On Wed, Oct 13, 2010 at 10:14 AM, Tom Lane <tgl@sss.pgh.pa.us> wrote:
> It's possible that at some point we'll try to introduce plan caching
> for non-inlined SQL functions.
hm, I think the search_path/function plan issue would have to be dealt
with before doing this -- a while back IIRC you suggested function
plans might be organized around search_path setting at plan time, or
this would break a fair amount of code (for example, mine) :-).
merlin