Re: doc: improve the restriction description of using indexes on REPLICA IDENTITY FULL table.

On Sat, Jul 8, 2023 at 1:49 PM Amit Kapila <amit.kapila16@gmail.com> wrote:
>
> On Fri, Jul 7, 2023 at 1:36 PM Masahiko Sawada <sawada.mshk@gmail.com> wrote:
> >
> > I prefer the first suggestion. I've attached the updated patch.
> >
>
> This looks mostly good to me but I think it would be better if we can
> also add the information that the leftmost index column must be a
> non-expression. So, how about: "Candidate indexes must be btree,
> non-partial, and the leftmost index column must be a non-expression
> and reference to a published table column (i.e. cannot consist of only
> expressions)."?

That part in parentheses ought to say "the index ..." because it is
referring to the full INDEX, not to the leftmost column. I think this
was missed when Sawada-san took my previous suggestion [1].

IMO it doesn't sound right to say the "index column must be a
non-expression". It is already a non-expression because it is a
column. So I think it would be better to refer to this as an INDEX
"field" instead of an INDEX column. Note that "field" is the same
terminology used in the docs for CREATE INDEX [2].

SUGGESTION
Candidate indexes must be btree, non-partial, and the leftmost index
field must be a column that references a published table column (i.e.
the index cannot consist of only expressions).

~~~~

What happened to the earlier idea of removing/merging the redundant
(?) function IsIndexOnlyOnExpression()?
- Something wrong with that?
- Chose not to do it?
- Will do it raised in another thread?

------
[1] my review v2 -
https://www.postgresql.org/message-id/CAHut%2BPsFdTZJ7DG1jyu7BpA_1d4hwEd-Q%2BmQAPWcj1ZLD_X5Dw%40mail.gmail.com
[2] create index - https://www.postgresql.org/docs/current/sql-createindex.html

Kind Regards,
Peter Smith.
Fujitsu Australia