Documentation improvement patch

On Fri, Oct 9, 2026 at 9:46 AM Oleg <o.sibiryakov@postgrespro.ru> wrote:

Dear all,

I have prepared a patch containing some minor inconsistencies in the documentation.
Please, take a look.

I will be looking forward to your feedback.

Mine would be: "Wow! What a sharp eye! Congratulations! I reviewed all your
proposed changes and agree with all of them, +1"

On Fri, Oct 9, 2026 at 10:43 AM Matemática A3K <matematica.a3k@gmail.com>
wrote:

On Fri, Oct 9, 2026 at 9:46 AM Oleg <o.sibiryakov@postgrespro.ru> wrote:

Dear all,

I have prepared a patch containing some minor inconsistencies in the documentation.
Please, take a look.

I will be looking forward to your feedback.

Mine would be: "Wow! What a sharp eye! Congratulations! I reviewed all
your proposed changes and agree with all of them, +1"

Neither of the first two changes seems correct to me.

Third one is good. So maybe it's just the wanting to use commas where they
are unnecessary or incorrect that I'd push back on.

First it correctly adds missing periods then incorrectly removes others.

Using substring makes sense.

Most of these do seem to be unambiguous corrections.

For Oid -> OID - the documentation is inconsistent but favors Oid ...

 /*
- * Retrieve the string value of an HBA option which was registered via
+ * Retrieve the string value of an HBA option, which was registered via
  * RegisterOAuthHBAOptions(). Usable only during validate_cb or
shutdown_cb.
  *

Another unneeded comma and passed up the chance to use 'that' like in prior
changes.

I stopped about half way. Suggest breaking these according to the problem
being corrected. But I don't trust your present ability to know when to
introduce commas to the prose. I am also unsure whether we really care
about adhering to Oxford commas.

David J.

Hi,

I had a look at the parts that were related to logical replication.

Everything seemed "correct" to me, but the patch makes just bare
minimal changes for single words or commas. That's OK if minimal
changes is the objective, but there are some places where you might
consider going further and fixing awkward sentences properly given you
are changing them anyway. e.g.

Current:
The ADD TABLES IN SCHEMA, SET TABLES IN SCHEMA, SET ALL TABLES, and
SET ALL SEQUENCES to a publication requires the invoking user to be a
superuser.

Patched ("requires" -> "require"):
The ADD TABLES IN SCHEMA, SET TABLES IN SCHEMA, SET ALL TABLES, and
SET ALL SEQUENCES to a publication require the invoking user to be a
superuser.

Suggested:
The invoking user must be a superuser to use ADD TABLES IN SCHEMA, SET
TABLES IN SCHEMA, SET ALL TABLES, or SET ALL SEQUENCES.

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

On Sun, Oct 11, 2026 at 6:42 PM Peter Smith <smithpb2250@gmail.com> wrote:

Hi,

I had a look at the parts that were related to logical replication.

Everything seemed "correct" to me, but the patch makes just bare
minimal changes for single words or commas. That's OK if minimal
changes is the objective, but there are some places where you might
consider going further and fixing awkward sentences properly given you
are changing them anyway. e.g.

Current:
The ADD TABLES IN SCHEMA, SET TABLES IN SCHEMA, SET ALL TABLES, and
SET ALL SEQUENCES to a publication requires the invoking user to be a
superuser.

Patched ("requires" -> "require"):
The ADD TABLES IN SCHEMA, SET TABLES IN SCHEMA, SET ALL TABLES, and
SET ALL SEQUENCES to a publication require the invoking user to be a
superuser.

Suggested:
The invoking user must be a superuser to use ADD TABLES IN SCHEMA, SET
TABLES IN SCHEMA, SET ALL TABLES, or SET ALL SEQUENCES.

I also came to this conclusion when I reviewed it. I think if punctuation
is put aside because
it is debatable, the patch improves reading by fixing typos and unclear
sentences.

Can we agree to proceed in this direction?