考虑：
List<String> someList = new ArrayList<String>();
// add "monkey", "donkey", "skeleton key" to someListfor (String item : someList) {
System.out.println(item);
}如果不使用for each语法for循环的等价物会是什么样的？
for (Iterator<String> i = someIterable.iterator(); i.hasNext();) {
String item = i.next();
System.out.println(item);
}请注意，如果您需要使用i.remove();在你的循环中，或以某种方式访问实际的迭代器，你不能使用for ( : ) ：）习语，因为实际的迭代器只是推断。
正如 Denis Bueno 所指出的，此代码适用于实现Iterable接口的任何对象。
此外，如果for (:)习惯用法的右侧是array而不是Iterable对象，则内部代码使用 int 索引计数器并检查array.length 。请参阅Java 语言规范 。
每个的构造对数组也有效。例如
String[] fruits = new String[] { "Orange", "Apple", "Pear", "Strawberry" };
for (String fruit : fruits) {
// fruit is an element of the `fruits` array.
}这基本上相当于
for (int i = 0; i < fruits.length; i++) {
String fruit = fruits[i];
// fruit is an element of the `fruits` array.
}总的来说：
[nsayer]以下是发生的事情的较长形式：
for(Iterator<String> i = someList.iterator(); i.hasNext(); ) { String item = i.next(); System.out.println(item); }请注意，如果您需要使用 i.remove（）; 在你的循环中，或以某种方式访问实际的迭代器，你不能使用 for（：）习语，因为实际的迭代器只是推断。
这是由 nsayer 的回答所暗示的，但值得注意的是，当 “someList” 是实现 java.lang.Iterable 的任何东西时，OP 的 for（..）语法将起作用 - 它不一定是列表，也不一定是来自 java.util 中。因此，即使您自己的类型也可以使用此语法。
在Java 5 中添加的foreach循环 （也称为 “增强 for 循环”）相当于使用java.util.Iterator - 对于同一事物的语法糖。因此，当逐个读取每个元素并按顺序时，应始终在迭代器上选择foreach ，因为它更方便和简洁。
for(int i : intList) {
System.out.println("An element in the list: " + i);
}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Iterator 。例如，尝试在使用foreach删除元素可以（将？）导致ConcurrentModificationException 。
foreach与for ：基本差异 for和foreach之间唯一的实际区别是，在可索引对象的情况下，您无权访问索引。需要基本for循环的示例：
for(int i = 0; i < array.length; i++) {
if(i < 5) {
// Do something special
} else {
// Do other stuff
}
}虽然您可以使用foreach手动创建单独的索引 int-variable，
int idx = -1;
for(int i : intArray) {
idx++;
...
}不推荐，因为可变范围是不理想的，基本for循环简直就是这个用例的标准和预期的格式。
foreach与for ：性能当访问收藏，一foreach是显著速度比基本for循环的数组访问。但是，当访问数组时 - 至少使用原始数据包和包装器数组 - 通过索引进行访问的速度要快得多。
当访问索引比迭代更快 23- 40％ int或Integer阵列。这是本文底部测试类的输出，它将 100 个元素的原始数组中的数字相加（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)我还为Integer数组运行了这个，索引仍然是明显的赢家，但速度只有 18％到 25％。
但是，对于Integers List ，迭代器是明显的赢家。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 test-function（ int[]到List<Integer> ， length to size()等）进行必要的更改：
[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，它们几乎相同，但是使用集合，迭代器获胜。
* 这篇文章基于我在 Stack Overflow 上写的两个答案：
更多信息： 哪个更有效，for-each 循环或迭代器？
在 Stack Overflow 上阅读这个问题之后，我创建了这个比较 - 时间 - 任意两件事的类：
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
}