Michael Paquier

Thanks for the patch. The patch you are proposing for the additional tests has no explanation about what's being done, like the interactions between the non-existing indexes and the hints....

I have spent some time studying the various patterns as well as the influence of parallel_aware here, and I am unclear what you claim is a bug here. Could you...

This has been fixed as of a9863af.

This does not seem to be relevant anymore, hence closing for now.

Looks like there is nothing left to do here, so closing.

Would you like to write a patch for it?

Yeah, this would be a new feature in itself. I don't think that anybody has bothered with this case yet, would you be interested in writing a patch?

I'll look at producing a patch about that. For the time being, let's give a poke at @rjuju for the interactions with hypopg.

> I agree with @HUST-Huyajun, checking and ignoring hypothetical indexes is likely the only thing that can be done here. Yep. I'm on the same line as you here. Ignoring...

The trick with explain_get_index_name_hook() does not seem to address all that properly: we still have problems with EXPLAIN plans where hypothetical indexes have names that do not show up correctly....