Menu
Inspired by this.
Suppose we have a non empty std::vector<T> v;
is there any difference between
for (int i = v.size() - 1; i >= 0; i--) {
///Stuff.
}
and:
for (int i = v.size(); i--; ) {
///Stuff.
}
?
I mean, I wouldn't do that for readability sake, but who knows what could be useful in life...
(Here just a test to see they are equivalent)
Edit: The point is having i as index to access backward the vector (in a case where having the index was preferable to having iterators)
Edit 2: Out of curiosity: They end up having slightly different assembly code. See this and this.
723
One could think of a for loop
for (<decl-init> ; <condition> ; <post-adjust>) <body>
as a rough equivalent to this while loop:
<decl-init>;
while (condition) {
<body>;
<post-adjust>;
}
The biggest difference between the for and the rewrite above is the scope of variables declared in the <decl-init> block, but this is not important for the analysis below.
Rewriting both loops as a while loop gives you this:
int i = v.size() - 1;
while ( i >= 0 ) {
<body>;
i--;
}
vs.
int i = v.size();
while (i--) {
<body>;
}
As you can see, the only difference is that i is decremented before entering the iteration, and the condition is started with i greater by 1 than in the first loop. These two adjustments "cancel each other", making your loops equivalent from the technical point of view. Aesthetics is a different thing, though: conditions with side effects are harder to understand than "pure" ones, so the first loop is more readable.
53
It is a legal way of doing it, but using i-- as a condition in a for loop is a terrible idea in terms of readability.
The for loop is meant to consist of 3 parts - why make people lives harder?
People who read your code will be happy if you stick to the traditional for loop:
for (int i = v.size() - 1; i >= 0; i--) {
///Stuff.
}
If you love us? You can donate to us via Paypal or buy me a coffee so we can maintain and grow! Thank you!
Donate Us With
