> * Make CLUSTER error out if the target index is not of an 'amindexnulls'
> index AM. This would amount to restricting CLUSTER to b-trees, which is
> annoying.
I think this solution is fine - we just need to fix GiST to index nulls
one day :)
> It occurs to me also that the same kind of pitfall exists for partial
> indexes: cluster on a partial index, you lose. However, I don't have
> a problem with simply refusing to cluster on partial indexes.
No problem with that...
> Comments? Any other ideas out there?
Is there any conceivable gain in doing a CLUSTER over a tsearch index?
Surely it's basically randomly accessed?
Chris