just a thought about "doing the compilers job"

[.rhavin grobert]

guess you have the following:

class X {
int fn1(); // -\ lets assume those fn's
int fn2(); // >- do something independent
int fn3(); // -/ from each other
}

class Y {
public:
X& X() {return m_x;};
private:
X m_x;
};

class Z {
public:
Y& Y() {return m_y;};
private:
Y m_y;
};

_______________ _______________ _

now when i do something like this:

Z z;
CallWhateverFun ctionTakingThre eInts(z.y().x() .fn1,z.y().x(). fn2,z.y().x().f n3);

is it as effective as writing ...

Z z;
X& x = z.y().x();
CallWhateverFun ctionTakingThre eInts(x.fn1,x.f n2,x.fn3);

[Victor Bazarov]

Re: just a thought about "doing the compilers job"

..rhavin grobert wrote:

guess you have the following:
>
class X {
int fn1(); // -\ lets assume those fn's
int fn2(); // >- do something independent
int fn3(); // -/ from each other
}
>
class Y {
public:
X& X() {return m_x;};
private:
X m_x;
};
>
class Z {
public:
Y& Y() {return m_y;};
private:
Y m_y;
};
>
_______________ _______________ _
>
now when i do something like this:
>
Z z;
CallWhateverFun ctionTakingThre eInts(z.y().x() .fn1,z.y().x(). fn2,z.y().x().f n3);
>
is it as effective as writing ...
>
Z z;
X& x = z.y().x();
CallWhateverFun ctionTakingThre eInts(x.fn1,x.f n2,x.fn3);
>
>

Generally speaking, the compiler is usually smart enough to take
advantage of inlining of your functions and also to cache the common
subexpression, so, yes it would be as effective in terms of the program
performance (if it isn't, you probably want to chuck that compiler and
get a better one). However, from the readability point of view, the
latter version is waaaaay better.

V
--
Please remove capital 'A's when replying by e-mail
I do not respond to top-posted replies, please don't ask