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

On 2017-09-18 02:53:03 -0700, Andres Freund wrote:
> On 2017-09-13 23:39:21 -0400, Tom Lane wrote:
> > The real problem in this area, to my mind, is that we're not testing that
> > code --- either end of it --- in any systematic way. If it's broken it
> > could take us quite a while to notice.
>
> Independent of the thrust of my question - why aren't we adding a
> 'force-v2' option to libpq? A test that basically does something like
> postgres[22923][1]=# \setenv PGFORCEV2 1
> postgres[22923][1]=# \c
> You are now connected to database "postgres" as user "andres".
> postgres[22924][1]=>
> seems easy enough to add, in fact I tested the above.
>
> And the protocol coverage of the v2 protocol seems small enough that a
> single not too large file ought to cover most if it quite easily.

Here's what I roughly was thinking of. I don't quite like the name, and
the way the version is specified for libpq (basically just the "raw"
integer). Not sure if others have an opinion on that. I personally
would lean towards not documenting this option...

There's a few things that I couldn't find easy ways to test:
- the v2 specific binary protocol - I don't quite see how we could test
that without writing C
- version error checks - pg_regress/psql errors out in non-interactive
mode if a connection fails to be established. This we could verify
with a s simple tap test.

Coverage of the relevant files is a good bit higher afterwards. Although
our libpq coverage is generally pretty damn awful.

Regards,

Andres