We need to kill C syntax for declaring function types

[Bruno Medeiros]

On 04/10/2010 10:07, Don wrote:
> A great example of how C syntax is hurting us.
> ---
> I found this bit of code in std.container, inside BinaryHeap:
>
> size_t insert(ElementType!Store value)
> {
> static if (is(_store.insertBack(value)))
> {
> ...
> }
> else ...
>
> What does the static if do? It's *intended* to check if _store has a
> member function insertBack(), which accepts type of a 'value'.
>
> But instead, it ALWAYS silently does the 'else' clause.
> Unless _store.insertBack is a valid *type*, (eg, alias int insertBack;).
> In which case it gives an error "declaration value is already defined".
>
> Why?
>
> This happens because
> x(y); is valid C syntax for declaring a type 'y', such that &y is of
> type 'x function()'.
>
> The C syntax is unspeakably ridiculous, useless, and downright
> dangerous. It shouldn't compile.
>

Whoa. I considered myself completely knowledgeable of the C language, 
but I had no idea about this syntax. (Note: by "completely 
knowledgeable" I don't mean I could recite the spec by memory, but 
rather that at least I knew what features, syntax and semantics ANSI C 
89 had available.)

Hum, your description of what "x(y);" means seems slightly incorrect 
though. Both in C and D, if x is a type, then it is the same as "x y;", 
that is, it declares a variable y with type x. If x is not a type but y 
is, it seems to be the same as "void x(y);", that is, it declares a 
function prototype named x. If both are not types, then it declares that 
strange thing I don't quite understand nor am I interested to...

I do vaguely recall learning about the first scenario, where a variable 
is declared, but I had not idea about the others. Is this mentioned in 
K&R TCPL 2nd edition?

Not that it matters, it's still horrid! I'm glad we're nuking it from D.

-- 
Bruno Medeiros - Software Engineer