Horton

Thread

Commits

  1. postgres_fdw: Fix thinko in comment for UserMappingPasswordRequired().

  1. Comment for UserMappingPasswordRequired in contrib/postgres_fdw

    Etsuro Fujita <etsuro.fujita@gmail.com> — 2026-02-18T20:23:26Z

    Hi,

While working on something else, I noticed $SUBJECT:

/*
 * Return true if the password_required is defined and false for this user
 * mapping, otherwise false. The mapping has been pre-validated.
 */
static bool
UserMappingPasswordRequired(UserMapping *user)
{
    ListCell   *cell;

    foreach(cell, user->options)
    {
        DefElem    *def = (DefElem *) lfirst(cell);

        if (strcmp(def->defname, "password_required") == 0)
            return defGetBoolean(def);
    }

    return true;
}

I think the former part of the comment should be: Return *false* if
the password_required is defined and false for this user mapping,
otherwise *true*.

Patch attached.

Best regards,
Etsuro Fujita

  2. Re: Comment for UserMappingPasswordRequired in contrib/postgres_fdw

    Andreas Karlsson <andreas@proxel.se> — 2026-02-18T20:30:06Z

    On 2/18/26 9:23 PM, Etsuro Fujita wrote:
> I think the former part of the comment should be: Return *false* if
> the password_required is defined and false for this user mapping,
> otherwise *true*.
I feel the wording of the comment is pretty awkward both before and 
after your correctness fix. I am not a native speaker but shouldn't it 
be something like the below which explains better what is actually going on.

/*
  * Checks the value of password_required, defaults to true
  * if not defined. The mapping has been pre-validated.
  */

Andreas

  3. Re: Comment for UserMappingPasswordRequired in contrib/postgres_fdw

    Etsuro Fujita <etsuro.fujita@gmail.com> — 2026-02-22T11:10:57Z

    On Thu, Feb 19, 2026 at 5:30 AM Andreas Karlsson <andreas@proxel.se> wrote:
> On 2/18/26 9:23 PM, Etsuro Fujita wrote:
> > I think the former part of the comment should be: Return *false* if
> > the password_required is defined and false for this user mapping,
> > otherwise *true*.
> I feel the wording of the comment is pretty awkward both before and
> after your correctness fix. I am not a native speaker but shouldn't it
> be something like the below which explains better what is actually going on.
>
> /*
>   * Checks the value of password_required, defaults to true
>   * if not defined. The mapping has been pre-validated.
>   */

I like your wording.  I am not a native speaker either, though.  This
would be nitpicking, but I think it is better to clearly mention what
the function returns.  How about modifying it a bit, like this?

/*
 * Check and return the value of password_required, if defined; otherwise,
 * return true, which is the default value of it.  The mapping has been
 * pre-validated.
 */

Anyway, thanks for the comment!

Best regards,
Etsuro Fujita

  4. Re: Comment for UserMappingPasswordRequired in contrib/postgres_fdw

    Andreas Karlsson <andreas@proxel.se> — 2026-02-22T23:01:31Z

    On 2/22/26 12:10 PM, Etsuro Fujita wrote:
>> /*
>>    * Checks the value of password_required, defaults to true
>>    * if not defined. The mapping has been pre-validated.
>>    */
> 
> I like your wording.  I am not a native speaker either, though.  This
> would be nitpicking, but I think it is better to clearly mention what
> the function returns.  How about modifying it a bit, like this?
> 
> /*
>   * Check and return the value of password_required, if defined; otherwise,
>   * return true, which is the default value of it.  The mapping has been
>   * pre-validated.
>   */
No strong opinion. I would be fine with either. I do not think saying 
that it returns is necessary since you can see that from the function 
definition but it does not really harm either.

-- 
Andreas Karlsson
Percona

  5. Re: Comment for UserMappingPasswordRequired in contrib/postgres_fdw

    Etsuro Fujita <etsuro.fujita@gmail.com> — 2026-02-23T09:36:38Z

    On Mon, Feb 23, 2026 at 8:01 AM Andreas Karlsson <andreas@proxel.se> wrote:
> On 2/22/26 12:10 PM, Etsuro Fujita wrote:
> >> /*
> >>    * Checks the value of password_required, defaults to true
> >>    * if not defined. The mapping has been pre-validated.
> >>    */
> >
> > I like your wording.  I am not a native speaker either, though.  This
> > would be nitpicking, but I think it is better to clearly mention what
> > the function returns.  How about modifying it a bit, like this?
> >
> > /*
> >   * Check and return the value of password_required, if defined; otherwise,
> >   * return true, which is the default value of it.  The mapping has been
> >   * pre-validated.
> >   */
> No strong opinion. I would be fine with either. I do not think saying
> that it returns is necessary since you can see that from the function
> definition but it does not really harm either.

Ok.  I like the modified version as it also keeps the existing comment
to some extent, so I'd like to go with it.  Updated patch attached.  I
will push and back-patch it to all supported versions if there are no
objections from others.

Thanks!

Best regards,
Etsuro Fujita

  6. Re: Comment for UserMappingPasswordRequired in contrib/postgres_fdw

    Etsuro Fujita <etsuro.fujita@gmail.com> — 2026-02-27T08:10:50Z

    On Mon, Feb 23, 2026 at 6:36 PM Etsuro Fujita <etsuro.fujita@gmail.com> wrote:
> On Mon, Feb 23, 2026 at 8:01 AM Andreas Karlsson <andreas@proxel.se> wrote:
> > On 2/22/26 12:10 PM, Etsuro Fujita wrote:
> > >> /*
> > >>    * Checks the value of password_required, defaults to true
> > >>    * if not defined. The mapping has been pre-validated.
> > >>    */
> > >
> > > I like your wording.  I am not a native speaker either, though.  This
> > > would be nitpicking, but I think it is better to clearly mention what
> > > the function returns.  How about modifying it a bit, like this?
> > >
> > > /*
> > >   * Check and return the value of password_required, if defined; otherwise,
> > >   * return true, which is the default value of it.  The mapping has been
> > >   * pre-validated.
> > >   */
> > No strong opinion. I would be fine with either. I do not think saying
> > that it returns is necessary since you can see that from the function
> > definition but it does not really harm either.
>
> Ok.  I like the modified version as it also keeps the existing comment
> to some extent, so I'd like to go with it.  Updated patch attached.  I
> will push and back-patch it to all supported versions if there are no
> objections from others.

Done.

Best regards,
Etsuro Fujita