Instead of using a vector of floats, i want to use a struct, so that it can store different types and is easier.
struct QuadVertex
{
glm::vec3 Position;
glm::vec4 Color;
glm::vec2 TexCoord;
int TexIndex;
};
But if i use this structure in a vertex buffer, c++ could use padding or alignment in the structure and openl wouldnt get the expected data. How can i safely use this structure to store vertices, and how should i modify the vertex attributes correctly?
Specifically, when setting up your vertex format and pointers in code, using sizeof for the stride in your glVertexAttribPointer calls will cause OpenGL to correctly set things up such that each vertex properly begins at the start of each new struct in your array.
Similarly, using offsetof for the position of each attribute will cause OpenGL to correctly retrieve each attribute from it's correct location in the struct.
This will work irrespective of padding or alignment, and will be portable across different architectures and compilers where these might differ. (Loading the data correctly into the structs is another issue, of course.)
