Re: Dynamic PL/pgSQL select query: value association propblem

Hi

2018-02-16 13:20 GMT+01:00 Thiemo Kellner <thiemo(at)gelassene-pferde(dot)biz>:

> Hi all
>
> I would like to have a generic trigger function that compares on insert if
> there is already a record in the table with the very same values. Using
> PL/pgSQL ( I am not bound to that) I know the insert record structure from
> the new record and I can build a select query dynamically from the
> catalogue, but I do not know to associate the new record values to the
> corresponding columns. An example
> Table T has columns Q and L, in that order. If I create an insert trigger
> function I have the new values in new.L and new.Q. From the catalogue I can
> create a the select query S_QUERY:
> select count(*) > 0 from T where A = $1 and B = $2. But when I want to
> EXECUTE S_QUERY USING it fails because I cannot use something like NEW[1].
>
>
Why you don't create query like

EXECUTE 'SELECT xxx FROM TAB WHERE A = $1.x AND B = $1.y' USING NEW;

I don't understand tou your case, but usually count(*) > 0 looks like
antipattern - probably you want to use EXISTS(...)

Regards

Pavel

> Is there a way to convert the record type into an array type? Or is there
> even a way to do it more directly like WHERE T.RECORD = NEW?
>
> Kind regards Thiemo
>
> --
> Öffentlicher PGP-Schlüssel: http://pgp.mit.edu/pks/lookup?
> op=get&search=0x8F70EFD2D972CBEF
>
> ----------------------------------------------------------------
> This message was sent using IMP, the Internet Messaging Program.
>
>