Обсуждение: Patch: doc for pg_logical_emit_message()

Hi, Hackers

pg_logical_emit_message() can be used by any user,
but the following document says that it can be used by only superuser.

> Table 9.88. Replication SQL Functions
> Use of these functions is restricted to superusers.

I think that pg_logicl_emit_message() should be used by any user.
Therefore, I attach the document patch.

Ryo
Matsumura

On Fri, Apr 19, 2019 at 3:21 PM Matsumura, Ryo
<matsumura.ryo@jp.fujitsu.com> wrote:
>
> Hi, Hackers
>
> pg_logical_emit_message() can be used by any user,
> but the following document says that it can be used by only superuser.
>
> > Table 9.88. Replication SQL Functions
> > Use of these functions is restricted to superusers.
>
> I think that pg_logicl_emit_message() should be used by any user.
> Therefore, I attach the document patch.

Thanks for the patch!

Use of not only pg_logical_emit_message() but also other replication
functions in Table 9.88 is not restricted to superuser. For example,
replication role can execute pg_create_physical_replication_slot().
So I think that the patch should fix also the description for those
replication functions. Thought?

Regards,

-- 
Fujii Masao

On Tue. Apr. 23, 2019 at 02:59 AM Masao, Fujii
<masao.fujii@gmail.com> wrote:

Thank you for the comment.

> So I think that the patch should fix also the description for those
> replication functions. Thought?

I think so too.
I attach a new patch.

Regards
Ryo Matsumura

On Wed, Apr 24, 2019 at 11:12 AM Matsumura, Ryo
<matsumura.ryo@jp.fujitsu.com> wrote:
>
> On Tue. Apr. 23, 2019 at 02:59 AM Masao, Fujii
> <masao.fujii@gmail.com> wrote:
>
> Thank you for the comment.
>
> > So I think that the patch should fix also the description for those
> > replication functions. Thought?
>
> I think so too.
> I attach a new patch.

Thanks for updating the patch!

+    Use of functions for replication origin is restricted to superusers.
+    Use of functions for replication slot is restricted to superusers
and replication roles.

"replication role" is a bit confusing. For example, we have
"replication role" related to session_replication_role. So
I think it's better to use something like "users having
<literal>REPLICATION</literal> privilege".

+    Only <function>pg_logical_emit_message</function> can be used by any users.

Not any user, I think. For example, what about a user not having
EXECUTE privilege on pg_logical_emit_message function?
I don't think that this description only for pg_logical_emit_message
is necessary.

Regards,

-- 
Fujii Masao

On Wed. Apr. 24, 2019 at 11:40 PM Masao, Fujii
<masao.fujii@gmail.com> wrote:

Thank you for the comment.
I understand about REPLICATION privilege and notice my unecessary words.
I update the patch.

Regards
Ryo Matsumura

On Fri, Apr 26, 2019 at 10:52 AM Matsumura, Ryo
<matsumura.ryo@jp.fujitsu.com> wrote:
>
> On Wed. Apr. 24, 2019 at 11:40 PM Masao, Fujii
> <masao.fujii@gmail.com> wrote:
>
> Thank you for the comment.
> I understand about REPLICATION privilege and notice my unecessary words.
> I update the patch.

Thanks! Pushed.

Regards,

-- 
Fujii Masao

On Thu. May. 9, 2019 at 01:48 AM Masao, Fujii
<masao.fujii@gmail.com> wrote:

> Thanks! Pushed.

Thank you.


Regards
Ryo Matsumura