Horton

Re: [HACKERS] Surjective functional indexes

Alvaro Herrera <alvherre@alvh.no-ip.org>

View thread View on postgr.es
From: Alvaro Herrera <alvherre@alvh.no-ip.org>
To: Konstantin Knizhnik <k.knizhnik@postgrespro.ru>
Cc: Stephen Frost <sfrost@snowman.net>, Michael Paquier <michael.paquier@gmail.com>, Simon Riggs <simon@2ndquadrant.com>, Christoph Berg <myon@debian.org>, Thomas Munro <thomas.munro@enterprisedb.com>, Peter Eisentraut <peter.eisentraut@2ndquadrant.com>, PostgreSQL Hackers <pgsql-hackers@postgresql.org>
Date: 2018-03-22T20:53:53Z
Lists: pgsql-hackers 
The whole IsProjectionFunctionalIndex looks kinda bogus/ugly to me.  Set
the boolean to false, but keep evaluating anyway?  But then, I thought
the idea was to do this based on the reloption, not by comparing the
expression cost to a magical (unmodifiable) value?

In RelationGetIndexAttrBitmap(), indexattrs no longer gets the columns
corresponding to projection indexes.  Isn't that weird/error
prone/confusing?  I think it'd be saner to add these bits to both
bitmaps.

Please update the comments ending in heapam.c:4188, and generally all
comments that you should update.

Please keep serial_schedule in sync with parallel_schedule.

Also, pgindent.

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

Commits

  1. Finish reverting "recheck_on_update" patch.