MIME-Version: 1.0
References: <CA+renyUApHgSZF9-nd-a0+OPGharLQLO=mDHcY4_qQ0+noCUVg@mail.gmail.com>
 <CACJufxHVg65raNhG2zBwXgjrD6jqace4NZbePyMhP8-_Q=iT8w@mail.gmail.com>
 <0af6a323-f026-4cd9-adf1-941038cd00de@illuminatedcomputing.com>
 <88518c81-dcdc-4c5b-9200-146b74a520ab@eisentraut.org> <113f5e25-4455-4bdb-98ec-56275705cc4e@illuminatedcomputing.com>
 <7bd1c8f9-a91a-41a3-990e-0f796ba692ec@eisentraut.org> <d1a793bd-d1f1-46d7-90b6-8b89b6feb1cc@illuminatedcomputing.com>
 <6c3edb2c-bc45-4f19-babd-8f4edf39b384@illuminatedcomputing.com>
 <41b57948-f4d3-455f-8880-e0ce7e61ffc1@illuminatedcomputing.com>
 <0d977c63-f4a6-43ba-aad0-35c06c55df11@eisentraut.org> <CACJufxFFeUHoBomd2856amDBccNVQZzA=Jrk1_FDdvSZ1+X7og@mail.gmail.com>
 <CACJufxEACgLY7YUvzo8j=iOB-z6-1QHAvMr8cUHRP7b_F_vo3A@mail.gmail.com>
 <CA+renyXHFFyq1n2+4cwcFoLXqzUhb6u6WCL6wjk8ZjP9WxTExw@mail.gmail.com>
 <CACJufxF3voTh5yT46w2C4Noy=YDxKNi_LW7r0C55MuPvk=r1Pw@mail.gmail.com>
 <CACJufxHTrJwLp2rtWi2LPrxef_7gU04M=_d_LP3VY0=+NFHrkw@mail.gmail.com> <9372f843-006f-4e79-a64a-b60178cf40e9@illuminatedcomputing.com>
In-Reply-To: <9372f843-006f-4e79-a64a-b60178cf40e9@illuminatedcomputing.com>
From: jian he <jian.universality@gmail.com>
Date: Tue, 19 Mar 2024 17:01:02 +0800
Message-ID: <CACJufxHuzsn_9V+KpONTY-1pb92uVZxesqF50ZS-eSS2qKSr-Q@mail.gmail.com>
Subject: Re: SQL:2011 application time
To: Paul Jungwirth <pj@illuminatedcomputing.com>
Cc: Peter Eisentraut <peter@eisentraut.org>, 
	PostgreSQL Hackers <pgsql-hackers@lists.postgresql.org>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Archived-At: <https://www.postgresql.org/message-id/CACJufxHuzsn_9V%2BKpONTY-1pb92uVZxesqF50ZS-eSS2qKSr-Q%40mail.gmail.com>
Precedence: bulk

On Tue, Mar 19, 2024 at 6:49=E2=80=AFAM Paul Jungwirth
<pj@illuminatedcomputing.com> wrote:
>
> Rebased to 846311051e.
>

Hi, I just found out some minor issues.

+ * types matching the PERIOD element. periodprocoid is a GiST support
function to
+ * aggregate multiple PERIOD element values into a single value
+ * (whose return type need not match its inputs,
+ * e.g. many ranges can be aggregated into a multirange).
  * And aggedperiodoperoid is also a ContainedBy operator,
- * but one whose rhs is anymultirange.
+ * but one whose rhs matches the type returned by aggedperiodoperoid.
  * That way foreign keys can compare fkattr <@ range_agg(pkattr).
  */
 void
-FindFKPeriodOpers(Oid opclass,
-  Oid *periodoperoid,
-  Oid *aggedperiodoperoid)
+FindFKPeriodOpersAndProcs(Oid opclass,
+  Oid *periodoperoid,
+  Oid *aggedperiodoperoid,
+  Oid *periodprocoid)

I think, aggedperiodoperoid is more descriptive than periodprocoid, in
0005, you don't need to rename it.
aslo do we need to squash v29 0001 to 0005 together?

--- a/doc/src/sgml/ref/create_table.sgml
+++ b/doc/src/sgml/ref/create_table.sgml
@@ -1167,7 +1167,8 @@ WITH ( MODULUS <replaceable
class=3D"parameter">numeric_literal</replaceable>, REM
       column(s) of some row of the referenced table.  If the <replaceable
       class=3D"parameter">refcolumn</replaceable> list is omitted, the
       primary key of the <replaceable class=3D"parameter">reftable</replac=
eable>
-      is used.  Otherwise, the <replaceable
class=3D"parameter">refcolumn</replaceable>
+      is used (omitting any part declared with <literal>WITHOUT
OVERLAPS</literal>).
+      Otherwise, the <replaceable class=3D"parameter">refcolumn</replaceab=
le>
       list must refer to the columns of a non-deferrable unique or primary=
 key
       constraint or be the columns of a non-partial unique index.
      </para>
I think this does not express that
foreign key is PERIOD, then the last column of refcolumn must specify PERIO=
D?

+     <para>
+      If the last column is marked with <literal>PERIOD</literal>,
+      it is treated in a special way.
+      While the non-<literal>PERIOD</literal> columns are compared for equ=
ality
+      (and there must be at least one of them),
+      the <literal>PERIOD</literal> column is not.
+      Instead the constraint is considered satisfied
+      if the referenced table has matching records
+      (based on the non-<literal>PERIOD</literal> parts of the key)
+      whose combined <literal>PERIOD</literal> values completely cover
+      the referencing record's.
+      In other words, the reference must have a referent for its
entire duration.
+      This column must be a column with a range type.
+      In addition the referenced table must have a primary key
+      or unique constraint declared with <literal>WITHOUT PORTION</literal=
>.
+     </para>
you forgot to change  <literal>WITHOUT PORTION</literal> to
<literal>WITHOUT OVERLAPS</literal>


Oid pf_eq_oprs[RI_MAX_NUMKEYS]; /* equality operators (PK =3D FK) */
Oid pp_eq_oprs[RI_MAX_NUMKEYS]; /* equality operators (PK =3D PK) */
Oid ff_eq_oprs[RI_MAX_NUMKEYS]; /* equality operators (FK =3D FK) */
in struct RI_ConstraintInfo, these comments need to be updated?