Re: Ad Hoc Indexes

Hans-Juergen Schoenig wrote:
> this kind of indexing makes sense if you visit the same data over and  
> over again. WITH-queries would be an example for that and self joins  
> could benefit from the this feature too.

Yes, for joins it is useful.  We have hash joins, which are kind of ad hoc 
hash indexes.  If anyone wants to implement a btree join, that might work. :)  
But I think we essentially support what the OP is asking for, it is just 
under different labels.

-- 
Peter Eisentraut
http://developer.postgresql.org/~petere/