Re: [HACKERS] path toward faster partition pruning

Amit Langote wrote:

> I will continue working on improving the comments / cleaning things up and
> post a revised version soon, but until then please look at the attached.

I tried to give this a read.  It looks pretty neat stuff -- as far as I
can tell, it follows Robert's sketch for how this should work.  The fact
that it's under-commented makes me unable to follow it too closely
though (I felt like adding a few "wtf?" comments here and there), so
it's taking me a bit to follow things in detail.  Please do submit
improved versions as you have them.

I think you're using an old version of pg_bsd_indent.

In particular need of commentary
 * match_clause_to_partition_key() should indicate which params are
 output and what do they get

 * get_steps_using_prefix already has a comment, but it doesn't really
 explain much.  (I'm not sure why you use the term "tuple" here.  I mean,
 mathematically it probably makes sense, but in the overall context it
 seems just confusing.)

-- 
Álvaro Herrera                https://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services