DOCS - typos and grammar issues across logical replication docs.

On Fri, Apr 24, 2026 at 03:32:24PM +1000, Peter Smith wrote:

I fed the logical replication online documentation into my favourite
AI tool to help me come up with the following patch that fixes many
minor issues (e.g., spelling, grammar, inconsistencies, vagueness,
etc.).

It is a mixed bag of changes. Some of the suggestions are correct
grammar fixes, some make things somewhat worse, some are irrelevant
(like the s/won't/will not/ or the s/please//).

-  When logical replication of a table typically starts, PostgreSQL takes
-  a snapshot of the table's data on the publisher database and copies it
+  Logical replication of a table typically starts by PostgreSQL taking
+  a snapshot of the table's data on the publisher database and copying it
   to the subscriber.  Once complete, changes on the publisher since the

This new sentence makes less sense here. And there's nothing bad with
the original.

Some of these mistakes are older than HEAD, and are user-visible, so
backpatched the most relevant ones.

Thanks.
--
Michael

On Mon, Apr 27, 2026 at 5:18 PM Michael Paquier <michael@paquier.xyz> wrote:

On Fri, Apr 24, 2026 at 03:32:24PM +1000, Peter Smith wrote:

I fed the logical replication online documentation into my favourite
AI tool to help me come up with the following patch that fixes many
minor issues (e.g., spelling, grammar, inconsistencies, vagueness,
etc.).

It is a mixed bag of changes. Some of the suggestions are correct
grammar fixes, some make things somewhat worse, some are irrelevant
(like the s/won't/will not/ or the s/please//).

-  When logical replication of a table typically starts, PostgreSQL takes
-  a snapshot of the table's data on the publisher database and copies it
+  Logical replication of a table typically starts by PostgreSQL taking
+  a snapshot of the table's data on the publisher database and copying it
to the subscriber.  Once complete, changes on the publisher since the

This new sentence makes less sense here. And there's nothing bad with
the original.

Some of these mistakes are older than HEAD, and are user-visible, so
backpatched the most relevant ones.

Thanks.

Hi Michael,

Thanks for pushing!

(My opinion differs about some of the rejected ones, but I am not
going to debate about them.)

======
Kind Regards
Peter Smith.
Fujitsu Australia