Horton

Thread

Commits

  1. doc: Improve description of pg_get_multixact_stats()

  1. Fix doc about pg_get_multixact_stats()

    Yingying Chen <cyy9255@gmail.com> — 2026-06-25T11:49:58Z

    Hi Hacker,

I found a small documentation issue in func-info.smgl:

   <para>
    The internal transaction ID type <type>xid</type> is 32 bits wide and
    wraps around every 4 billion transactions.  However,
    the functions shown in <xref linkend="functions-pg-snapshot"/>, except
    <function>age</function>, <function>mxid_age</function>, and
    <function>pg_get_multixact_members</function>, use a
    64-bit type <type>xid8</type> that does not wrap around during the life
    of an installation and can be converted to <type>xid</type> by casting
if
    required;  see <xref linkend="transaction-id"/> for details.

Since pg_get_multixact_stats() was added in PG19, it should be added to
this exception list.

Regards,
Yingying Chen

  2. Re: Fix doc about pg_get_multixact_stats()

    Michael Paquier <michael@paquier.xyz> — 2026-06-25T22:27:03Z

    On Thu, Jun 25, 2026 at 07:49:58PM +0800, Yingying Chen wrote:
> Since pg_get_multixact_stats() was added in PG19, it should be added to
> this exception list.

Indeed.  Thanks for the patch.  Will fix.
--
Michael

  3. Re: Fix doc about pg_get_multixact_stats()

    Chao Li <li.evan.chao@gmail.com> — 2026-06-25T23:15:37Z

    
> On Jun 26, 2026, at 06:27, Michael Paquier <michael@paquier.xyz> wrote:
> 
> On Thu, Jun 25, 2026 at 07:49:58PM +0800, Yingying Chen wrote:
>> Since pg_get_multixact_stats() was added in PG19, it should be added to
>> this exception list.
> 
> Indeed.  Thanks for the patch.  Will fix.
> --
> Michael

While reviewing this patch, I found this statement for pg_get_multixact_stats() is a little misleading or unclear:
```
       <para>
        To use this function, you must have privileges of the
        <literal>pg_read_all_stats</literal> role.
       </para></entry>
```

From the implementation, for unprivileged (not super user and not pg_read_all_stats role) users, pg_get_multixact_stats doesn't raise an error, but just returns all NULL values. The current statement doesn't seem to explain that behavior clearly.

How about rephrasing it like this?
```
      <para>
        Only superusers and roles with privileges of the
        <literal>pg_read_all_stats</literal> role can see details; other users
        receive null values.
      </para></entry>
```

Best regards,
--
Chao Li (Evan)
HighGo Software Co., Ltd.
https://www.highgo.com/

  4. Re: Fix doc about pg_get_multixact_stats()

    Michael Paquier <michael@paquier.xyz> — 2026-06-25T23:41:39Z

    On Fri, Jun 26, 2026 at 07:15:37AM +0800, Chao Li wrote:
> From the implementation, for unprivileged (not super user and not
> pg_read_all_stats role) users, pg_get_multixact_stats doesn't raise
> an error, but just returns all NULL values. The current statement
> doesn't seem to explain that behavior clearly.
> 
> How about rephrasing it like this?
> ```
>       <para>
>         Only superusers and roles with privileges of the
>         <literal>pg_read_all_stats</literal> role can see details; other users
>         receive null values.
>       </para></entry>
> ```

Yeah, that looks like an improvement over the statu-quo.  I was
looking for something similar but we don't have many functions hidden
behind a role privilege.
--
Michael