Question

Let’s say I have three arrays of same size. I have to do something with all objects. If I would use a standard C array, I would write something like

for (i = 0; i < size; i++) {
    doSomething(array1[i]);   // or [[array1 objectAtIndex:i] doSomething];
    doSomethingElse(array2[i]);   // or [[array2 objectAtIndex:i] doSomethingElse];
    doSomethingReallySpecial(array3[i]);   // or [[array3 objectAtIndex:i] doSomethingReallySpecial];
}

With Objective C we got more ways to cycle through objects in NSArray: fast enumeration, block-based enumeration and using enumerators. Which one should I use and why? What’s the difference?

Edit
Actually this question may be formulated like this: if one needs to use the index of an item of an array, which enumeration should be used?

Answer 1

You should use the one that most clearly expresses the intent of your code to readers of the code. Unless you have identified this loops as a significant performance bottleneck, ability to maintain your code is far more important.

That said, fast enumeration alone (i.e for(id i in array)) does not provide the index. If you don’t want to use a C-style loop, -[NSArray enumerateObjectsUsingBlock:] provides both the current item and the index and could be used to zip multiple arrays. I suspect, that for multiple arrays, the C-style approach will be often be effectively equivalent in performance—and perhaps clearer to the reader, but you should profile it to be sure.

What you’re looking for is a common idiom in functional programming and could certainly be implemented on top of GCD using blocks, but I don’t know of any implementations.