1. Are there any plans for standardizing empty initializer lists?
struct foo { int a; void *p; };
struct foo f = {0}; // legal C, f->p initialized like a static variable
struct foo f = {}; // not legal but supported by gcc
To me it would make sense that there is no need to specify a value for any of the members that are intended to be initialized exactly like static variables (and the first member is not special so I shouldn't have to explicitly assign a zero?). However the syntax currently demands at least one initializer.
--
2. I recall seeing a proposal for allowing declarations after case labels:
switch (foo) {
case 1:
int var;
// ...
}
This is currently not allowed and you'd have to wrap the lines after case in braces, or insert a semicolon after the case label. Is this making it to c2x?
--
3. I've run into some recent controversy w.r.t. having multiple functions called main (and this has come up in production code). In particular, I ran into a program programs that has a static main() function (with parameters that are not void or int and char[]), which is not intended to be the* main function that is the program's entry point.
gcc warns about this because the parameters disagree with what's prescribed for the program entry point. It's not clear to me whether this is intended to be legal or not.
--
4. Looking at the requirements for main brings up another question: it says how main should be defined (no static or extern keyword). However, the definition could be preceded by a static declaration, which then affects the definition that follows:
If the declaration of an identifier for a function has no storage-class specifier, its linkage is determined exactly as if it were declared with the storage-class specifier extern.
For an identifier declared with the storage-class specifier extern in a scope in which a prior declaration of that identifier is visible, if the prior declaration specifies internal or external linkage, the linkage of the identifier at the later declaration is the same as the linkage specified at the prior declaration.
Therefore, it is possible to have a main function with internal linkage and a definition that exactly matches the one given in the spec:
static int main(int, char *[]);
int main(int argc, char *argv[]) { /* ... */ }
As one might guess, this program doesn't make it through the linker when compiled with gcc. Is this supposed to be legal? Should the spec perhaps require main to have external linkage, and then allow other functions called main with internal linkage (and parameters that do not match what is required of the external one)?
EDIT: ---
Are the fixes w.r.t. reserved identifiers going to make it in c2x? Can I finally have a function called toilet() without undefined behavior?