Handling rectangle coordinates in a 2D space - c++

Question

I would like to know if there's such a very basic method to know where exactly the translated object is in 2D space. I would like to clarify my thoughts regarding OpenGL matrices.
Please have a time to take a look at this very few code examples:
class Rectangle
{
public:
//this function does amazing magic!
//..so continue reading
void draw();
inline bool contains(int x, int y) const {
return x >= this->x && x <= this->x + width &&
y >= this->y && y <= this->y + height;
}
int x;
int y;
uint32_t width;
uint32_t height;
protected:
std::vector<Rectangle *> children;
}
solution #1
void Rectangle::draw()
{
glPushMatrix();
glTranslatef(x, y, 0);
...
//popped out the previous matrix so I have a fresh one
glPopMatrix();
//outside the matrix of parent
//not logical (?)
for (auto &i : children)
{
i->draw();
}
}
solution #2
void Rectangle::draw()
{
glPushMatrix();
glTranslatef(x, y, 0);
...
//inside the matrix of parent
//good! the origin now is parent's coordinate.
for (auto &i : children)
{
//previous matrix is not popped out so the origin is parent's (x,y)
i->draw();
}
glPopMatrix();
}
In solution #1, it is simple and very easy to locate the coordinates of each Rectangle::children, however, its quite hard to manage the code and its not an obvious choice for me because those Rectangle::children's origin position must be their parent container, right?
On the other hand, I much preferred the solution #2 and it is an obvious choice because, for every children I will create, their origin will be the parent's location. But, it's very hard for me to manage those Rectangle::children's coordinates. For example, I want to have a hover effect applied on certain Rectangle::children, I would do in implementation #1 as simple as this:
solution #1
for (auto &i : children)
{
if (i->contains(mouse().x, mouse().y)
{
//hover effect!
}
}
solution #2
for (auto &i : children)
{
if (i->contains(mouse().x - this->x,
mouse().y - this->y)
{
//hover effect!
}
}
The main problem I'm really concern of is that I want to have a fixed function in the Rectangle class like isHovering() this will accept 0 number of parameters and I want it be a cv-qualified. What this will do is: will check that the mouse cursor is inside the rectangle.
I think I would mess up the code for this function if I implement the solution #2 here. What do you think? Is there any nice way algorithm existing there I didn't know yet? I would like to ask first before I proceed along way.
Or...there's an existing magic way to implement in the drawing function to be able to achieve this?
Thanks for reading!
Update
Suppose the drawing hierarchy I want to achieve is something like this:
glClear(GL_COLOR_BUFFER_BIT);
glLoadIdentity();
//one matrix for every parent-rectangle and its child-rectangle component
xPushMatrix();
//parent
glTranslatef(parent.x, parent.y, 0);
glDrawables();
//children
glTranslatef(child.x, child.y, 0);
glDrawables();
...to N child
xPopMatrix();
xSwapBuffers(window);
xPollEvents();
The problem: What is the right(efficient, clean code) way to find the coordinates of Rectangle::children ?
Hope you get what I mean.