Re: Documenting removal of nonnullvalue() and friends - Mailing list pgsql-docs

From Josh Kupershmidt
Subject Re: Documenting removal of nonnullvalue() and friends
Date
Msg-id AANLkTim=ExkXOAqGBAd6Uqnnq9X9L6UC9q2_bZGR792a@mail.gmail.com
Whole thread Raw
In response to Documenting removal of nonnullvalue() and friends  (Josh Kupershmidt <schmiddy@gmail.com>)
Responses Re: Documenting removal of nonnullvalue() and friends  (Tom Lane <tgl@sss.pgh.pa.us>)
List pgsql-docs
On Thu, Oct 14, 2010 at 10:00 AM, Peter Eisentraut <peter_e@gmx.net> wrote:
> The sets of intentionally documented and undocumented functions is part
> of the API specification of a release, and we're not changing that after
> the release, especially not when a future release ends up reverting the
> change.

Fair enough. I still think that we shouldn't be intentionally leaving
features undocumented, as it just leads to confusion down the road
(e.g. trying to figure out what some legacy code is doing, without
documentation).

I'm curious now, though: are there any other functions which we're
leaving intentionally undocumented? The only similar one I've run
across are the deprecated "-a" and "-A" arguments to the createuser
command. (There's another argument for documenting deprecated
features/arguments: it's easier to know what needs to be ripped out in
future releases).

Josh

pgsql-docs by date:

Previous
From: Peter Eisentraut
Date:
Subject: Re: Documenting removal of nonnullvalue() and friends
Next
From: Tom Lane
Date:
Subject: Re: Documenting removal of nonnullvalue() and friends