考虑:
List<String> someList = new ArrayList<String>();
// add "monkey", "donkey", "skeleton key" to someList
for (String item : someList) {
System.out.println(item);
}
如果不使用for each语法,等价的for循环会是什么样子?
考虑:
List<String> someList = new ArrayList<String>();
// add "monkey", "donkey", "skeleton key" to someList
for (String item : someList) {
System.out.println(item);
}
如果不使用for each语法,等价的for循环会是什么样子?
当前回答
public static Boolean Add_Tag(int totalsize)
{
List<String> fullst = new ArrayList<String>();
for(int k=0; k<totalsize; k++)
{
fullst.addAll();
}
}
其他回答
这是一个等价的表达式。
for(Iterator<String> sit = someList.iterator(); sit.hasNext(); ) {
System.out.println(sit.next());
}
nsayer的答案暗示了这一点,但值得注意的是,当“someList”是实现java.lang.Iterable的任何东西时,OP的for(..)语法都会起作用——它不必是列表,也不必是java.util的集合。因此,甚至您自己的类型也可以与此语法一起使用。
在Java5中添加的for each循环(也称为“增强的for循环”)相当于使用Java.util.Interator——这是一个语法糖。因此,在逐个按顺序读取每个元素时,应始终在迭代器上选择for each,因为它更方便和简洁。
对于每个
for (int i : intList) {
System.out.println("An element in the list: " + i);
}
迭代器
Iterator<Integer> intItr = intList.iterator();
while (intItr.hasNext()) {
System.out.println("An element in the list: " + intItr.next());
}
在某些情况下,您必须直接使用迭代器。例如,在使用For each时尝试删除元素可能会(将?)导致ConcurrentModificationException。
对于每个与For循环:基本区别
for循环和for each之间的唯一实际区别是,在可索引对象的情况下,您无权访问索引。需要基本for循环时的示例:
for (int i = 0; i < array.length; i++) {
if(i < 5) {
// Do something special
} else {
// Do other stuff
}
}
尽管您可以手动为每个创建单独的索引int变量,
int idx = -1;
for (int i : intArray) {
idx++;
...
}
…不建议这样做,因为变量范围并不理想,基本的for循环只是这个用例的标准和预期格式。
对于每个与For循环:性能
访问集合时,for each比基本for循环的数组访问快得多。然而,当访问数组时(至少是使用原始数组和包装数组时),通过索引访问的速度要快得多。
为基元int数组的迭代器和索引访问之间的差异计时
在访问int或Integer数组时,索引比迭代器快23-40%。这是本文底部测试类的输出,它将100个元素的原始int数组中的数字相加(a是迭代器,B是索引):
[C:\java_code\]java TimeIteratorVsIndexIntArray 1000000
Test A: 358,597,622 nanoseconds
Test B: 269,167,681 nanoseconds
B faster by 89,429,941 nanoseconds (24.438799231635727% faster)
[C:\java_code\]java TimeIteratorVsIndexIntArray 1000000
Test A: 377,461,823 nanoseconds
Test B: 278,694,271 nanoseconds
B faster by 98,767,552 nanoseconds (25.666236154695838% faster)
[C:\java_code\]java TimeIteratorVsIndexIntArray 1000000
Test A: 288,953,495 nanoseconds
Test B: 207,050,523 nanoseconds
B faster by 81,902,972 nanoseconds (27.844689860906513% faster)
[C:\java_code\]java TimeIteratorVsIndexIntArray 1000000
Test A: 375,373,765 nanoseconds
Test B: 283,813,875 nanoseconds
B faster by 91,559,890 nanoseconds (23.891659337194227% faster)
[C:\java_code\]java TimeIteratorVsIndexIntArray 1000000
Test A: 375,790,818 nanoseconds
Test B: 220,770,915 nanoseconds
B faster by 155,019,903 nanoseconds (40.75164734599769% faster)
[C:\java_code\]java TimeIteratorVsIndexIntArray 1000000
Test A: 326,373,762 nanoseconds
Test B: 202,555,566 nanoseconds
B faster by 123,818,196 nanoseconds (37.437545972215744% faster)
我还对Integer数组运行了这个,索引仍然是明显的赢家,但只快了18%到25%。
对于集合,迭代器比索引更快
然而,对于整数列表来说,迭代器显然是赢家。只需将测试类中的int数组更改为:
List<Integer> intList = Arrays.asList(new Integer[] {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, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100});
并对测试函数进行必要的更改(int[]到List<Integer>,length到size()等):
[C:\java_code\]java TimeIteratorVsIndexIntegerList 1000000
Test A: 3,429,929,976 nanoseconds
Test B: 5,262,782,488 nanoseconds
A faster by 1,832,852,512 nanoseconds (34.326681820485675% faster)
[C:\java_code\]java TimeIteratorVsIndexIntegerList 1000000
Test A: 2,907,391,427 nanoseconds
Test B: 3,957,718,459 nanoseconds
A faster by 1,050,327,032 nanoseconds (26.038700083921256% faster)
[C:\java_code\]java TimeIteratorVsIndexIntegerList 1000000
Test A: 2,566,004,688 nanoseconds
Test B: 4,221,746,521 nanoseconds
A faster by 1,655,741,833 nanoseconds (38.71935684115413% faster)
[C:\java_code\]java TimeIteratorVsIndexIntegerList 1000000
Test A: 2,770,945,276 nanoseconds
Test B: 3,829,077,158 nanoseconds
A faster by 1,058,131,882 nanoseconds (27.134122749113843% faster)
[C:\java_code\]java TimeIteratorVsIndexIntegerList 1000000
Test A: 3,467,474,055 nanoseconds
Test B: 5,183,149,104 nanoseconds
A faster by 1,715,675,049 nanoseconds (32.60101667104192% faster)
[C:\java_code\]java TimeIteratorVsIndexIntList 1000000
Test A: 3,439,983,933 nanoseconds
Test B: 3,509,530,312 nanoseconds
A faster by 69,546,379 nanoseconds (1.4816434912159906% faster)
[C:\java_code\]java TimeIteratorVsIndexIntList 1000000
Test A: 3,451,101,466 nanoseconds
Test B: 5,057,979,210 nanoseconds
A faster by 1,606,877,744 nanoseconds (31.269164666060377% faster)
在一次测试中,它们几乎是等价的,但在集合中,迭代器获胜。
*这篇文章基于我在Stack Overflow上写的两个答案:
Java中for每个循环的用法和语法我应该使用迭代器还是forloop进行迭代?
更多信息:for-each循环和迭代器哪个更有效?
完整测试类
在阅读了关于堆栈溢出的问题后,我创建了一个比较类做任何两件事所需的时间:
import java.text.NumberFormat;
import java.util.Locale;
/**
<P>{@code java TimeIteratorVsIndexIntArray 1000000}</P>
@see <CODE><A HREF="https://stackoverflow.com/questions/180158/how-do-i-time-a-methods-execution-in-java">https://stackoverflow.com/questions/180158/how-do-i-time-a-methods-execution-in-java</A></CODE>
**/
public class TimeIteratorVsIndexIntArray {
public static final NumberFormat nf = NumberFormat.getNumberInstance(Locale.US);
public static final void main(String[] tryCount_inParamIdx0) {
int testCount;
// Get try-count from a command-line parameter
try {
testCount = Integer.parseInt(tryCount_inParamIdx0[0]);
}
catch(ArrayIndexOutOfBoundsException | NumberFormatException x) {
throw new IllegalArgumentException("Missing or invalid command line parameter: The number of testCount for each test. " + x);
}
//Test proper...START
int[] intArray = new int[] {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, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100};
long lStart = System.nanoTime();
for(int i = 0; i < testCount; i++) {
testIterator(intArray);
}
long lADuration = outputGetNanoDuration("A", lStart);
lStart = System.nanoTime();
for(int i = 0; i < testCount; i++) {
testFor(intArray);
}
long lBDuration = outputGetNanoDuration("B", lStart);
outputGetABTestNanoDifference(lADuration, lBDuration, "A", "B");
}
private static final void testIterator(int[] int_array) {
int total = 0;
for(int i = 0; i < int_array.length; i++) {
total += int_array[i];
}
}
private static final void testFor(int[] int_array) {
int total = 0;
for(int i : int_array) {
total += i;
}
}
//Test proper...END
//Timer testing utilities...START
public static final long outputGetNanoDuration(String s_testName, long l_nanoStart) {
long lDuration = System.nanoTime() - l_nanoStart;
System.out.println("Test " + s_testName + ": " + nf.format(lDuration) + " nanoseconds");
return lDuration;
}
public static final long outputGetABTestNanoDifference(long l_aDuration, long l_bDuration, String s_aTestName, String s_bTestName) {
long lDiff = -1;
double dPct = -1.0;
String sFaster = null;
if(l_aDuration > l_bDuration) {
lDiff = l_aDuration - l_bDuration;
dPct = 100.00 - (l_bDuration * 100.0 / l_aDuration + 0.5);
sFaster = "B";
}
else {
lDiff = l_bDuration - l_aDuration;
dPct = 100.00 - (l_aDuration * 100.0 / l_bDuration + 0.5);
sFaster = "A";
}
System.out.println(sFaster + " faster by " + nf.format(lDiff) + " nanoseconds (" + dPct + "% faster)");
return lDiff;
}
//Timer testing utilities...END
}
使用较旧的Java版本(包括Java7),可以使用foreach循环,如下所示。
List<String> items = new ArrayList<>();
items.add("A");
items.add("B");
items.add("C");
items.add("D");
items.add("E");
for(String item : items) {
System.out.println(item);
}
以下是Java8中使用for-each循环的最新方法(使用forEach+lambda表达式或方法引用循环List)。
Lambda公司
// Output: A,B,C,D,E
items.forEach(item->System.out.println(item));
方法参考
// Output: A,B,C,D,E
items.forEach(System.out::println);
有关详细信息,请参阅“Java 8 For Each examples”。
正如许多好答案所说,如果一个对象想要使用for each循环,它必须实现Iterable接口。
我将发布一个简单的示例,并尝试以不同的方式解释for each循环的工作原理。
对于每个循环示例:
public class ForEachTest {
public static void main(String[] args) {
List<String> list = new ArrayList<String>();
list.add("111");
list.add("222");
for (String str : list) {
System.out.println(str);
}
}
}
然后,如果我们使用javap来反编译这个类,我们将得到这个字节码示例:
public static void main(java.lang.String[]);
flags: ACC_PUBLIC, ACC_STATIC
Code:
stack=2, locals=4, args_size=1
0: new #16 // class java/util/ArrayList
3: dup
4: invokespecial #18 // Method java/util/ArrayList."<init>":()V
7: astore_1
8: aload_1
9: ldc #19 // String 111
11: invokeinterface #21, 2 // InterfaceMethod java/util/List.add:(Ljava/lang/Object;)Z
16: pop
17: aload_1
18: ldc #27 // String 222
20: invokeinterface #21, 2 // InterfaceMethod java/util/List.add:(Ljava/lang/Object;)Z
25: pop
26: aload_1
27: invokeinterface #29, 1 // InterfaceMethod java/util/List.iterator:()Ljava/util/Iterator;
从示例的最后一行可以看到,编译器将在编译时自动将for每个关键字的使用转换为Iterator的使用。这可以解释为什么对象(它没有实现Iterable接口)在尝试使用for each循环时会抛出异常。