Re: Document NULL

On Wed, Dec 11, 2024 at 8:09 AM Marcos Pegoraro <marcos(at)f10(dot)com(dot)br> wrote:

> Em ter., 10 de dez. de 2024 às 20:00, David G. Johnston <
> david(dot)g(dot)johnston(at)gmail(dot)com> escreveu:
>
>
>> When I finalize the examples I'm probably going to \pset null <NULL>.
>>
> Yes, much better than an empty space in the examples, but you need to show
> what PSET you did, maybe
> <literal>\pset null</literal> meta-command</link> to specify the
> textual output of null values
> it encounters in query results. To get same results as you are seeing
> on this page, do "\pset null <NULL>"
>

Yes, I will be doing this in some form. My indecision at the moment is
that this ideally belongs in the preamble material for the section but
wants the reader to have seen the usage output sub-section.

> typo in func.sgml
> <link linkend="nullvalues">three-valued</link> typed
> results (true, false, or null).
> should remove that comma, right ?
> results (true, false or null).
>

I include the Oxford commas unless I'm forbidden to. There isn't any
forbiddance here that I am aware of. But now I'm questioning whether it
should read "true, false, and null" instead of "true, false, or
null"...though I'm confident that any of the four options is going to be
understood by the reader and this is basically purely stylistic in an area
we haven't codified. Any specific arguments or prior-art to consider for
choosing one over the others?

>
> Would be good to mention on nullvalues-json section that nulls on JSON
> values are case sensitive, so NULL or Null won't work
>
> [1] -
> https://www.postgresql.org/message-id/CAB-JLwYaauh-QRjnqN-DLfgZ8xFmy_uyGe545tiuejtiX0-T%2BQ%40mail.gmail.com
>
>
Agreed.

David J.