Paul Eggert wrote:
> > Do we need the compiler to verify that the parameter is really unused?
> > No, we don't. Even if the code is later refactored in such a way that
> > the parameter gets used, there is no damage.
> ...
> I see several things going on here. First, it's useful to glance at a
> function F's head and see whether F's arguments are really needed, or
> simply pacify an API. MAYBE_UNUSED doesn't tell me that; UNNAMED does.
Yes, but in doing that, it's first of all more useful to know what a
parameter is about. It would be *horrible* to have a function defined as
void window_size_changed (int, int, int w, int)
{ ... }
It is much better to not omit the identifier:
void window_size_changed (int /*x*/, int /*y*/, int w, int /*h*/)
{ ... }
or
void window_size_changed (_GL_UNUSED_PARAMETER int x, _GL_UNUSED_PARAMETER
int y,
int w, _GL_UNUSED_PARAMETER int h)
{ ... }
or
void window_size_changed (int _GL_UNNAMED (x), int _GL_UNNAMED (y),
int w, int _GL_UNNAMED (h))
{ ... }
> Second, not everyone agrees about whether and when unused parameters
> should be diagnosed and/or unnamed.
This is mostly because unused parameters arise from different situations:
- as implementations of a function type / API,
- or from refactorings and bitrot.
> Also witness C++, where a core guideline is "Unused parameters should be
> unnamed"
> <https://github.com/isocpp/CppCoreGuidelines/blob/master/CppCoreGuidelines.md#rf-unused>.
You must be joking? I can't take a style guideline from Bjarne Stroustrup
seriously. That is the guy who invented the syntax " = 0 " to denote an
unimplemented method and the parentheses syntax for the initializer of a
variable.
> I found many
> arguments that were formerly used in old Emacs versions but the code had
> bitrotted. The patch to fix this, which I have not applied yet, is over
> 4000 lines long; it does not use MAYBE_UNUSED or UNNAMED, but simply
> omits arguments.
I hope it does not omit the identifiers? That would be horrible, see
above.
> That project of using -Wunused-parameter in Emacs unfortunately also
> resulted in many false positives.
Maybe we should have a way to declare that a function is meant to be an
implementation of a function type / API ?
Bruno
