What differences exist between Objective-C categories and .NET extension methods?

I'm learning about Categories in Objective-C and they seem really similar to extension methods in .NET.

Are there any subtle differences or gotchas between the two that can cause problems when moving from one language to开发者_JAVA百科 the other, or are they for all intents and purposes identical concepts?

In my experience using the two concepts in C# and ObjC, I use the features for the same problems. Adding new instance or static methods to built-in types when subclassing either isn't allowed or ill advised.

I tend to try and avoid Categories/Extensions because they can be a code smell that you are relying on built-in classes too heavily instead of building your own data structures to represent your system.

With that said, I haven't run into any strange gotchas once you understand the syntactical differences.

In the .NET world, an extension method will be called even if the object instance is null. For example:

public static class Extensions {
    public static string CountCharacters(this string inputString) {
        if (inputString == null) return "None!";
        return inputString.Length.ToString;
    }
}
public void TestExtension() {
    string testString = null;
    string result = testString.CountCharacters(); // returns "None!"
}

In the Objective-C world, instance methods defined in a category will not be called on an nil object instance.

@implementation NSString (CountCharacters)
    - (NSString *)countCharacters {
        if (self)
            return [self length];
        return @"None!";
    }
@end
-(void)testCategoryMethod {
    NSString *testString = nil;
    NSString *result = [testString countCharacters];
    // 'result' will remain nil
}