Tom Lane wrote:
> Jan Wieck <JanWieck@yahoo.com> writes:
> > This isn't local recompilation in current backend. It's
> > recreation of the pg_rewrite entry for a relation, including
> > propagation.
>
> Where I'd like to go (see my previous mail) is that pg_rewrite,
> pg_attrdef, and friends store *only* the source text of rules,
> default expressions, etc. No compiled trees at all in the database.
> So there's no need to update the database entries, but there is a
> need for something like a shared-cache-invalidation procedure to cause
> backends to recompile things that depend on updated relations.
Hmmm,
are you sure that this doesn't have a severe performance impact?
When and how often are these parsetrees read? IIRC these parsetree strings are interpreted somehow during
heap_open(). Now you want to run a flex/bison plus tons of syscache lookups for operator and function
candidatesand possible casting in this place?
Jan
--
#======================================================================#
# It's easier to get forgiveness for being wrong than for being right. #
# Let's break this rule - forgive me. #
#================================================== JanWieck@Yahoo.com #
_________________________________________________________
Do You Yahoo!?
Get your free @yahoo.com address at http://mail.yahoo.com
