You don't need to; PGclear is called automatically some time after the Result is garbage collected. This is done by using PQclear as a finalizer on the foreign pointer that represents a Result: http://hackage.haskell.org/packages/archive/postgresql-libpq/0.8.2/doc/html/... There is "unsafeFreeResult", which triggers the finalizer at a deterministic point in time. But I don't recommend it's use unless you really understand the basics of GHC's implementation : you can easily have a reference to the Result sitting around in a thunk even "after" you have computed some value from the Result and thrown away all other references. (Because, in fact, you haven't computed the value completely; some part of that value is actually a thunk that needs the Result to compute the rest of the value.) And even then, I'm skeptical that very many applications would truly benefit from it's use. Best, Leon On Fri, Jun 8, 2012 at 3:03 PM, Kirk Peterson <necrobious@gmail.com> wrote:

Hi Leon,

I'm working with your haskell postgresql-libpq package and noticed that the examples form: http://www.postgresql.org/docs/9.1/static/libpq-example.html release a Result using the PGclear() function, however there is no clear function in postgresql-libpq.

I'm new to libpq in general, so perhaps this is an easy question: what is the idiomatic way to clear a result using postgresql-libpq?

cheers, -kirk

-- ⑆gmail.com⑆necrobious⑈