support extending indexscan,heapscan and customizing plan

Hi,

we currently want to employ PG and Gin Index for light-weight full text
search engine. however,

we encounter a problem that Gin Index is not quick enough for relevence,
because it must

scan all heaptuples which statisfy the condition, then invoke ts_rank to
compute the rank of tuple.

an idea comes to us: add some foundamental ranking features to the Gin index
structure;

then we can find the most relevance tuples just from index, which result in
a problem that we must

add a type of indexscan and heapscan,customize plan. but PG don't support
the behaviors,.

As far as I know, oracle support customizing plan. My question is how can i
slove it, and whether PG

will support extending indexscan and heapscan,customizing plan in future.

Thanks

--zhuxiang

Yahoo! CN