What I want to do is:
int const bitsPerInt = log2(X);
bitset<bitsPerInt> bits(a random number...);
but I get this error:
'bitsPerInt' cannot appear in a constant expression
error: template argument 1 is invalid
If you really need this to work, make your own log2 that works in compile-time and pass it to bitset's template argument.
constexpr unsigned Log2(unsigned n, unsigned p = 0) {
return (n <= 1) ? p : Log2(n / 2, p + 1);
}
constexpr size_t bitCount = Log2(X);
std::bitset<bitCount> bits;
Live example.
Here's the solution using template meta-programming i.e. without using constexpr:
template<int N,unsigned int P=0>
struct Log2 { enum { value = Log2<N/2,P+1>::value }; };
template <unsigned p>
struct Log2<0, p> { enum { value = p }; };
template <unsigned p>
struct Log2<1, p> { enum { value = p }; };
std::bitset<Log2<4>::value> bits;
Live example.
This version should work in both C++03 and C++11; however, if you've access to C++11, I'd still recommend the constexpr way since it's cleaner (easier to understand).
Template parameter needs to be known(and constant if it is a value and not a type) at compile time. This is how templates work in C++. Templates actually generate real code for each specific version of the generic code.
Related
I'm trying to implement a BitField template class for C++11/14, my base idea is:
template <typename T, size_t... Bits>
class BitField
{
public:
BitField();
private:
T value;
};
template <typename T, size_t... Bits>
BitField<T, Bits...>::BitField()
{
}
and then as an example instantiate it like this:
BitField<uint8_t, 2, 3, 3> bitfield;
where 2, 3, 3 are the sizes of the 3 bitfields, offsets come consequently (please let's ignore bit order by now).
Of course this is just a skeleton to be filled with appropriate setters and getters methods, now my question is: how can I get Bits values in the constructor to check if their sum fits the type, possibly at compile time?
More in general, can this be a good approach for this problem? I say other solutions but for many reasons I don't like them too much.
Thanks, Matteo
You can check size with (C++17)
static_assert(sizeof(T) == (Bits + ...));
Before fold-expression of C++17, there are workaround more verbose.
Create a function
template <std::size_t... Bits>
constexpr std::size_t sum()
{
const std::size_t numbers[] = {Bits...};
// std::accumulate in constexpr only since C++20
std::size_t res = 0;
for (auto number : numbers) {
res += number;
}
return res;
}
and then
static_assert(sizeof(T) == sum<Bits...>());
You can even use lambda instead of helper function if wanted.
Lets say that you have a function which generates some security token for your application, such as some hash salt, or maybe a symetric or asymetric key.
Now lets say that you have this function in your C++ as a constexpr and that you generate keys for your build based on some information (like, the build number, a timestamp, something else).
You being a diligent programmer make sure and call this in the appropriate ways to ensure it's only called at compile time, and thus the dead stripper removes the code from the final executable.
However, you can't ever be sure that someone else isn't going to call it in an unsafe way, or that maybe the compiler won't strip the function out, and then your security token algorithm will become public knowledge, making it more easy for would be attackers to guess future tokens.
Or, security aside, let's say the function takes a long time to execute and you want to make sure it never happens during runtime and causes a bad user experience for your end users.
Are there any ways to ensure that a constexpr function can never be called at runtime? Or alternately, throwing an assert or similar at runtime would be ok, but not as ideal obviously as a compile error would be.
I've heard that there is some way involving throwing an exception type that doesn't exist, so that if the constexpr function is not deadstripped out, you'll get a linker error, but have heard that this only works on some compilers.
Distantly related question: Force constexpr to be evaluated at compile time
In C++20 you can just replace constexpr by consteval to enforce a function to be always evaluated at compile time.
Example:
int rt_function(int v){ return v; }
constexpr int rt_ct_function(int v){ return v; }
consteval int ct_function(int v){ return v; }
int main(){
constexpr int ct_value = 1; // compile value
int rt_value = 2; // runtime value
int a = rt_function(ct_value);
int b = rt_ct_function(ct_value);
int c = ct_function(ct_value);
int d = rt_function(rt_value);
int e = rt_ct_function(rt_value);
int f = ct_function(rt_value); // ERROR: runtime value
constexpr int g = rt_function(ct_value); // ERROR: runtime function
constexpr int h = rt_ct_function(ct_value);
constexpr int i = ct_function(ct_value);
}
Pre C++20 workaround
You can enforce the use of it in a constant expression:
#include<utility>
template<typename T, T V>
constexpr auto ct() { return V; }
template<typename T>
constexpr auto func() {
return ct<decltype(std::declval<T>().value()), T{}.value()>();
}
template<typename T>
struct S {
constexpr S() {}
constexpr T value() { return T{}; }
};
template<typename T>
struct U {
U() {}
T value() { return T{}; }
};
int main() {
func<S<int>>();
// won't work
//func<U<int>>();
}
By using the result of the function as a template argument, you got an error if it can't be solved at compile-time.
A theoretical solution (as templates should be Turing complete) - don't use constexpr functions and fall back onto the good-old std=c++0x style of computing using exclusively struct template with values. For example, don't do
constexpr uintmax_t fact(uint n) {
return n>1 ? n*fact(n-1) : (n==1 ? 1 : 0);
}
but
template <uint N> struct fact {
uintmax_t value=N*fact<N-1>::value;
}
template <> struct fact<1>
uintmax_t value=1;
}
template <> struct fact<0>
uintmax_t value=0;
}
The struct approach is guaranteed to be evaluated exclusively at compile time.
The fact the guys at boost managed to do a compile time parser is a strong signal that, albeit tedious, this approach should be feasible - it's a one-off cost, maybe one can consider it an investment.
For example:
to power struct:
// ***Warning: note the unusual order of (power, base) for the parameters
// *** due to the default val for the base
template <unsigned long exponent, std::uintmax_t base=10>
struct pow_struct
{
private:
static constexpr uintmax_t at_half_pow=pow_struct<exponent / 2, base>::value;
public:
static constexpr uintmax_t value=
at_half_pow*at_half_pow*(exponent % 2 ? base : 1)
;
};
// not necessary, but will cut the recursion one step
template <std::uintmax_t base>
struct pow_struct<1, base>
{
static constexpr uintmax_t value=base;
};
template <std::uintmax_t base>
struct pow_struct<0,base>
{
static constexpr uintmax_t value=1;
};
The build token
template <uint vmajor, uint vminor, uint build>
struct build_token {
constexpr uintmax_t value=
vmajor*pow_struct<9>::value
+ vminor*pow_struct<6>::value
+ build_number
;
}
In the upcoming C++20 there will be consteval specifier.
consteval - specifies that a function is an immediate function, that is, every call to the function must produce a compile-time constant
Since now we have C++17, there is an easier solution:
template <auto V>
struct constant {
constexpr static decltype(V) value = V;
};
The key is that non-type arguments can be declared as auto. If you are using standards before C++17 you may have to use std::integral_constant. There is also a proposal about the constant helper class.
An example:
template <auto V>
struct constant {
constexpr static decltype(V) value = V;
};
constexpr uint64_t factorial(int n) {
if (n <= 0) {
return 1;
}
return n * factorial(n - 1);
}
int main() {
std::cout << "20! = " << constant<factorial(20)>::value << std::endl;
return 0;
}
Have your function take template parameters instead of arguments and implement your logic in a lambda.
#include <iostream>
template< uint64_t N >
constexpr uint64_t factorial() {
// note that we need to pass the lambda to itself to make the recursive call
auto f = []( uint64_t n, auto& f ) -> uint64_t {
if ( n < 2 ) return 1;
return n * f( n - 1, f );
};
return f( N, f );
}
using namespace std;
int main() {
cout << factorial<5>() << std::endl;
}
Lets say that you have a function which generates some security token for your application, such as some hash salt, or maybe a symetric or asymetric key.
Now lets say that you have this function in your C++ as a constexpr and that you generate keys for your build based on some information (like, the build number, a timestamp, something else).
You being a diligent programmer make sure and call this in the appropriate ways to ensure it's only called at compile time, and thus the dead stripper removes the code from the final executable.
However, you can't ever be sure that someone else isn't going to call it in an unsafe way, or that maybe the compiler won't strip the function out, and then your security token algorithm will become public knowledge, making it more easy for would be attackers to guess future tokens.
Or, security aside, let's say the function takes a long time to execute and you want to make sure it never happens during runtime and causes a bad user experience for your end users.
Are there any ways to ensure that a constexpr function can never be called at runtime? Or alternately, throwing an assert or similar at runtime would be ok, but not as ideal obviously as a compile error would be.
I've heard that there is some way involving throwing an exception type that doesn't exist, so that if the constexpr function is not deadstripped out, you'll get a linker error, but have heard that this only works on some compilers.
Distantly related question: Force constexpr to be evaluated at compile time
In C++20 you can just replace constexpr by consteval to enforce a function to be always evaluated at compile time.
Example:
int rt_function(int v){ return v; }
constexpr int rt_ct_function(int v){ return v; }
consteval int ct_function(int v){ return v; }
int main(){
constexpr int ct_value = 1; // compile value
int rt_value = 2; // runtime value
int a = rt_function(ct_value);
int b = rt_ct_function(ct_value);
int c = ct_function(ct_value);
int d = rt_function(rt_value);
int e = rt_ct_function(rt_value);
int f = ct_function(rt_value); // ERROR: runtime value
constexpr int g = rt_function(ct_value); // ERROR: runtime function
constexpr int h = rt_ct_function(ct_value);
constexpr int i = ct_function(ct_value);
}
Pre C++20 workaround
You can enforce the use of it in a constant expression:
#include<utility>
template<typename T, T V>
constexpr auto ct() { return V; }
template<typename T>
constexpr auto func() {
return ct<decltype(std::declval<T>().value()), T{}.value()>();
}
template<typename T>
struct S {
constexpr S() {}
constexpr T value() { return T{}; }
};
template<typename T>
struct U {
U() {}
T value() { return T{}; }
};
int main() {
func<S<int>>();
// won't work
//func<U<int>>();
}
By using the result of the function as a template argument, you got an error if it can't be solved at compile-time.
A theoretical solution (as templates should be Turing complete) - don't use constexpr functions and fall back onto the good-old std=c++0x style of computing using exclusively struct template with values. For example, don't do
constexpr uintmax_t fact(uint n) {
return n>1 ? n*fact(n-1) : (n==1 ? 1 : 0);
}
but
template <uint N> struct fact {
uintmax_t value=N*fact<N-1>::value;
}
template <> struct fact<1>
uintmax_t value=1;
}
template <> struct fact<0>
uintmax_t value=0;
}
The struct approach is guaranteed to be evaluated exclusively at compile time.
The fact the guys at boost managed to do a compile time parser is a strong signal that, albeit tedious, this approach should be feasible - it's a one-off cost, maybe one can consider it an investment.
For example:
to power struct:
// ***Warning: note the unusual order of (power, base) for the parameters
// *** due to the default val for the base
template <unsigned long exponent, std::uintmax_t base=10>
struct pow_struct
{
private:
static constexpr uintmax_t at_half_pow=pow_struct<exponent / 2, base>::value;
public:
static constexpr uintmax_t value=
at_half_pow*at_half_pow*(exponent % 2 ? base : 1)
;
};
// not necessary, but will cut the recursion one step
template <std::uintmax_t base>
struct pow_struct<1, base>
{
static constexpr uintmax_t value=base;
};
template <std::uintmax_t base>
struct pow_struct<0,base>
{
static constexpr uintmax_t value=1;
};
The build token
template <uint vmajor, uint vminor, uint build>
struct build_token {
constexpr uintmax_t value=
vmajor*pow_struct<9>::value
+ vminor*pow_struct<6>::value
+ build_number
;
}
In the upcoming C++20 there will be consteval specifier.
consteval - specifies that a function is an immediate function, that is, every call to the function must produce a compile-time constant
Since now we have C++17, there is an easier solution:
template <auto V>
struct constant {
constexpr static decltype(V) value = V;
};
The key is that non-type arguments can be declared as auto. If you are using standards before C++17 you may have to use std::integral_constant. There is also a proposal about the constant helper class.
An example:
template <auto V>
struct constant {
constexpr static decltype(V) value = V;
};
constexpr uint64_t factorial(int n) {
if (n <= 0) {
return 1;
}
return n * factorial(n - 1);
}
int main() {
std::cout << "20! = " << constant<factorial(20)>::value << std::endl;
return 0;
}
Have your function take template parameters instead of arguments and implement your logic in a lambda.
#include <iostream>
template< uint64_t N >
constexpr uint64_t factorial() {
// note that we need to pass the lambda to itself to make the recursive call
auto f = []( uint64_t n, auto& f ) -> uint64_t {
if ( n < 2 ) return 1;
return n * f( n - 1, f );
};
return f( N, f );
}
using namespace std;
int main() {
cout << factorial<5>() << std::endl;
}
How do you insert a transformed integer_sequence (or similar since I am targeting C++11) into a variadic template argument?
For example I have a class that represents a set of bit-wise flags (shown below). It is made using a nested-class because you cannot have two variadic template arguments for the same class. It would be used like typedef Flags<unsigned char, FLAG_A, FLAG_B, FLAG_C>::WithValues<0x01, 0x02, 0x04> MyFlags. Typically, they will be used with the values that are powers of two (although not always, in some cases certain combinations would be made, for example one could imagine a set of flags like Read=0x1, Write=0x2, and ReadWrite=0x3=0x1|0x2). I would like to provide a way to do typedef Flags<unsigned char, FLAG_A, FLAG_B, FLAG_C>::WithDefaultValues MyFlags.
template<class _B, template <class,class,_B> class... _Fs>
class Flags
{
public:
template<_B... _Vs>
class WithValues :
public _Fs<_B, Flags<_B,_Fs...>::WithValues<_Vs...>, _Vs>...
{
// ...
};
};
I have tried the following without success (placed inside the Flags class, outside the WithValues class):
private:
struct _F {
// dummy class which can be given to a flag-name template
template <_B _V> inline constexpr
explicit _F(std::integral_constant<_B, _V>) { } };
// we count the flags, but only in a dummy way
static constexpr unsigned _count = sizeof...(_Fs<_B, _F, 1>);
static inline constexpr
_B pow2(unsigned exp, _B base = 2, _B result = 1) {
return exp < 1 ?
result :
pow2(exp/2,
base*base,
(exp % 2) ? result*base : result);
}
template <_B... _Is> struct indices {
using next = indices<_Is..., sizeof...(_Is)>;
using WithPow2Values = WithValues<pow2(_Is)...>;
};
template <unsigned N> struct build_indices {
using type = typename build_indices<N-1>::type::next;
};
template <> struct build_indices<0> {
using type = indices<>;
};
//// Another attempt
//template < _B... _Is> struct indices {
// using WithPow2Values = WithValues<pow2(_Is)...>;
//};
//template <unsigned N, _B... _Is> struct build_indices
// : build_indices<N-1, N-1, _Is...> { };
//template < _B... _Is> struct build_indices<0, _Is...>
// : indices<_Is...> { };
public:
using WithDefaultValues =
typename build_indices<_count>::type::WithPow2Values;
Of course, I would be willing to have any other alternatives to the whole situation (supporting both flag names and values in the same template set, etc).
I have included a "working" example at ideone: http://ideone.com/NYtUrg - by "working" I mean compiles fine without using default values but fails with default values (there is a #define to switch between them).
Thanks!
So I figured it out on my own, I guess I posted too soon.
I was able to get around the error generated with the given code with a dummy template argument in the build_indices and indices template classes above. The argument has to be a typename as they currently have variadic integral types.
(Note: still using the improper _F names here - I have corrected my personal code to not use these reserved name - thanks for the tip)
Here is a working solution which results in the WithValues<...> template to be filled with powers of 2 (1, 2, 4, 8, 16, ...) based on the size of the Flags variadic template.
private:
// dummy class which can be given to a flag-name template
struct _F
{
template <_B _V> inline constexpr
explicit _F(std::integral_constant<_B, _V>) { }
};
// we count the flags, but only in a dummy way
static constexpr unsigned _count = sizeof...(_Fs<_B, _F, 1>);
static inline constexpr
_B pow2(unsigned exp, _B base = 2, _B result = 1)
{
return exp < 1 ?
result :
pow2(exp/2, base*base, (exp % 2) ? result*base : result);
}
template <class dummy, _B... _Is> struct indices
{
using next = indices<dummy, _Is..., sizeof...(_Is)>;
using WithPow2Values = WithValues<pow2(_Is)...>;
};
template <class dummy, unsigned N> struct build_indices
{
using type = typename build_indices<dummy, N-1>::type::next;
};
template <class dummy> struct build_indices<dummy, 0>
{
using type = indices<dummy>;
};
public:
using WithDefaultValues =
typename build_indices<void, _count>::type::WithPow2Values;
I have a template for a data structure / container optimized for sizes in the power of 2 and will not work properly unless the size parameter is a power 2.
template <typename T, unsigned int __pow2int>
class CustomContainer {
}
What is the best way / is it even possible... to enforce a compile-time check to ensure that __pow2int is a power of 2?
I'm new to C++ and I've been looking at pages like this: http://www.stroustrup.com/bs_faq2.html#constraints but I find syntax such as...
static void constraints(T1 a, T2 b) { T2 c = a; b = a; }
Can_copy() { void(*p)(T1,T2) = constraints; }
totally confusing, I'm not even sure if that is the way to go about what I'm trying to achieve, I got totally lost trying to follow that FAQ.
The way I see it, perhaps I should declare a user-defined type that only creates power of 2 integers and use that as a template type?
I tried to implement it, but I ended up getting a "non-type template parameter cannot...." error.
class intb2 {
const std::uint32_t _output;
public:
intb2(std::uint8_t bit) : _output([&]() {
uint8_t rbit=(bit == 0) ? 1 : bit;
std::uint32_t i=1;
return (i << (rbit-1));
}()) {}
const std::uint32_t& operator()() {
return _output;
}
};
template <typename T, intb2 __pow2int>
class CustomContainer {....
The usual trick to check whether a number x is a power of two is to bitwise-and it with x-1 and see if the result is zero:
x != 0 && (x & (x−1)) == 0
Now use that in a static assert declaration:
template <typename T, unsigned int N>
class CustomContainer {
static_assert( N != 0 && (N & (N−1)) == 0 , "Not a power of two!");
};
The easiest way to solve this is to use the exponent as the argument.
template <typename T, unsigned int exp>
class CustomContainer
{
unsigned int __pow2int = 1<<exp;
};