"Frederick Ross" <madhadron@gmail.com> writes:
> But say I have a table with an integer column. If I had a pure SQL
> function, I could call
> select twohundreds(i) from table;
> and get a properly concatenated list of the rows resulting from
> applying twohundreds to each row of table. This is deprecated
> according to the documentation, and it doesn't work at all with
> PLPGSQL, so how should I go about doing it?
I believe you can make an end-run around the implementation restriction
by creating a SQL-language function that's a wrapper around your plpgsql
function.
regards, tom lane