[Scummvm-devel] Revisiting coding convention for empty loop body

Adrian Frühwirth scummvm-devel at bonki.net
Sat Apr 21 13:57:47 CEST 2018


Hi again,

This, too, has the advantage that it would probably silence potential compiler warnings should we turn -Wempty-body back
on at some future point.

+1

-Adrian

On 21.04.2018 12:54, Paul Gilbert wrote:
> Hmmm.. if we'd need to do some form of change to existing code to , I'd be more in favor of mandating the use of {}
> rather than the semicolon, as opposed to requiring an entire comment just to say that it's intentional that the loop
> doesn't have a body. It would convey the same information, in a more compact form. And it would have the advantage of
> bypassing the question of whether semicolons should have spaces in front of them.
> 
> Paul.
> 
> On Fri, Apr 20, 2018 at 4:30 PM, Adrian Frühwirth <scummvm-devel at bonki.net <mailto:scummvm-devel at bonki.net>> wrote:
> 
>     Hi everybody,
> 
>     Our CFCs currently state that trailing semicolons belonging to empty loops should be preceded by a space when on the
>     same line _and_ that the loop should be marked intentional with an appropriate comment. This was added to our CFCs after
>     a short discussion on -devel in 2010 [1].
> 
>     A quick grep reveals that there is not a single occurrence with such a comment (at least not where the semicolon is on
>     the same line) and most also skip the preceding space.
> 
>     The reason why this was proposed was that gcc warns about empty loop bodies when using -Wextra or -Wempty-body [2].
>     However, looking at our configure script, this warning seems to be disabled currently [3].
> 
>     If we intend to keep the warning disabled I would like to drop the requirement for either the extra space or the
>     comment, having both seems a bit redundant. I'm in strong favor of keeping the comment (or rather, add it where
>     necessary) and drop the space before the semicolon.
> 
>     Is anybody against this change?
> 
>     -Adrian
> 
>     [1] http://lists.scummvm.org/pipermail/scummvm-devel/2010-April/007554.html
>     <http://lists.scummvm.org/pipermail/scummvm-devel/2010-April/007554.html>
>     [2] https://gcc.gnu.org/onlinedocs/gcc/Warning-Options.html <https://gcc.gnu.org/onlinedocs/gcc/Warning-Options.html>
>     [3] https://github.com/scummvm/scummvm/blob/master/configure#L2033
>     <https://github.com/scummvm/scummvm/blob/master/configure#L2033>
> 
>     _______________________________________________
>     Scummvm-devel mailing list
>     Scummvm-devel at lists.scummvm.org <mailto:Scummvm-devel at lists.scummvm.org>
>     http://lists.scummvm.org/listinfo/scummvm-devel <http://lists.scummvm.org/listinfo/scummvm-devel>
> 
> 
> 
> 
> _______________________________________________
> Scummvm-devel mailing list
> Scummvm-devel at lists.scummvm.org
> http://lists.scummvm.org/listinfo/scummvm-devel
> 



More information about the Scummvm-devel mailing list