Native partitioning tablespace inheritance - Mailing list pgsql-hackers

From David G. Johnston
Subject Native partitioning tablespace inheritance
Date
Msg-id CAKFQuwb3Vg+A9cJOwJSksoPn2j7mD_fVNqKKnbdkyfjREY6qVQ@mail.gmail.com
Whole thread Raw
In response to Re: Native partitioning tablespace inheritance  (Robert Haas <robertmhaas@gmail.com>)
List pgsql-hackers
On Thursday, April 12, 2018, Robert Haas <robertmhaas@gmail.com> wrote:
On Thu, Apr 12, 2018 at 2:40 PM, Jonathan S. Katz
<jonathan.katz@excoventures.com> wrote:
> If there are no strong objections I am going to add this to the “Older Bugs”
> section of Open Items in a little bit.

I strongly object.  This is not a bug.  The TABLESPACE clause doing
exactly what it was intended to do, which is determine where all of
the storage associated with the partitioned table itself goes.  It so
happens that there is no storage, so now somebody would like to
repurpose the same option to do something different.  

The part about accepting an option that is basically invalid is reasonably bug-like.  Having tablespace and partition by clauses be mutually exclusive would be worthy of fixing though it couldn't be back-patched.  Documentation is good but outright prevention is better.

If we can't agree on the future behavior we should at least prevent the existing situation in v11.  I'm doubting whether redefine behavior of the existing option to anything other than an error would be acceptable.

David J.


pgsql-hackers by date:

Previous
From: "Jonathan S. Katz"
Date:
Subject: Re: Native partitioning tablespace inheritance
Next
From: Peter Marreck
Date:
Subject: Proposal: Remove "no" from the default english.stop word list