Re: Is it time to kill support for very old servers?

On Wed, Oct 11, 2017 at 10:28 PM, Robert Haas <robertmhaas(at)gmail(dot)com> wrote:
> On Tue, Oct 10, 2017 at 9:14 PM, Andres Freund <andres(at)anarazel(dot)de> wrote:
>> I'm actually inclined not to, and keep this as a undocumented debugging
>> option. Limiting the use of this option to people willing to read the
>> code seems like a good idea to me.
>
> -1. I use the documentation to find things, even though I am a
> developer, and I don't think hiding things from users is a good idea
> anyway. We can say that we recommend against using the option and
> that it's just for testing, but it should be documented anyway.

You would be disappointed with the current state of things then.
fe-connect.c lists a couple of debug options:
- authtype, which is not used anywhere, still kept to not reject
connection strings using it. I can get that this is not documented.
- tty, same argument as authtype.
- replication, which is still available, and that I have for example
used a couple of times for debugging the replication parser (psql can
work directly with some replication commands as you know). But this is
not documented.

FWIW, I would like to see things properly documented as well. And
please note that I am fine to write such a patch as well.
--
Michael