Re: add_partial_path() may remove dominated path but still in use

From: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
To: Amit Kapila <amit(dot)kapila16(at)gmail(dot)com>
Cc: Kohei KaiGai <kaigai(at)heterodb(dot)com>, Kyotaro HORIGUCHI <horiguchi(dot)kyotaro(at)lab(dot)ntt(dot)co(dot)jp>, Robert Haas <robertmhaas(at)gmail(dot)com>, Pg Hackers <pgsql-hackers(at)postgresql(dot)org>
Subject: Re: add_partial_path() may remove dominated path but still in use
Date: 2019-02-09 15:27:24
Message-ID: 6941.1549726044@sss.pgh.pa.us
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

Amit Kapila <amit(dot)kapila16(at)gmail(dot)com> writes:
> It seems you would also like to see this back-patched. I am not sure
> if that is a good idea as there is some risk of breaking existing
> usage. Tom, do you have any opinion on this patch? It seems to me
> you were thinking to have a separate hook for partial paths, but the
> patch has solved the problem by moving the hook location.

I was expecting Haas to take point on this, but since he doesn't seem
to be doing so, I'll push it. I don't think there's any material
risk of breaking things --- the only functionality lost is the ability to
remove or modify baserel Gather paths, which I doubt anybody is interested
in doing. Certainly that's way less useful than the ability to add
partial paths and have them be included in Gather-building.

In a green field I'd rather have had a separate hook for adding partial
paths, but it's not clear that that really buys much of anything except
logical cleanliness ... against which it adds cost since the using
extension(s) have to figure out what's going on twice.

Also this way does have the advantage that it retroactively fixes things
for extensions that may be trying to make partial paths today.

regards, tom lane

In response to

Browse pgsql-hackers by date

  From Date Subject
Next Message Alvaro Herrera 2019-02-09 15:56:13 Re: Fixing findDependentObjects()'s dependency on scan order (regressions in DROP diagnostic messages)
Previous Message Peter Eisentraut 2019-02-09 15:06:10 Re: pgsql: Restrict the use of temporary namespace in two-phase transaction