On 16 January 2012 23:53, Tom Lane <tgl@sss.pgh.pa.us> wrote:
> Well, short of seeing an acceptable patch for the larger thing, I don't
> want to accept a patch to add that field to Const, because I think it's
> a kluge. I'm still feeling that there must be a better way ...
What does an acceptable patch look like? Does your objection largely
hinge on the fact that the serialisation is performed after the
re-writing stage rather on the raw parse tree, or is it something
else?
Despite my full plate this commitfest, I am determined that this
feature be available in 9.2, as I believe that it is very important.
Instrumentation of queries is something that it just isn't possible to
do well right now, with each of the available third party solutions or
pg_stat_statements. That really needs to change.
--
Peter Geoghegan http://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Training and Services