I am using compiler g++ 6.3.0 (c++14).
In the code-
#include<iostream>
int f(auto a){return a;}
int f1(auto (*g)(int),int a) {return g(a);}
main()
{
std::cout<< f1(f,8);
}
Compiler is not able to deduce the return type of g.
It shows following error-
temp.cpp: In function 'int main()':
temp.cpp:9:20: error: no matching function for call to 'f1(<unresolved overloaded function type>, int)'
std::cout<< f1(f,8);
^
temp.cpp:5:5: note: candidate: template<class auto:2> int f1(auto:2 (*)(int), int)
int f1(auto (*g)(int),int a) {return g(a);}
^~
temp.cpp:5:5: note: template argument deduction/substitution failed:
temp.cpp:9:20: note: couldn't deduce template parameter 'auto:2'
std::cout<< f1(f,8);
^
But no error comes in the code-
#include<iostream>
int f(int /* <<<<< */ a){return a;} // only (auto a) is changed to (int a)
int f1(auto (*g)(int),int a) {return g(a);}
main()
{
std::cout<< f1(f,8);
}
Help me understand the error...
int f(auto a){return a;}
is equivalent to
template <typename T>
int f(T a){return a;}
You cannot take the address of a template (or overload set) - that is why you're seeing that error. Workarounds:
Take the address of the instantiation you want:
return f1(f<int>,8);
Make f1 accept auto and pass a lambda:
int f1(auto g, int a) {return g(a);}
int main()
{
std::cout<< f1([](auto x){ f(x); },8);
}
Related
Surprisingly (embarrassingly?) I cannot get the syntax of the static_const of a const member function right. In short (details below) if the member function is not marked const I use:
static_cast<std::vector<double> (mymodule::Foo::*)(const std::vector<double>&)>(&mymodule::Foo::bar)
but marking the member function Foo::bar(...) const the compiler does not know what to do:
error: address of overloaded function 'bar' cannot be static_cast to type 'std::vector<double> (mymodule::Foo::*)(const std::vector<double> &)'
Where should I put the function's constness?
Details
I'm trying to create Python binding for the following module:
namespace mymodule {
class Foo
{
public:
Foo() = default;
template <class T>
T bar(const T& a) const
{
T ret = a;
for (auto& i : ret) {
i *= 2.0;
}
return ret;
}
template <class T>
T bar(const T& a, double f) const
{
T ret = a;
for (auto& i : ret) {
i *= f;
}
return ret;
}
};
} // namespace mymodule
whereby I write the Python bindings with pybind11:
#include <pybind11/pybind11.h>
namespace py = pybind11;
PYBIND11_MODULE(example, m)
{
py::class_<mymodule::Foo>(m, "Foo")
.def(py::init<>())
.def("bar",
static_cast<std::vector<double> (mymodule::Foo::*)(const std::vector<double>&)>(&mymodule::Foo::bar),
py::arg("a"))
.def("bar",
static_cast<std::vector<double> (mymodule::Foo::*)(const std::vector<double>&, double)>(&mymodule::Foo::bar),
py::arg("a"),
py::arg("f"));
}
which fails to compile:
.../example.cpp:54:14: error: address of overloaded function 'bar' cannot be static_cast to type 'std::vector<double> (mymodule::Foo::*)(const std::vector<double> &)'
static_cast<std::vector<double> (mymodule::Foo::*)(const std::vector<double>&)>(&mymodule::Foo::bar),
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
.../example.cpp:19:7: note: candidate function template
T bar(const T& a) const
^
.../example.cpp:29:7: note: candidate function template
T bar(const T& a, double f) const
^
.../example.cpp:58:14: error: address of overloaded function 'bar' cannot be static_cast to type 'std::vector<double> (mymodule::Foo::*)(const std::vector<double> &, double)'
static_cast<std::vector<double> (mymodule::Foo::*)(const std::vector<double>&, double)>(&mymodule::Foo::bar),
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
.../example.cpp:19:7: note: candidate function template
T bar(const T& a) const
^
.../example.cpp:29:7: note: candidate function template
T bar(const T& a, double f) const
^
2 errors generated.
You should add const at last as:
static_cast<std::vector<double> (mymodule::Foo::*)(const std::vector<double>&) const>(&mymodule::Foo::bar),
// ^^^^^
In the following code, I want to call a template function by implicitly converting an int to a Scalar<int> object.
#include<iostream>
using namespace std;
template<typename Dtype>
class Scalar{
public:
Scalar(Dtype v) : value_(v){}
private:
Dtype value_;
};
template<typename Dtype>
void func(int a, Scalar<Dtype> b){
cout << "ok" <<endl;
}
int main(){
int a = 1;
func(a, 2);
//int b = 2;
//func(a, b);
return 0;
}
Why does the template argument deduction/substitution fail? And the commented-codes are also wrong.
test.cpp: In function ‘int main()’:
test.cpp:19:12: error: no matching function for call to ‘func(int&, int)’
func(a, 2);
^
test.cpp:19:12: note: candidate is:
test.cpp:13:6: note: template<class Dtype> void func(int, Scalar<Dtype>)
void func(int a, Scalar<Dtype> b){
^
test.cpp:13:6: note: template argument deduction/substitution failed:
test.cpp:19:12: note: mismatched types ‘Scalar<Dtype>’ and ‘int’
func(a, 2);
Because template argument deduction is not that smart: it does not (by design) consider user-defined conversions. And int -> Scalar<int> is a user-defined conversion.
If you want to use TAD, you need to convert your argument at the caller site:
func(a, Scalar<int>{2});
or define a deduction guide1 for Scalar and call f:
func(a, Scalar{2}); // C++17 only
Alternatively, you can explicitly instantiate f:
func<int>(a, 2);
1) The default deduction guide is sufficient: demo.
template<typename Dtype>
void func(int a, Scalar<Dtype> b){
cout << "ok" <<endl;
}
template<typename Dtype>
void func(int a, Dtype b){
func(a, Scalar<Dtype>(std::move(b)));
}
template argument deduction is pattern matching, and it only matches the types or their base types exactly. It does no conversion.
Conversion is done later, at overload resolution & function call time.
Here, we add another overload that explicitly forwards to the one you want.
While writing some template code, I ran into <unresolved overloaded function type> errors which can be reduced to the following.
template <int N>
auto bar()
{
return N;
}
int main(int, char* [])
{
auto foo = [] (auto func) {
return func();
};
foo(bar<3>);
}
With the errors being:
unresolved_overload.cpp: In function 'int main(int, char**)':
unresolved_overload.cpp:26:28: error: no match for call to '(main(int, char**)::<lambda(auto:1)>) (<unresolved overloaded function type>)'
std::cout << foo(bar<3>) << std::endl;
^
unresolved_overload.cpp:21:29: note: candidate: template<class auto:1> constexpr main(int, char**)::<lambda(auto:1)>::operator decltype (((const main(int, char**)::<lambda(auto:1)>*)((const main(int, char**)::<lambda(auto:1)>* const)0))->operator()(static_cast<auto:1&&>(<anonymous>))) (*)(auto:1)() const
auto foo = [] (auto func) {
^
unresolved_overload.cpp:21:29: note: template argument deduction/substitution failed:
unresolved_overload.cpp:26:28: note: couldn't deduce template parameter 'auto:1'
std::cout << foo(bar<3>) << std::endl;
^
unresolved_overload.cpp:21:29: note: candidate: template<class auto:1> main(int, char**)::<lambda(auto:1)>
auto foo = [] (auto func) {
^
unresolved_overload.cpp:21:29: note: template argument deduction/substitution failed:
unresolved_overload.cpp:26:28: note: couldn't deduce template parameter 'auto:1'
std::cout << foo(bar<3>) << std::endl;
If we replace the auto-return with the explicit return type, int, the example will compile fine.
Why does auto-return run into these issues? I looked into template argument deduction and substitution but the search was largely unfruitful. I thought it might have something to do with the order of template instantiation / etc but couldn't make too much sense of it...
Per AndyG's suggestion, I found the same issue on GCC's bug list. Bug 64194. First reported in 2014. Thus the conclusion seems to be that this is a GCC bug and thankfully not another special case for templates.
Working around this just requires having something else to trigger the instantiation (e.g. assign to a variable, a using declaration).
Try this:
template <typename func>
auto bar(func&& f)->decltype(f())
{
return f();
}
int main()
{
int i = 100;
auto f = [=]()
{
return i;
};
bar(f);
return 0;
}
I'm working on a function which invokes a supplied function with a variable number of arguments. It compiles and works correctly on Visual Studio 2015, but fails to compile on Clang . I've prepared a demonstration which shows what I'm trying to do. The error I get in Clang is:
prog.cpp: In function 'int main()': prog.cpp:31:2: error: no matching
function for call to 'run(std::vector&, void ()(int&, const
int&), const int&)' ); ^ prog.cpp:7:6: note: candidate:
template void
run(std::vector&, const std::function&,
mutrArgs ...) void run(
^ prog.cpp:7:6: note: template argument deduction/substitution failed: prog.cpp:31:2: note: mismatched types 'const
std::function' and 'void ()(int&, const
int&)' );
#include <functional>
#include <iostream>
#include <vector>
using namespace std;
template<int RepeatTimes, class ... mutrArgs>
void run(
vector<int>& vec,
const function<void(int&, mutrArgs ...)>& mutr,
mutrArgs ... args
)
{
for (int times{0} ; times < RepeatTimes ; ++times)
for (auto& item : vec)
mutr(item, args...);
}
void adder(int& i, const int& val)
{
i += val;
}
int main()
{
vector<int> v{0,1,2,3,4,5,6,7,8,9};
const int addValue{4};
run<2, const int&>(
v,
&adder,
addValue
);
for (auto i : v)
cout << i << " ";
cout << endl;
return 0;
}
run<2, const int&> just state the first argument, but doesn't deactivate deduction.
run<2, const int&>(v, &adder, addValue);
has 2 places to deduce mutrArgs:
addValue -> mutrArgs = { const int& }
&adder which is not a std::function and so fail.
Taking address of function fix that problem
auto call_run = &run<2, const int&>;
call_run(v, &adder, addValue);
Strangely, clang doesn't support the inlined usage contrary to gcc :/
(&run<2, const int&>)(v, &adder, addValue);
If you want to disable deduction, you may make your template arg non deducible:
template <typename T> struct identity { using type = T; };
template <typename T> using non_deducible_t = typename identity<T>::type;
And then
template<int RepeatTimes, class ... mutrArgs>
void run(
std::vector<int>& vec,
const std::function<void(int&, non_deducible_t<mutrArgs> ...)>& mutr,
non_deducible_t<mutrArgs> ... args
)
Demo
Even if in your case a simple typename F as suggested by Joachim Pileborg seems better.
If you look at all standard library algorithm function, at least the ones taking a "predicate" (a callable object) they take that argument as a templated type.
If you do the same it will build:
template<int RepeatTimes, typename F, class ... mutrArgs>
void run(
vector<int>& vec,
F mutr,
mutrArgs ... args
)
{
...
}
See here for an example of you code. Note that you don't need to provide all template arguments, the compiler is able to deduce them.
The following code does not compile with G++ (although I believe it should):
#include <iostream>
template <unsigned N>
struct foo_traits {
typedef const char ArrayArg[N];
typedef int Function (ArrayArg *);
};
template <unsigned N>
int foo (typename foo_traits<N>::Function *ptr) {
return ptr(&"good");
}
int bar (const char (*x)[5]) {
std::cout << *x << "\n";
return 0;
}
int main ()
{
return foo(bar);
}
I checked this with GCC 4.4 through 4.7, and I get a template argument deduction failure. With 4.7.1:
prog.cpp: In function ‘int main()’:
prog.cpp:21:19: error: no matching function for call to ‘foo(int (&)(const char (*)[5]))’
prog.cpp:21:19: note: candidate is:
prog.cpp:10:5: note: template<unsigned int N> int foo(typename foo_traits<N>::Function*)
prog.cpp:10:5: note: template argument deduction/substitution failed:
prog.cpp:21:19: note: couldn't deduce template parameter ‘N’
If I use an explicit template argument (i.e., foo<5>(bar)), it compiles fine. If I use a version of the code without the typedefs, it compiles fine:
#include <iostream>
template <unsigned N>
int fixfoo (int (*ptr) (const char (*)[N])) {
return ptr(&"good");
}
int bar (const char (*x)[5]) {
std::cout << *x << "\n";
return 0;
}
int main ()
{
return fixfoo(bar);
}
Is the failing code supposed to compile (i.e., did I make a silly mistake)?
int foo(typename foo_traits<N>::Function *ptr);
The signature makes it a non-deductible context, so you must include the template arguments so that the value N is known and so consequentially the type of the pointer ptr be known as well.
Your second example compiles because the type of the signature through bar can be deduced.