Re: JSONPATH documentation

From: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
To: Alexander Korotkov <a(dot)korotkov(at)postgrespro(dot)ru>
Cc: Jeff Janes <jeff(dot)janes(at)gmail(dot)com>, pgsql-hackers <pgsql-hackers(at)postgresql(dot)org>
Subject: Re: JSONPATH documentation
Date: 2019-09-22 22:03:32
Message-ID: 16968.1569189812@sss.pgh.pa.us
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

Alexander Korotkov <a(dot)korotkov(at)postgrespro(dot)ru> writes:
> On Sun, Sep 22, 2019 at 9:18 PM Jeff Janes <jeff(dot)janes(at)gmail(dot)com> wrote:
> Currently description of jsonpath is divided between datatypes section
> and functions and operators section. And yes, this looks cumbersome.

Agreed, but ...

> I think we should move the whole description to the one section.
> Probably we should move jsonpath description to datatypes section
> (assuming jsonpath is a datatype) leaving functions and operators
> section with just SQL-level functions and operators. What do you
> think?

... I don't think that's an improvement. We don't document detailed
behavior of a datatype's functions in datatype.sgml, and this seems
like it would be contrary to that layout. If anything, I'd merge
the other way, with only a very minimal description of jsonpath
(perhaps none?) in datatype.sgml.

While we're whining about this, I find it very off-putting that
the jsonpath stuff was inserted in the JSON functions section
ahead of the actual JSON functions. I think it should have
gone after them, because it feels like a barely-related interjection
as it stands. Maybe there's even a case that it should be
its own <sect1>, after the "functions-json" section.

regards, tom lane

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Thomas Munro 2019-09-22 22:29:06 Re: scorpionfly needs more semaphores
Previous Message David Fetter 2019-09-22 21:58:04 Re: Efficient output for integer types