Re: [PATCH] Remove __ELF__ conditionals

["Carlos O'Donell" <carlos at systemhalted dot org>]

On Sun, Jan 29, 2012 at 8:59 AM, Joseph S. Myers
<joseph@codesourcery.com> wrote:
> On Sat, 28 Jan 2012, Carlos O'Donell wrote:
>
>> Please correct me if I'm wrong but the cleanup is part of the general
>> "remove things we don't use" philosophy we want to adopt? In this case
>
> Yes. ?Various parts of (long-dead and bitrotten) non-ELF support were
> removed as per <http://sourceware.org/bugzilla/show_bug.cgi?id=13551>.
> It turns out there are still some conditionals left - and it makes sense
> to remove them as well to simplify things for people reading and working
> on the code in future. ?Similarly, pre-standard C was actually desupported
> by glibc many years ago, __STDC__ conditionals removal is simply cleaning
> up some relics that remain of the old support.

Excellent.

>> We now support ranged notion for copyright years per README.
>
> FWIW, GDB plans to move to using the simplest copyright notice form: a
> single range of years, <first>-2012, with the last year in the range
> updated for all files by a script at the start of the year. ?(It was
> confirmed in <http://sourceware.org/ml/gdb-patches/2012-01/msg00936.html>
> that using just a single range like this is OK. ?It's been recommended
> practice for some time to update the list of years in all files at the
> start of the year so people don't need to worry about it for the rest of
> the year. ?gnulib has a script for such start-of-year updates.)

This is a great idea.

Is the GNU Coding Standard going to be updated with this information?

Should we wait for the standard to get updated before we adopt this policy?

> What do people think about similarly simplifying the copyright notices in
> glibc, and doing the automatic start-of-year updates, so that there's one
> fewer thing for contributors to get right and one fewer thing for
> reviewers to check when reviewing patches? ?That would be for all files
> with FSF copyright notices except those originating from other projects
> such as GMP and gettext and config.{sub,guess} (and subject to dealing
> properly with generated files, i.e. regenerating them from their sources).

I think this is a great idea, I just think we need to get confirmation that
this is applicable to all FSF projects.

I have not yet read, or know if I can read, the case ruling mentioned by
Joel Brobecker.

Cheers,
Carlos.