开发者

incompatible types in assignment of ‘float’ to ‘float [16]’

开发者 https://www.devze.com 2023-03-28 06:10 出处:网络
I have struct: struct mat4 { 开发者_如何学JAVAfloat m[16]; mat4(); ... float *getFloat(); } float *mat4::getFloat() {

I have struct:

struct mat4 {
    开发者_如何学JAVAfloat m[16];
    mat4();
    ...
    float *getFloat();
}

float *mat4::getFloat() {
    return m;
}

Now I want to make m equal to m from newly created matrix r:

void mat4::rotate(vec3 v) {
    mat4 r, rx, ry, rz;
    ...
    matrix calculations
    ...
    m = *r.getFloat();
}

But this gives me error "incompatible types in assignment of ‘float’ to ‘float [16]’" I have searched Google and tried different ways but no success so far. Please tell me how could I do that?


getFloat() returns a pointer. m is an array. If you want to copy all of what is returned into m you will need to use std::copy or std::memcpy:

std::memcpy(m, r.getFloat(), sizeof(m));

If you meant to get just one float from getFloat() you could do:

  • m[0] = *r.getFloat();


r.getFloat() is returning a pointer to a single float. This is dereferenced to give a single float, and then assigned to an array of 16 floats.

Assuming that m contains the entire state of mat4, you can use the built in assignment operator:

*this = r;

The compiler will automatically implement the struct dump/ memcpy to copy all 16 floats.


Don't use naked C arrays. This is C++, we can do a lot better:

#include <tr1/array>  // or <array> if you're in MSVC or GCC

typedef std::tr1::array<float, 16> myfloats;

int main()
{
  myfloats a, b;
  a[0] = /* ... fill the array */
  b = a; // done
}

You can also put the array into your matrix structure:

struct mat4
{
  typedef std::tr1::array<float, 16> myfloats;

  myfloats & data() { return m_data; }

  mat4(); /* ... etc.  ... */

private:
   myfloats m_data;
};

You should be able to just assign variables of type mat4 to each other directly!


Use std::copy as:

#include <algorithm>

std::copy(r.m, r.m+16, m);

Also, if you can use r.m directly, why call r.getFloat() then?


r.getFloat() returns a pointer to "its" m. Arrays cannot be copied with simple assignment; you would have to work either with memcpy() or memmove() or with a for loop over the array elements.

0

精彩评论

暂无评论...
验证码 换一张
取 消