Re: plpython vs _POSIX_C_SOURCE

On Tue, Jan 24, 2023 at 11:37 PM Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> wrote:
> Andres Freund <andres(at)anarazel(dot)de> writes:
> > Patches attached.
>
> +1 for 0001. I'm still nervous about 0002. However, maybe the
> cases that we had trouble with are legacy issues that nobody cares
> about anymore in 2023. We can always look for another answer if
> we get complaints, I guess.

It feels like things are changing so fast these days that whatever was
happening 12 years ago is not likely to be relevant. Compilers change
enough to cause warnings and even errors in just a few years. A decade
is long enough for an entire platform to become irrelevant.

Plus, the cost of experimentation here seems very low. Sure, something
might break, but if it does, we can just change it back, or change it
again. That's not really a big deal. The thing that would be a big
deal, maybe, is if we released and only found out afterward that this
caused some subtle and horrible problem for which we had no
back-patchable fix, but that seems pretty unlikely.

--
Robert Haas
EDB: http://www.enterprisedb.com