Re: emacs-28 9c222b9: Port to C compilers that lack size-0 arrays

[Paul Eggert]

On 12/3/21 00:16, Eli Zaretskii wrote:

> We have already started the pretest.

Yes, and the pretest is why I found the problems I fixed: I built and 
tested on less-common platforms that I normally don't bother with.

> How serious is the problem this attempts to solve, and what bad things
> will happen if we release Emacs 28.1 without those changes?

What I experienced was a ton of diagnostics with both the Oracle Solaris 
and the IBM AIX C compilers, which made it difficult for me to see the 
real problems on those platforms (of which there were some). Although I 
was able to finish the builds anyway, the old code was a clear violation 
of the C89/C99/C11 standards and a good indication that there will be 
problems building on other less-common platforms that I don't have 
access to.

Conforming to the C standard doesn't affect behavior here (it possibly 
doesn't even change the machine code, though I didn't check this) so I 
considered the change to be a safe prophylactic.