Ken Tanzer <ken.tanzer@gmail.com> writes:
> Getting a build environment together seemed more painful, so here's a test
> case.
Wow, that's pretty broken. I'll dig into it tomorrow.
> Just for fun, I tried this in another database on a different
> machine (and with 9.0.08). I got the same results, so it doesn't seem to
> be a case of something wacky with my particular database.
Yeah, seems to fail the same way in all active branches ...
> p.s., Not your problem I know, but I need to deal with this somehow and
> rather soon.
If you just need a work-around-it-right-now solution, I'd suggest
introducing an "OFFSET 0" optimization fence into one or another of the
levels of view below the outer joins. I've not experimented but I think
that ought to fix it, at some possibly-annoying cost in query
optimization. Hopefully I'll have another answer tomorrow.
regards, tom lane