AKA ".stringof strikes again", or ".attackof.stringof"...

Not sure if this is right or not:

------------------------------
void foo(){}
pragma(msg, foo.stringof);
------------------------------

Outputs "foo()", but shouldn't it just be "foo" instead? Or am I overlooking something?

And on top of that, giving foo a parameter:

------------------------------
void foo(int i){}
pragma(msg, foo.stringof);
------------------------------

Error: function main.foo (int) does not match parameter types ()
Error: expected 1 arguments, not 0

WTF?

Ok, so maybe that stupid optional-parens function-invocation "feature" is kicking in where it's not wanted. D's standard way to refer to a function itself is supposed to be '&':

------------------------------
void foo(int i){}
pragma(msg, &foo.stringof);
------------------------------
main.d(2): Error: function main.foo (int) does not match parameter types ()
main.d(2): Error: expected 1 arguments, not 0
main.d(2): Error: "foo()"c is not an lvalue
main.d(2): Error: pragma msg string expected for message, not '&"foo()"c'

FFPPJTTdD!!!!!

Associativity problem?

------------------------------
void foo(int i){}
pragma(msg, (&foo).stringof);
------------------------------
& foo

Argh! (Not to be confused with "Args!")

Sooooooooo.......

If I'm writing a template that takes in a varadic list of variables and functions, and does something with their names, what's the right way to do that (if any)? Trivial example:

------------------------------
template makeBools(idents...)
{
    const char[] foo = "bool _generated_from_"~idents[0].stringof~"_name;" ~
        foo!(idents[1..$]);
}
int i;
void func1(){}
void func2(int x){}
mixin(makeBools!(i, func1, func2)); // Thoroughly fucks up.
------------------------------

I suppose I could resort to passing in string literals, but I'd really rather not have to.

"Nick Sabalausky" <a@a.a> wrote in message news:hdj3dk$1r5k$1@digitalmars.com...
> AKA ".stringof strikes again", or ".attackof.stringof"...
>

Ok, *now* I see all the reports of this on bugzilla, now that I searched for just "stringof" and dug through the pile of results, instead of searching for both "stringof" and "function"...Real pain for metaprogramming...

On Fri, Nov 13, 2009 at 12:05 AM, Nick Sabalausky <a@a.a> wrote:
> "Nick Sabalausky" <a@a.a> wrote in message news:hdj3dk$1r5k$1@digitalmars.com...
>> AKA ".stringof strikes again", or ".attackof.stringof"...
>>
>
> Ok, *now* I see all the reports of this on bugzilla, now that I searched for just "stringof" and dug through the pile of results, instead of searching for both "stringof" and "function"...Real pain for metaprogramming...

At any rate I think a single stringof for a function is not sufficient.
You might want any of:
    foo
    foo(int, int)
    foo(int a, int b)
There should probably be some __traits functions for getting these
different things, if there aren't already.
--bb

Bill Baxter wrote:
> On Fri, Nov 13, 2009 at 12:05 AM, Nick Sabalausky <a@a.a> wrote:
>> "Nick Sabalausky" <a@a.a> wrote in message
>> news:hdj3dk$1r5k$1@digitalmars.com...
>>> AKA ".stringof strikes again", or ".attackof.stringof"...
>>>
>> Ok, *now* I see all the reports of this on bugzilla, now that I searched for
>> just "stringof" and dug through the pile of results, instead of searching
>> for both "stringof" and "function"...Real pain for metaprogramming...
> 
> At any rate I think a single stringof for a function is not sufficient.
> You might want any of:
>     foo
>     foo(int, int)
>     foo(int a, int b)
> There should probably be some __traits functions for getting these
> different things, if there aren't already.
> --bb
Funny thing --  .stringof was a direct response from Walter to my 'meta.nameof' module.
It included  nameOf(xxx), qualifiedNameOf(xxx) and prettyNameOf(xxx), which were the 3 cases you listed above.