Hi, On 9/12/18 1:34 PM, Michael Banck wrote: > Hi, > > On Fri, May 11, 2018 at 11:08:52AM -0400, Bruce Momjian wrote: >> I have committed the first draft of the Postgres 11 release notes. I >> will add more markup soon. You can view the most current version here: >> >> http://momjian.us/pgsql_docs/release-11.html > > The first item of section 'E.1.3.10. Server Applications' is mine and > has this TODO (though maybe that should be better marked up?) text: > > |IS IT CLEAR FROM THE DOCS THAT THE REPLICATION SLOT IS NOT TEMPORARY? > > So I guess we need to decide on that before release. The current doc is > AFAIK: > > | This option causes the replication slot specified by the > | option <literal>--slot</literal> to be created before starting the > | backup. In this case, an error is raised if the slot already exists. > > I guess it does not hurt to have something like "to be (permanently) > created before starting", but what do others think? Is it clear enough?
Attached are two patches. The first modifies the major improvements section to align with what we have discovered during the beta period + the general advocacy push. It also modifies the "last updated date" + narrows down a prospective release month. The second modifies/removes the questions/placeholders in the release notes, such as Michael's comment above (which I agreed with and added a one sentence explanation after). Thanks, Jonathan
From 2733f0df599aeda48408a3d27b796e36fbdd6fc3 Mon Sep 17 00:00:00 2001
From: "Jonathan S. Katz" <jonathan.k...@excoventures.com>
Date: Sat, 15 Sep 2018 10:43:44 -0400
Subject: [PATCH 1/2] Updates to major improvements section of release notes.
---
doc/src/sgml/release-11.sgml | 47 ++++++++++++++++++++++++++++++--------------
1 file changed, 32 insertions(+), 15 deletions(-)
diff --git a/doc/src/sgml/release-11.sgml b/doc/src/sgml/release-11.sgml
index 684d34c091..9ed34e72e8 100644
--- a/doc/src/sgml/release-11.sgml
+++ b/doc/src/sgml/release-11.sgml
@@ -6,7 +6,7 @@
<formalpara>
<title>Release date:</title>
- <para>2018-??-?? (CURRENT AS OF 2018-07-30)</para>
+ <para>2018-10-?? (CURRENT AS OF 2018-09-20)</para>
</formalpara>
<sect2>
@@ -22,11 +22,11 @@
<listitem>
<para>
- Major improvements to partitioning:
+ Improvements to partitioning functionality, including:
<itemizedlist>
<listitem>
<para>
- Partitioning by a hash key
+ Partitioning by a hash key (a.k.a. "hash partitioning")
</para>
</listitem>
<listitem>
@@ -37,9 +37,8 @@
</listitem>
<listitem>
<para>
- Improved <command>SELECT</command> query performance due to
- enhanced partition elimination during query processing and
- execution
+ Improved <command>SELECT</command> performance from enhanced partition
+ elimination strategies during query processing and execution
</para>
</listitem>
<listitem>
@@ -48,29 +47,37 @@
KEY</literal>, indexes, and triggers on partitioned tables
</para>
</listitem>
+ <listitem>
+ <para>
+ Having a "default" partition for storing data that does not match a
+ partition key
+ </para>
+ </listitem>
</itemizedlist>
</para>
</listitem>
<listitem>
<para>
- Improvements to parallelism:
+ Improvements to parallelism, including:
<itemizedlist>
<listitem>
<para>
- Parallelized hash joins
+ B-tree indexes can now be built in parallel with
+ <command>CREATE INDEX</command>
</para>
</listitem>
<listitem>
<para>
- Parallelized <command>CREATE INDEX</command> for B-tree indexes
+ Parallelized <command>CREATE TABLE .. AS</command>,
+ <command>CREATE MATERIALIZED VIEW</command>, and certain
+ queries using <literal>UNION</literal>
</para>
</listitem>
<listitem>
<para>
- Parallelized <command>CREATE TABLE .. AS</command>,
- <command>CREATE MATERIALIZED VIEW</command>, and certain
- queries using <literal>UNION</literal>
+ Performance improvements for parallelized hash joins and parallelized
+ sequential scans
</para>
</listitem>
</itemizedlist>
@@ -79,14 +86,17 @@
<listitem>
<para>
- SQL stored procedures, with support for embedded transactions
+ SQL stored procedures that support embedded transactions. Stored
+ procedures can be created with <link linkend="sql-createprocedure">
+ <command>CREATE PROCEDURE</command></link> and executed with
+ <link linkend="sql-call"><command>CALL</command></link>
</para>
</listitem>
<listitem>
<para>
- JIT compilation of some SQL code, including support for fast evaluation
- of expressions
+ Introduction of <link linkend="jit">just-in-time</link> (JIT) compilation
+ during query execution
</para>
</listitem>
@@ -99,6 +109,13 @@
</para>
</listitem>
+ <listitem>
+ <para>
+ Covering indexes, which can be utilized using the
+ <literal>INCLUDE</literal> clause of <command>CREATE INDEX</command>
+ </para>
+ </listitem>
+
<listitem>
<para>
Many other useful performance improvements, including making
--
2.14.3 (Apple Git-98)
From feb08e3747e5dc043e21b60e9e681f44a0562896 Mon Sep 17 00:00:00 2001
From: "Jonathan S. Katz" <jonathan.k...@excoventures.com>
Date: Sat, 15 Sep 2018 11:01:35 -0400
Subject: [PATCH 2/2] Remove or modify placeholders for v11 release notes.
---
doc/src/sgml/release-11.sgml | 21 ++++++---------------
1 file changed, 6 insertions(+), 15 deletions(-)
diff --git a/doc/src/sgml/release-11.sgml b/doc/src/sgml/release-11.sgml
index 9ed34e72e8..e8af023140 100644
--- a/doc/src/sgml/release-11.sgml
+++ b/doc/src/sgml/release-11.sgml
@@ -196,10 +196,6 @@
would be dumped without such specifications if the database locale
and encoding matched the old cluster's defaults.
</para>
-
- <para>
- DID I GET EVERYTHING?
- </para>
</listitem>
<listitem>
@@ -614,8 +610,7 @@
The new command <link linkend="sql-alterindex"><command>ALTER
INDEX ATTACH PARTITION</command></link> allows indexes to be
attached to partitions. This does not behave as a global index
- since the contents are private to each index. WARN WHEN USING
- AN EXISTING INDEX?
+ since the contents are private to each index.
</para>
</listitem>
@@ -924,7 +919,7 @@ same commits as above
</para>
<para>
- This reduces the likelihood of serialization conflicts. ACCURATE?
+ This reduces the likelihood of serialization conflicts.
</para>
</listitem>
@@ -1991,7 +1986,6 @@ same commits as above
<command>CALL</command>s or in nested PL/pgSQL DO and
<command>CALL</command> blocks that only contain other PL/pgSQL
<command>DO</command> and <command>CALL</command> blocks.
- ACCURATE?
</para>
</listitem>
@@ -2414,12 +2408,8 @@ same commits as above
The option <option>--create-slot</option> creates
the named replication slot (<option>--slot</option>)
when the <acronym>WAL</acronym> streaming method
- (<option>--wal-method=stream</option>) is used.
- </para>
-
- <para>
- IS IT CLEAR FROM THE DOCS THAT THE REPLICATION SLOT IS NOT
- TEMPORARY?
+ (<option>--wal-method=stream</option>) is used. The replication slot is
+ available until it is explicitly dropped.
</para>
</listitem>
@@ -2915,7 +2905,8 @@ same commits as above
</para>
<para>
- This is useful for KNN-GiST searches. HOW?
+ This is useful for KNN-GiST searches when looking for coordinates in
+ descending order.
</para>
</listitem>
--
2.14.3 (Apple Git-98)
signature.asc
Description: OpenPGP digital signature
