Possible Duplicate:
Is literal creation of an NSMutableDictionary less efficient than the class helper method?
According to the WWDC video that introduces ObjectiveC literals, NSMutableArray
s can be initialized like so:
[NSMutableArray arrayWithArray:@[]];
but what if we were to do it like this:
[@[] mutableCopy];
I know that one is initializing the array, and the other is just providing a shallow copy; thus the memory management is different. But if we are using ARC, what are the disadvantages of using the latter? Is it more expensive? ARC probably handles both differently, but I'm just wondering if using mutableCopy
is 'worse' or not.
No. Just as how there isn't a syntax for creating an NSMutableString either. Mutable objects are not particularly suited to literal values.
The mutableCopy method returns the object created by implementing NSMutableCopying protocol's mutableCopyWithZone: By sending: NSString* myString; NSMutableString* newString = [myString mutableCopy]; The return value WILL be mutable.
In the former case, you end up with two objects added to the current autorelease pool: the NSArray literal and the NSMutableArray that you’re creating from it. In the latter, the literal NSArray gets added to the autorelease pool but the NSMutableArray copied from it does not. So the first case is slightly worse, performance-wise, in that draining the autorelease pool (e.g. at the end of the run-loop cycle) requires releasing both objects in succession, while in the second case the copy will get released separately from the autoreleased literal. In practice, the time difference will be infinitesimal.
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