Re: pgsql: Unbreak the build.

Hi,

On 2022-03-23 11:38:39 -0400, Robert Haas wrote:
> On Wed, Mar 23, 2022 at 10:46 AM Andres Freund <andres(at)anarazel(dot)de> wrote:
> > There's new warnings that sound reasonable introduced in the prior commit that didn't get removed in this one:
> >
> > https://cirrus-ci.com/task/5259487073271808?logs=mingw_cross_warning#L392
>
> That link takes me to a screen that shows no warnings.

Hm, apparently copied the link with slightly off line numbers. Odd.

> Scrolling around the only thing I see that doesn't seem to be addressed by
> this commit is a complaint about get_bc_algorithm_name falling off the end.

Well, there's also the test failure on windows...

> I've added a dummy return statement to hopefully address that.

Assert(false); won't help a compiler to see the path is unreachable when
building without assertions. Might be nicer to use pg_unreachable().

Greetings,

Andres Freund