I'm trying to gain a deeper understanding of how Swift copies value types:
The behavior you see in your code will always be as if a copy took
place. However, Swift only performs an actual copy behind the scenes
when it is absolutely necessary to do so.
To advance my understanding, I'd like to get the memory address of a value type. I tried unsafeAddressOf(), but this doesn't work with structs and it seems to cast Swift's standard library types to reference types (e.g. String is cast to NSString).
How can I get the memory address of a value type, like an instance of Int, or a custom struct in Swift?
According to Martin R' s answer
addressOf() cannot be used with struct variables. String is a struct, however, it is automatically bridged to NSString when passed to a function expecting an object.
According to nschum's answer, you can get the (stack) address of a struct, build-in type or object reference like this:
import UIKit
func address(o: UnsafePointer<Void>) -> Int {
return unsafeBitCast(o, Int.self)
}
func addressHeap<T: AnyObject>(o: T) -> Int {
return unsafeBitCast(o, Int.self)
}
struct myStruct {
var a: Int
}
class myClas {
}
//struct
var struct1 = myStruct(a: 5)
var struct2 = struct1
print(NSString(format: "%p", address(&struct1))) // -> "0x10f1fd430\n"
print(NSString(format: "%p", address(&struct2))) // -> "0x10f1fd438\n"
//String
var s = "A String"
var aa = s
print(NSString(format: "%p", address(&s))) // -> "0x10f43a430\n"
print(NSString(format: "%p", address(&aa))) // -> "0x10f43a448\n"
//Class
var class1 = myClas()
var class2 = class1
print(NSString(format: "%p", addressHeap(class1))) // -> 0x7fd5c8700970
print(NSString(format: "%p", addressHeap(class2))) // -> 0x7fd5c8700970
unsafeAddressOf(class1) //"UnsafePointer(0x7FD95AE272E0)"
unsafeAddressOf(class2) //"UnsafePointer(0x7FD95AE272E0)"
//Int
var num1 = 55
var num2 = num1
print(NSString(format: "%p", address(&num1))) // -> "0x10f1fd480\n"
print(NSString(format: "%p", address(&num2))) // -> "0x10f1fd488\n"
One thing I found is, if myStruct
has no value, the address will be retain same:
struct myStruct {
}
var struct1 = myStruct()
var struct2 = struct1
print(NSString(format: "%p", address(&struct1))) // -> ""0xa000000000070252\n""
print(NSString(format: "%p", address(&struct2))) // -> ""0xa000000000070252\n""
Swift 2.0 :
You can use this unsafeAddressOf(someObject)
or in Swift 3.0:
use withUnsafePointer(to: someObejct) { print("\($0)") }
I'm not sure if there's a "recommended" way to do that, but one method is to use withUnsafePointer(_:_:)
, like this:
var s: String = "foo"
withUnsafePointer(&s) { NSLog("\($0)") }
This printed 0x00007ffff52a011c8
on my machine.