On Tue, Sep 3, 2013 at 9:13 AM, Tom Lane <t...@sss.pgh.pa.us> wrote: > Andres Freund <and...@2ndquadrant.com> writes: >> On 2013-09-03 08:59:53 -0500, Merlin Moncure wrote: >>> While playing around with Andres's trick, I noticed that it works but >>> will not match against operators taking "any" although those will >>> match with explicit schema declaration (FWICT it goes through the >>> search_path trying to explicitly match int/int operator then goes >>> again matches "any"). That's pretty weird: > >> Not surprising. We look for the best match for an operator and >> explicitly matching types will be that. If there were no operator(int, >> int) your anyelement variant should get called. > > Yeah, this has exactly nothing to do with operator precedence. > Precedence is about which operator binds tighter in cases like "A+B*C".
That all makes perfect sense -- thanks guys. For posterity, Andres's trick worked and did end up saving me some coding after all -- in my case I have to eval() some externally generated fairly complex expressions in SQL (via pl/pgsql EXECUTE) in the context of a much larger query. My de-parsing code ended up having bugs and it was much easier to tip-toe around the search_path (via SET LOCAL) and force the modified operator. merlin
