Re: Common Table Expressions (WITH RECURSIVE) patch

Gregory Stark <stark(at)enterprisedb(dot)com> writes:
> Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> writes:
>> Yeah, I'd been running into that issue too. Adding an explicit pointer
>> to the CTE into the RTE doesn't work because it renders the parse tree
>> un-copiable (at least without something a lot more sophisticated than
>> copyObject; and saving/loading rule parsetrees would be tough too).

> Well the alternative to direct pointers is as you did with subqueries, turning
> the set into a flat array and storing indexes into the array. I'm not sure if
> that applies here or not.

I think that just changes the problem into "where can I find the array?" ...

The real issue here is that simplicity of copying etc requires that
child nodes in a parse tree never have back-links leading up to their
parent. If we were willing to drop that requirement for Query then
we'd not need any auxiliary data structures to chase up to upper-level
rtables or CTEs. I'm not sure this cure is better than the disease
though.

regards, tom lane