SLetovsky(at)aol(dot)com writes:
> Am I correct in interpreting your comments as saying you believe that
> if we could lose the OR and the strand constraint PG would probably
> use the index properly?
No, I said I thought it could do it without that ;-). But yes, you'd
have a much less fragile query if you could lose the OR condition.
Have you looked into using a UNION ALL instead of OR to merge the two
sets of results? It sounds grotty, but might be faster...
regards, tom lane