Re: Should we optimize the `ORDER BY random() LIMIT x` case?

Nico,

> > Can the detection of such queries be done by the yacc/bison parser
> > grammar?
>
> Maybe the `sortby` rule could check if the expression is `random()`,
> then `sort_clause` could check if `$3` is a one-item `sortby_list` of
> just `random()` and mark `$$` as special -- this should be cheap, yes?
> We'd still need to check for `LIMIT` somewhere else.

Although partially implementing an optimizer on the parser level is
possible, it doesn't strike me as a right long-term solution. Firstly,
it's a hacky solution because maybe it will work for random() but for
some reason will not work for -random()*2. Secondly, imagine adding a
dozen optimizations like this in the upcoming years and all of them
interacting with each other. Imagine you are the person who has to
maintain this and not break things when adding another optimization.
All in all, this would be a poor design choice in my opinion.

--
Best regards,
Aleksander Alekseev