> - Prove IS NOT NULL inference for large arrays > > NR: No idea.
As the fairly new author of this patch, my perspective is that this patch got quite a bit of review, albeit without a formal "yes or no" response. I'm obviously interested in getting it committed, and I believe it's a fairly simple patch to look at (though since it's in predtest probably requires extra brain cycles to be careful we don't make spurious assumptions in the optimizer). It's also an obvious performance win for queries that can use partial indexes with almost no additional optimizer overhead. But I'm also interested in feedback on how patches like this work in the review process -- particularly when given the fair amount of discussion/cleanup but without a final word. As a new patch author a lot of understanding how this works feels very much like a "learn as you go", but since there's not a lot of information directly written on it I figured asking explicitly is the best way to learn the process better. Thanks, James Coleman