On Tue, Aug 18, 2020 at 1:31 PM Matthias van de Meent
wrote:
> Would this extra ordering not effectively be an extra tiebreaker in
> the ordering, applied before the TID? I do not know the full
> implications of that, but I believe that would not result in the
> limitations that you are mentioning
On Tue, 18 Aug 2020 at 22:00, Peter Geoghegan wrote:
>
> On Tue, Aug 18, 2020 at 11:52 AM Matthias van de Meent
> wrote:
> > Given that the above could work, the current btree tuple ordering is
> > not optimized for opclass-equal but datum image-distinct values:
> > ordering of opclass-equal valu
On Tue, 18 Aug 2020 at 18:44, Peter Geoghegan wrote:
>
> On Mon, Aug 17, 2020 at 11:44 PM Matthias van de Meent
> wrote:
> > But, if the ordering of operator-class equal tuples is already
> > system-defined, could the physical ordering of index tuples in a btree
> > (with deduplication enabled fo
On Tue, Aug 18, 2020 at 11:52 AM Matthias van de Meent
wrote:
> Given that the above could work, the current btree tuple ordering is
> not optimized for opclass-equal but datum image-distinct values:
> ordering of opclass-equal values is currently determined only by tid,
> with as an example curre
On Tue, Aug 18, 2020 at 11:52 AM Matthias van de Meent
wrote:
> Deduplication does not need to destroy semantic differences? 'equal'
> can (in my book) mean:
> - 'opclass-equal', that is the opclass returns true for an equality check
> - 'binary equal' or 'datum-equal' (? maybe incorrect term), th
On Tue, Aug 18, 2020 at 9:44 AM Peter Geoghegan wrote:
> If we wanted to fix this for numeric, we'd have to invent a new
> numeric datatype (called numeric2, say). That probably isn't as hard
> as it sounds, since it could be part of the same B-Tree operator
> family as numeric. It could also be i
On Mon, Aug 17, 2020 at 11:44 PM Matthias van de Meent
wrote:
> But, if the ordering of operator-class equal tuples is already
> system-defined, could the physical ordering of index tuples in a btree
> (with deduplication enabled for "unsafe" opclasses) be updated from
> [index_columns, tid] to [i
Hi,
I was reading through the new features of PG13 (beta), and noticed
that deduplication is disabled for float(4, 8) and numeric (and jsonb,
...) due to that the datums of those types could be not binary equal,
but equal according for the opclass used.
But, if the ordering of operator-class equa
