使用for、forin和block遍历NSArray的效率的比较

代码：

复制代码#import <Foundation/Foundation.h> extern uint64_t dispatch_benchmark(size_t count, void (^block)(void)); void processNumber(NSNumber *n); int main(int argc, const char * argv[]) { @autoreleasepool { // 创建一个数组并用随机数填充 NSUInteger const capacity = 15; NSMutableArray *array = [NSMutableArray arrayWithCapacity:capacity]; for (NSUInteger i = 0; i < capacity; ++i) { [array addObject:@(arc4random_uniform(100))]; } size_t count = 10000000; // 使用for循环遍历数组 uint64_t time1 = dispatch_benchmark(count, ^{ for (NSUInteger i = 0; i < capacity; ++i) { processNumber(array[i]); } }); NSLog(@"for = %lluns", time1); // 使用forin循环遍历数组 uint64_t time2 = dispatch_benchmark(count, ^{ for (NSNumber *num in array) { processNumber(num); } }); NSLog(@"forin = %lluns", time2); // 使用block遍历数组 uint64_t time3 = dispatch_benchmark(count, ^{ [array enumerateObjectsUsingBlock:^(NSNumber *num, NSUInteger idx, BOOL *stop) { processNumber(num); }]; }); NSLog(@"block = %lluns", time3); } return 0; } void processNumber(NSNumber *n) { // code... }

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
#import <Foundation/Foundation.h>
extern uint64_t dispatch_benchmark(size_t count, void (^block)(void));
void processNumber(NSNumber *n);
int main(int argc, const char * argv[]) {
@autoreleasepool {
// 创建一个数组并用随机数填充
NSUInteger const capacity = 15;
NSMutableArray *array = [NSMutableArray arrayWithCapacity:capacity];
for (NSUInteger i = 0; i < capacity; ++i) {
[array addObject:@(arc4random_uniform(100))];
}
size_t count = 10000000;
// 使用for循环遍历数组
uint64_t time1 = dispatch_benchmark(count, ^{
for (NSUInteger i = 0; i < capacity; ++i) {
processNumber(array[i]);
}
});
NSLog(@"for = %lluns", time1);
// 使用forin循环遍历数组
uint64_t time2 = dispatch_benchmark(count, ^{
for (NSNumber *num in array) {
processNumber(num);
}
});
NSLog(@"forin = %lluns", time2);
// 使用block遍历数组
uint64_t time3 = dispatch_benchmark(count, ^{
[array enumerateObjectsUsingBlock:^(NSNumber *num, NSUInteger idx, BOOL *stop) {
processNumber(num);
}];
});
NSLog(@"block = %lluns", time3);
}
return 0;
}
void processNumber(NSNumber *n) {
// code...
}

输出：

复制代码for = 326ns forin = 237ns block = 638ns

1
2
3
for = 326ns
forin = 237ns
block = 638ns

但是如果把所有的processNumber函数注释掉，输出的结果如下：

复制代码for = 20ns forin = 106ns block = 496ns

1
2
3
for = 20ns
forin = 106ns
block = 496ns

所以对于遍历数组，不能武断的得出某种方式效率一定最高的结论，实际的执行效果往往取决于代码编写的情况。