On 06/03/2018 11:50 PM, Tom Lane wrote:
> Tomas Vondra <tomas.vondra@2ndquadrant.com> writes:
>> The main remaining question I have is what do do with back-branches.
>> Shall we back-patch this or not?
>
> Given the behavioral changes involved, I'd say "no way". That's
> reinforced by the lack of field complaints; if there were lots of
> complaints, maybe we'd be willing to break backwards compatibility,
> but ...
>
Fair enough, I tend to over-estimate importance of bugfixes and
under-estimate breakage due to behavior change. But if we don't want to
back-patch this, I'm fine with that. I was a bit worried about making
future backpatches more painful, but this code received only ~20 commits
over the past files, half of that due tot pgindent, so that seems to be
a non-issue.
But now I'm wondering what does this mean for existing indexes? Doesn't
this effectively mean those are unlikely to give meaningful responses
(in the old or new semantics)?
regards
--
Tomas Vondra http://www.2ndQuadrant.com
PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services