes |