Java对象容器
范型和类型安全容器
因为数组是固定大小的，不能符合程序原型时的动态元素的管理，因此Java提供了一系列的容器对象。在读取一般的数组或者集合，例如ArrayList的用法，非常简单，但是在添加元素到集合的时候并没有检查所有的元素类型是否相同，这样做的后果就是在取出元素的时候会出现错误，而且是运行时错误，如下：
//: holding/ApplesAndOrangesWithoutGenerics.java
// Simple container example (produces compiler warnings).
// {ThrowsException}
import java.util.*;
class Apple {
private static long counter;
private final long id = counter++;
public long id() { return id; }
}
class Orange {}
public class ApplesAndOrangesWithoutGenerics {
@SuppressWarnings("unchecked")
public static void main(String[] args) {
ArrayList apples = new ArrayList();
for(int i = 0; i < 3; i++)
apples.add(new Apple());
// Not prevented from adding an Orange to apples:
apples.add(new Orange());
for(int i = 0; i < apples.size(); i++)
((Apple)apples.get(i)).id();
// Orange is detected only at run time
}
} /* (Execute to see output) *///:~
于是我们可以使用范型来保证类型的安全，使得添加的元素属于同一种数据类型，避免了读取元素时候的元素类型不安全的问题，关于范型的用法如下：
//: holding/ApplesAndOrangesWithGenerics.java
import java.util.*;
public class ApplesAndOrangesWithGenerics {
public static void main(String[] args) {
ArrayList<Apple> apples = new ArrayList<Apple>();
for(int i = 0; i < 3; i++)
apples.add(new Apple());
// Compile-time error:
// apples.add(new Orange());
for(int i = 0; i < apples.size(); i++)
System.out.println(apples.get(i).id());
// Using foreach:
for(Apple c : apples)
System.out.println(c.id());
}
} /* Output:
0
1
2
0
1
2
*///:~
如果上塑类型统一，那么也可以将范型对应类型的子类型上塑添加到范型数组中
基本概念
Java的容器类分为两类：Collection，一系列元素，其中List中的元素按照插入的顺序进行排列；Set中不能有重复的元素；Queue按照先进先出的顺序进行数组的添加和读取。Map，一系列的键值集合，ArrayList使用数字下表进行元素读取，因此从一定意义上来说也是一种键值集合；map允许通过一种对象来查找另一种对象。Maps是一个非常有力的编程工具。
在使用的时候通常是与这些集合的接口的进行交互，而在创建对象的时候才会去使用具体的类型，如List<Apple> apples = new ArrayList<Apple>();
添加一组元素
在Arrays和Collections中有添加一组元素的方法，Collection.Arrays.asList()可以将一个数组或者逗号分开的一列元素转换为List对象。Collections.addAll()将一个Collection对象或者一个数组或者一系列用都好分开的元素添加到Collection当中。如下面的例子：
//: holding/AddingGroups.java
// Adding groups of elements to Collection objects.
import java.util.*;
public class AddingGroups {
public static void main(String[] args) {
Collection<Integer> collection =new ArrayList<Integer>(Arrays.asList(1, 2, 3, 4, 5));
Integer[] moreInts = { 6, 7, 8, 9, 10 };
collection.addAll(Arrays.asList(moreInts));
// Runs significantly faster, but you can’t
// construct a Collection this way:
Collections.addAll(collection, 11, 12, 13, 14, 15);
Collections.addAll(collection, moreInts);
// Produces a list "backed by" an array:
List<Integer> list = Arrays.asList(16, 17, 18, 19, 20);
list.set(1, 99); // OK -- modify an element
// list.add(21); // Runtime error because the
// underlying array cannot be resized.
}
} ///:~
Collection的构造器可以接收Collection对象来初始化自己，但是Collections.addAll运行速度更快，因此更推荐使用默认构造器创建Collection然后再调用Collections.addAll()。Collection.addAll方法只能接受Collection对象作为参数，但是Array.asList或者Collections.addAll()使用的是参数列表。可以直接使用Arrays.asList，将得到的对象作为List使用，但是底层仍然是作为数组来表示的，当你试图添加或者删除元素的时候，实际上是要试图改变数组的大小，因此会得到一个Unsuported Operation运行时错误。
Arrays.asList的一个缺陷就是他对传入的类型做最好的猜测，而不管你指定的类型如何，有时候这样会产生错误，如下：
//: holding/AsListInference.java
// Arrays.asList() makes its best guess about type.
import java.util.*;
class Snow {}
class Powder extends Snow {}
class Light extends Powder {}
class Heavy extends Powder {}
class Crusty extends Snow {}
class Slush extends Snow {}
public class AsListInference {
public static void main(String[] args) {
List<Snow> snow1 = Arrays.asList(new Crusty(), new Slush(), new Powder());
//Won’t compile:
//List<Snow> snow2 = Arrays.asList(new Light(), new Heavy());
//Compiler says:
//found: java.util.List<Powder>
//required: java.util.List<Snow>
// Collections.addAll() doesn’t get confused:
List<Snow> snow3 = new ArrayList<Snow>();
Collections.addAll(snow3, new Light(), new Heavy());
// Give a hint using an explicit type argument specification:
List<Snow> snow4 = Arrays.<Snow>asList(new Light(), new Heavy());
}
} ///:~
容器打印
//: holding/PrintingContainers.java
// Containers print themselves automatically.
import java.util.*;
import static net.mindview.util.Print.*;
public class PrintingContainers {
static Collection fill(Collection<String> collection) {
collection.add("rat");
collection.add("cat");
collection.add("dog");
collection.add("dog");
return collection;
}
static Map fill(Map<String,String> map) {
map.put("rat", "Fuzzy");
map.put("cat", "Rags");
map.put("dog", "Bosco");
map.put("dog", "Spot");
return map;
}
public static void main(String[] args) {
print(fill(new ArrayList<String>()));
print(fill(new LinkedList<String>()));
print(fill(new HashSet<String>()));
print(fill(new TreeSet<String>()));
print(fill(new LinkedHashSet<String>()));
print(fill(new HashMap<String,String>()));
print(fill(new TreeMap<String,String>()));
print(fill(new LinkedHashMap<String,String>()));
}
} /* Output:
[rat, cat, dog, dog]
[rat, cat, dog, dog]
[dog, cat, rat]
Holding Your Objects
281
[cat, dog, rat]
[rat, cat, dog]
{dog=Spot, cat=Rags, rat=Fuzzy}
{cat=Rags, dog=Spot, rat=Fuzzy}
{rat=Fuzzy, cat=Rags, dog=Spot}
*///:~
List
List按照添加的顺序存放元素，List接口比Collection接口增加了允许插入和删除的方法，List有两种，ArrayList随机获取数据较快，但是在插入和一出元素的时候较慢；LinkedList用来顺序获取元素，插入和移除元素较快，但是随机获取元素较慢，但是比ArrayList有更多的特性：
//: holding/ListFeatures.java
import typeinfo.pets.*;
import java.util.*;
import static net.mindview.util.Print.*;
public class ListFeatures {
public static void main(String[] args) {
Random rand = new Random(47);
List<Pet> pets = Pets.arrayList(7);
print("1: " + pets);
Hamster h = new Hamster();
pets.add(h); // Automatically resizes
print("2: " + pets);
print("3: " + pets.contains(h));
pets.remove(h); // Remove by object
Pet p = pets.get(2);
print("4: " + p + " " + pets.indexOf(p));
Pet cymric = new Cymric();
print("5: " + pets.indexOf(cymric));
print("6: " + pets.remove(cymric));
// Must be the exact object:
print("7: " + pets.remove(p));
print("8: " + pets);
pets.add(3, new Mouse()); // Insert at an index
print("9: " + pets);
List<Pet> sub = pets.subList(1, 4);
print("subList: " + sub);
print("10: " + pets.containsAll(sub));
Collections.sort(sub); // In-place sort
print("sorted subList: " + sub);
// Order is not important in containsAll():
print("11: " + pets.containsAll(sub));
Collections.shuffle(sub, rand); // Mix it up
print("shuffled subList: " + sub);
print("12: " + pets.containsAll(sub));
List<Pet> copy = new ArrayList<Pet>(pets);
sub = Arrays.asList(pets.get(1), pets.get(4));
print("sub: " + sub);
copy.retainAll(sub);
print("13: " + copy);
copy = new ArrayList<Pet>(pets); // Get a fresh copy
copy.remove(2); // Remove by index
print("14: " + copy);
copy.removeAll(sub); // Only removes exact objects
print("15: " + copy);
copy.set(1, new Mouse()); // Replace an element
print("16: " + copy);
copy.addAll(2, sub); // Insert a list in the middle
print("17: " + copy);
print("18: " + pets.isEmpty());
pets.clear(); // Remove all elements
print("19: " + pets);
print("20: " + pets.isEmpty());
pets.addAll(Pets.arrayList(4));
print("21: " + pets);
Object[] o = pets.toArray();
print("22: " + o[3]);
Pet[] pa = pets.toArray(new Pet[0]);
print("23: " + pa[3].id());
}
} /* Output:
1: [Rat, Manx, Cymric, Mutt, Pug, Cymric, Pug]
2: [Rat, Manx, Cymric, Mutt, Pug, Cymric, Pug, Hamster]
3: true
4: Cymric 2
5: -1
6: false
7: true
8: [Rat, Manx, Mutt, Pug, Cymric, Pug]
9: [Rat, Manx, Mutt, Mouse, Pug, Cymric, Pug]
subList: [Manx, Mutt, Mouse]
10: true
sorted subList: [Manx, Mouse, Mutt]
11: true
shuffled subList: [Mouse, Manx, Mutt]
12: true
sub: [Mouse, Pug]
13: [Mouse, Pug]
14: [Rat, Mouse, Mutt, Pug, Cymric, Pug]
15: [Rat, Mutt, Cymric, Pug]
16: [Rat, Mouse, Cymric, Pug]
17: [Rat, Mouse, Mouse, Pug, Cymric, Pug]
18: false
19: []
20: true
21: [Manx, Cymric, Rat, EgyptianMau]
22: EgyptianMau
284
Thinking in Java
Bruce Eckel
23: 14
*///:~
历遍器
有些时候我们不知道我们将要使用的容器类型，或者说我们要对多种容器进行相同的历遍操作，这个时候如何做到代码与容器类型的分离，以便实现代码的重用呢？Java中提供了历遍器可以有效地将操作与容器类型相分离，如下：
//: holding/CrossContainerIteration.java
import typeinfo.pets.*;
import java.util.*;
public class CrossContainerIteration {
public static void display(Iterator<Pet> it) {
while(it.hasNext()) {
Pet p = it.next();
System.out.print(p.id() + ":" + p + " ");
}
System.out.println();
}
public static void main(String[] args) {
ArrayList<Pet> pets = Pets.arrayList(8);
LinkedList<Pet> petsLL = new LinkedList<Pet>(pets);
HashSet<Pet> petsHS = new HashSet<Pet>(pets);
TreeSet<Pet> petsTS = new TreeSet<Pet>(pets);
display(pets.iterator());
display(petsLL.iterator());
display(petsHS.iterator());
display(petsTS.iterator());
}
} /* Output:
0:Rat 1:Manx 2:Cymric 3:Mutt 4:Pug 5:Cymric 6:Pug 7:Manx
0:Rat 1:Manx 2:Cymric 3:Mutt 4:Pug 5:Cymric 6:Pug 7:Manx
4:Pug 6:Pug 3:Mutt 1:Manx 5:Cymric 7:Manx 2:Cymric 0:Rat
5:Cymric 2:Cymric 7:Manx 1:Manx 3:Mutt 6:Pug 4:Pug 0:Rat
*///:~
Iterator只能实现的功能有准备读取序列中的第一个元素；获取下一个元素；查看是否还有下一个元素；移除刚刚获取到的元素。ListIterator比Iterator功能更加强大，但是只对List有作用，Iterator只能前向移动，但是ListIterator是双向的，可以获取当前元素的前一个和下一个元素，可以设置当前元素的值，可以使用ListIterator来准备读取第一个元素，也可以通过传递索引来从指定的索引处开始读取，如下：
//: holding/ListIteration.java
import typeinfo.pets.*;
import java.util.*;
public class ListIteration {
public static void main(String[] args) {
List<Pet> pets = Pets.arrayList(8);
ListIterator<Pet> it = pets.listIterator();
while(it.hasNext())
System.out.print(it.next() + ", " + it.nextIndex() + ", " + it.previousIndex() + "; ");
System.out.println();
// Backwards:
while(it.hasPrevious())
System.out.print(it.previous().id() + " ");
System.out.println();
System.out.println(pets);
it = pets.listIterator(3);
while(it.hasNext()) {
it.next();
it.set(Pets.randomPet());
}
System.out.println(pets);
}
} /* Output:
Rat, 1, 0; Manx, 2,
5; Pug, 7, 6; Manx,
7 6 5 4 3 2 1 0
[Rat, Manx, Cymric,
[Rat, Manx, Cymric,
*///:
LinkedList
LinkedList继承自List接口，但是在插入和移除方面更加有效率，但是在随机读取方面效率较低，LinkedList添加了一些方法使得自身的性能更像是队列或者双向队列：
//: holding/LinkedListFeatures.java
import typeinfo.pets.*;
import java.util.*;
import static net.mindview.util.Print.*;
public class LinkedListFeatures {
public static void main(String[] args) {
LinkedList<Pet> pets = new LinkedList<Pet>(Pets.arrayList(5));
print(pets);
// Identical:
print("pets.getFirst(): " + pets.getFirst());
print("pets.element(): " + pets.element());
// Only differs in empty-list behavior:
print("pets.peek(): " + pets.peek());
// Identical; remove and return the first element:
print("pets.remove(): " + pets.remove());
print("pets.removeFirst(): " + pets.removeFirst());
// Only differs in empty-list behavior:
print("pets.poll(): " + pets.poll());
print(pets);
pets.addFirst(new Rat());
print("After addFirst(): " + pets);
pets.offer(Pets.randomPet());
print("After offer(): " + pets);
pets.add(Pets.randomPet());
print("After add(): " + pets);
pets.addLast(new Hamster());
print("After addLast(): " + pets);
print("pets.removeLast(): " + pets.removeLast());
}
} /* Output:
[Rat, Manx, Cymric, Mutt, Pug]
pets.getFirst(): Rat
pets.element(): Rat
pets.peek(): Rat
pets.remove(): Rat
pets.removeFirst(): Manx
pets.poll(): Cymric
[Mutt, Pug]
After addFirst(): [Rat, Mutt, Pug]
After offer(): [Rat, Mutt, Pug, Cymric]
After add(): [Rat, Mutt, Pug, Cymric, Pug]
After addLast(): [Rat, Mutt, Pug, Cymric, Pug, Hamster]
pets.removeLast(): Hamster
*///:~
Stack
栈是一种先进后出的数据结构，LinkedList有一些方法可以直接应用于栈，因此可以使用LinkedList而不是使用栈，但是有些时候栈可以更好的表达我们想要的东西：
//: net/mindview/util/Stack.java
// Making a stack from a LinkedList.
package net.mindview.util;
import java.util.LinkedList;
public class Stack<T> {
private LinkedList<T> storage = new LinkedList<T>();
public void push(T v) { storage.addFirst(v); }
public T peek() { return storage.getFirst(); }
public T pop() { return storage.removeFirst(); }
public boolean empty() { return storage.isEmpty(); }
public String toString() { return storage.toString(); }
} ///:~
如果想要获得栈的性质，这里使用继承是不合适的。因为这样派生出来的类不仅仅具有栈的功能，还具备了LinkedList的其他功能，下面就是这个我们自己的栈和java.util中的Stack的用法：
//: holding/StackCollision.java
import net.mindview.util.*;
public class StackCollision {
public static void main(String[] args) {
net.mindview.util.Stack<String> stack = new net.mindview.util.Stack<String>();
for(String s : "My dog has fleas".split(" "))
stack.push(s);
while(!stack.empty())
System.out.print(stack.pop() + " ");
System.out.println();
java.util.Stack<String> stack2 = new java.util.Stack<String>();
for(String s : "My dog has fleas".split(" "))
stack2.push(s);
while(!stack2.empty())
System.out.print(stack2.pop() + " ");
}
} /* Output:
fleas has dog My
fleas has dog My
*///:~
虽然java.util.Stack存在，但是用LinkedList创建的Stack性能更好，因此更推荐使用我们自己创建的栈结构。
Set
Set只能保存不想同的元素，如果存储相同的元素多次，将会报错。Set最长用的场景是测试元素之间的关系，例如你可以询问某个对象是否在Set中。因为这样，Set中最常用也是最重要的功能是查找，同长我们选择HashSet来进行这个工作。Set继承了Collection接口，而且除了这个接口没有实现任何其他的功能。
HashSet采用哈系算法，没有对元素进行排序，TreeSet采用了红黑树数据结构，因此HashSet得到的结果是无序的，TreeSet得到的结果是有序的。
Set中最长用的功能就是contains来测试集合之间的关系，如下：
//: holding/SetOperations.java
import java.util.*;
import static net.mindview.util.Print.*;
public class SetOperations {
public static void main(String[] args) {
Set<String> set1 = new HashSet<String>();
Collections.addAll(set1,"A B C D E F G H I J K L".split(" "));
set1.add("M");
print("H: " + set1.contains("H"));
print("N: " + set1.contains("N"));
Set<String> set2 = new HashSet<String>();
Collections.addAll(set2, "H I J K L".split(" "));
print("set2 in set1: " + set1.containsAll(set2));
set1.remove("H");
print("set1: " + set1);
print("set2 in set1: " + set1.containsAll(set2));
set1.removeAll(set2);
print("set2 removed from set1: " + set1);
Collections.addAll(set1, "X Y Z".split(" "));
print("‘X Y Z’ added to set1: " + set1);
}
} /* Output:
H: true
N: false
set2 in set1: true
set1: [D, K, C, B, L, G, I, M, A, F, J, E]
set2 in set1: false
set2 removed from set1: [D, C, B, G, M, A, F, E]
‘X Y Z’ added to set1: [Z, D, C, B, G, M, A, F, Y, X, E]
*///:~
Set中不能存储重复对象的特性非常有用，例如列举出出现的所有单词，还可以使用TreeSet将结果按照顺序输出：
//: holding/UniqueWordsAlphabetic.java
// Producing an alphabetic listing.
import java.util.*;
import net.mindview.util.*;
public class UniqueWordsAlphabetic {
public static void main(String[] args) {
Set<String> words = new TreeSet<String>(String.CASE_INSENSITIVE_ORDER);
words.addAll(new TextFile("SetOperations.java", "\\W+"));
System.out.println(words);
}
} /* Output:
[A, add, addAll, added, args, B, C, class, Collections, contains,
containsAll, D, E, F, false, from, G, H, HashSet, holding, I, import,
in, J, java, K, L, M, main, mindview, N, net, new, Output, Print,
public, remove, removeAll, removed, Set, set1, set2, SetOperations,
split, static, String, to, true, util, void, X, Y, Z]
*///:~
Map
在编程中通过一种对象索引另外一个对象非常有用。例如下面的代码用来检测java生成的随机数是否平均分布：
//: holding/Statistics.java
// Simple demonstration of HashMap.
import java.util.*;
public class Statistics {
public static void main(String[] args) {
Random rand = new Random(47);
Map<Integer,Integer> m = new HashMap<Integer,Integer>();
for(int i = 0; i < 10000; i++) {
// Produce a number between 0 and 20:
int r = rand.nextInt(20);
Integer freq = m.get(r);
m.put(r, freq == null ? 1 : freq + 1);
}
System.out.println(m);
}
} /* Output
另外Map还可以嵌套使用，只需要将其中的值设置为Map结构就可以，如下：
//: holding/MapOfList.java
package holding;
import typeinfo.pets.*;
import java.util.*;
import static net.mindview.util.Print.*;
public class MapOfList {
public static Map<Person, List<? extends Pet>>
petPeople = new HashMap<Person, List<? extends Pet>>();
static {
petPeople.put(new Person("Dawn"),
Arrays.asList(new Cymric("Molly"),new Mutt("Spot")));
petPeople.put(new Person("Kate"),
Arrays.asList(new Cat("Shackleton"),
new Cat("Elsie May"), new Dog("Margrett")));
petPeople.put(new Person("Marilyn"),
Arrays.asList(
new Pug("Louie aka Louis Snorkelstein Dupree"),
new Cat("Stanford aka Stinky el Negro"),
new Cat("Pinkola")));
petPeople.put(new Person("Luke"),
Arrays.asList(new Rat("Fuzzy"), new Rat("Fizzy")));
petPeople.put(new Person("Isaac"),
Arrays.asList(new Rat("Freckly")));
}
public static void main(String[] args) {
print("People: " + petPeople.keySet());
print("Pets: " + petPeople.values());
for(Person person : petPeople.keySet()) {
print(person + " has:");
for(Pet pet : petPeople.get(person))
print(" " + pet);
}
}
} /* Output:
Queue
队列是一种先进先出的数据结构，出队的顺序和入队的顺序相同。队列在对对象从一处转移到另外一处的时候非常有用，尤其是在并发编程的时候。LinkedList有支持队列特性的方法，同时还继承了Queue接口，因此一个LinkedList可以当作Queue来使用，通过将LinkedList上塑造型到Queue，可以使用队列的一系列方法：
//: holding/QueueDemo.java
// Upcasting to a Queue from a LinkedList.
import java.util.*;
public class QueueDemo {
public static void printQ(Queue queue) {
while(queue.peek() != null)
System.out.print(queue.remove() + " ");
System.out.println();
}
public static void main(String[] args) {
Queue<Integer> queue = new LinkedList<Integer>();
Random rand = new Random(47);
for(int i = 0; i < 10; i++)
queue.offer(rand.nextInt(i + 10));
printQ(queue);
Queue<Character> qc = new LinkedList<Character>();
for(char c : "Brontosaurus".toCharArray())
qc.offer(c);
printQ(qc);
}
} /* Output:
8 1 1 1 5 14 3 1 0 1
B r o n t o s a u r u s
*///:~
上面介绍的是一般队列，按照等待的时间进行先进先出，但是有些事后我们需要根据对象被需要的等级来安排出队的顺序，这个时候我们需要使用优先级队列，PriorityQueue。当使用offer添加对象到PriorityQueue的时候，对象在队列中被排序，默认的排序方法是使用对象的自然顺序，但是可以通过提供自己的Comparator来改变对象的排列顺序，你首先得到的对象是最高优先等级的对象：
//: holding/PriorityQueueDemo.java
import java.util.*;
public class PriorityQueueDemo {
public static void main(String[] args) {
PriorityQueue<Integer> priorityQueue = new PriorityQueue<Integer>();
Random rand = new Random(47);
for(int i = 0; i < 10; i++)
priorityQueue.offer(rand.nextInt(i + 10));
QueueDemo.printQ(priorityQueue);
List<Integer> ints = Arrays.asList(25, 22, 20,18, 14, 9, 3, 1, 1, 2, 3, 9, 14, 18, 21, 23, 25);
priorityQueue = new PriorityQueue<Integer>(ints);
QueueDemo.printQ(priorityQueue);
priorityQueue = new PriorityQueue<Integer>(ints.size(), Collections.reverseOrder());
priorityQueue.addAll(ints);
QueueDemo.printQ(priorityQueue);
String fact = "EDUCATION SHOULD ESCHEW OBFUSCATION";
List<String> strings = Arrays.asList(fact.split(""));
PriorityQueue<String> stringPQ = new PriorityQueue<String>(strings);
QueueDemo.printQ(stringPQ);
stringPQ = new PriorityQueue<String>(strings.size(), Collections.reverseOrder());
stringPQ.addAll(strings);
QueueDemo.printQ(stringPQ);
Set<Character> charSet = new HashSet<Character>();
for(char c : fact.toCharArray())
charSet.add(c); // Autoboxing
PriorityQueue<Character> characterPQ = new PriorityQueue<Character>(charSet);
QueueDemo.printQ(characterPQ);
}
} /* Output:
0 1 1 1 1 1 3 5 8 14
1 1 2 3 3 9 9 14 14 18 18 20 21 22 23 25 25
25 25 23 22 21 20 18 18 14 14 9 9 3 3 2 1 1
A A B C C C D D E E E F H H I I L N N O O O O S S S T T U U U W
W U U U T T S S S O O O O N N L I I H H F E E E D D C C C B A A
A B C D E F H I L N O S T U W
*///:~
Integer，String和Character都可以与PriorityQueue配合，因为他们已经有内置的排序方法，如果需要使用自己创建的类与PriorityQueue使用，你必须包含附加的排序方法，或者Comparator。
集合vs历遍器
Collection提供了所有有序序列的容器接口，另外，java.utilAbstractCollection类提供了Collection的实现，你可以通过创建AbstractCollection的子类而不用书写重复的代码。使用Collection接口而不使用具体类型的好处就是代码重用，任何继承自Collection接口的类型都可以在这个方法中使用。另外，Java中Collection和iterator是绑定到一起的，继承了Collection就意味着提供了iterator方法：
//: holding/InterfaceVsIterator.java
import typeinfo.pets.*;
import java.util.*;
public class InterfaceVsIterator {
public static void display(Iterator<Pet> it) {
while(it.hasNext()) {
Pet p = it.next();
System.out.print(p.id() + ":" + p + " ");
}
System.out.println();
}
public static void display(Collection<Pet> pets) {
for(Pet p : pets)
System.out.print(p.id() + ":" + p + " ");
System.out.println();
}
public static void main(String[] args) {
List<Pet> petList = Pets.arrayList(8);
Set<Pet> petSet = new HashSet<Pet>(petList);
Map<String,Pet> petMap = new LinkedHashMap<String,Pet>();
String[] names = ("Ralph, Eric, Robin, Lacey, " + "Britney, Sam, Spot, Fluffy").split(", ");
for(int i = 0; i < names.length; i++)
petMap.put(names[i], petList.get(i));
display(petList);
display(petSet);
display(petList.iterator());
display(petSet.iterator());
System.out.println(petMap);
System.out.println(petMap.keySet());
display(petMap.values());
display(petMap.values().iterator());
}
} /* Output:
0:Rat 1:Manx 2:Cymric 3:Mutt 4:Pug 5:Cymric 6:Pug 7:Manx
4:Pug 6:Pug 3:Mutt 1:Manx 5:Cymric 7:Manx 2:Cymric 0:Rat
0:Rat 1:Manx 2:Cymric 3:Mutt 4:Pug 5:Cymric 6:Pug 7:Manx
4:Pug 6:Pug 3:Mutt 1:Manx 5:Cymric 7:Manx 2:Cymric 0:Rat
{Ralph=Rat, Eric=Manx, Robin=Cymric, Lacey=Mutt, Britney=Pug,
Sam=Cymric, Spot=Pug, Fluffy=Manx}
[Ralph, Eric, Robin, Lacey, Britney, Sam, Spot, Fluffy]
0:Rat 1:Manx 2:Cymric 3:Mutt 4:Pug 5:Cymric 6:Pug 7:Manx
0:Rat 1:Manx 2:Cymric 3:Mutt 4:Pug 5:Cymric 6:Pug 7:Manx
*///:~
Iterator的使用在涉及到不能上塑造型到Collection的时候就显得尤为重要。例如我们通过继承一个承载了Pet对象的类来创建一个Collection的实现，我们必须实现Collection的所有方法，即使我们在display方法中不许要使用他们。虽然这可以通过继承自AbstractCollection来实现，你还是必须实现iterator()方法和size()来弥补AbstractCollection没有实现的方法，而且这些方法有可能在AbstractCollection的其他方法中使用到：
//: holding/CollectionSequence.java
import typeinfo.pets.*;
import java.util.*;
public class CollectionSequence
extends AbstractCollection<Pet> {
private Pet[] pets = Pets.createArray(8);
public int size() { return pets.length; }
public Iterator<Pet> iterator() {
return new Iterator<Pet>() {
private int index = 0;
public boolean hasNext() {
return index < pets.length;
}
public Pet next() { return pets[index++]; }
public void remove() { // Not implemented
throw new UnsupportedOperationException();
}
};
}
public static void main(String[] args) {
CollectionSequence c = new CollectionSequence();
InterfaceVsIterator.display(c);
InterfaceVsIterator.display(c.iterator());
}
} /* Output:
0:Rat 1:Manx 2:Cymric 3:Mutt 4:Pug 5:Cymric 6:Pug 7:Manx
0:Rat 1:Manx 2:Cymric 3:Mutt 4:Pug 5:Cymric 6:Pug 7:Manx
*///:~
从上面的例子中我们可以看到，如果需要继承Collection，你必须实现iterator方法，但是如果你的类已经继承了其他的类，那么你就不能继承自AbstractCollection，这个时候，继承自Collection你就必须实现接口中的所有方法，因此通过继承和添加iterator就显得方便许多：
//: holding/NonCollectionSequence.java
import typeinfo.pets.*;
import java.util.*;
class PetSequence {
protected Pet[] pets = Pets.createArray(8);
}
public class NonCollectionSequence extends PetSequence {
public Iterator<Pet> iterator() {
return new Iterator<Pet>() {
private int index = 0;
public boolean hasNext() {
return index < pets.length;
}
public Pet next() { return pets[index++]; }
public void remove() { // Not implemented
throw new UnsupportedOperationException();
}
};
}
public static void main(String[] args) {
NonCollectionSequence nc = new NonCollectionSequence();
InterfaceVsIterator.display(nc.iterator());
}
} /* Output:
0:Rat 1:Manx 2:Cymric 3:Mutt 4:Pug 5:Cymric 6:Pug 7:Manx
*///:~
通过提供Iterator是最松耦合的将序列与历遍序列相联系的方法，而且要比继承Collection要少得许多的限制。
Foreach和历遍器
freach语法可以对所有Collection对象使用。之所以可以的原因是Java SE5中引入了新的接口Iterable，包含了iterator方法用来生成iterator，foreach语法就是通过Iterable接口来实现历遍的，任何继承了Iterable接口的对象都可以使用foreach语法：
//: holding/IterableClass.java
// Anything Iterable works with foreach.
import java.util.*;
public class IterableClass implements Iterable<String> {
protected String[] words = ("And that is how " + "we know the Earth to be banana-shaped.").split(" ");
public Iterator<String> iterator() {
return new Iterator<String>() {
private int index = 0;
public boolean hasNext() {
return index < words.length;
}
public String next() { return words[index++]; }
public void remove() { // Not implemented
throw new UnsupportedOperationException();
}
};
}
public static void main(String[] args) {
for(String s : new IterableClass())
System.out.print(s + " ");
}
} /* Output:
And that is how we know the Earth to be banana-shaped.
*///:~
foreach语句可以用于数组或者任何Iterable接口，但是这并不意味着一个数组可以自动地转换成为Iterable，有没有任何装箱操作发生：
//: holding/ArrayIsNotIterable.java
import java.util.*;
public class ArrayIsNotIterable {
static <T> void test(Iterable<T> ib) {
for(T t : ib)
System.out.print(t + " ");
}
public static void main(String[] args) {
test(Arrays.asList(1, 2, 3));
String[] strings = { "A", "B", "C" };
// An array works in foreach, but it’s not Iterable:
//! test(strings);
// You must explicitly convert it to an Iterable:
test(Arrays.asList(strings));
}
} /* Output:
1 2 3 A B C
*///:~
如果是已经有了一个继承自Iterable的类，但是因为值提供了一种历遍方法，现在我希望有不同的历遍方法，例如前向和反向历遍，以便我可以选择使用那种历遍方法，该如何解决呢？一种方案就是使用适配器模式，当需要一种接口来适合另外一个接口的时候可以使用这种模式。下面添加一种逆向的历遍方法，不能重写，只能添加一个方法来产生一个Iterable对象，然后使用这个对象来完成foreach语句，这样就可以完成多种历遍方式：
//: holding/AdapterMethodIdiom.java
// The "Adapter Method" idiom allows you to use foreach
// with additional kinds of Iterables.
import java.util.*;
class ReversibleArrayList<T> extends ArrayList<T> {
public ReversibleArrayList(Collection<T> c) { super(c); }
public Iterable<T> reversed() {
return new Iterable<T>() {
public Iterator<T> iterator() {
return new Iterator<T>() {
int current = size() - 1;
public boolean hasNext() { return current > -1; }
public T next() { return get(current--); }
public void remove() { // Not implemented
throw new UnsupportedOperationException();
}
};
}
};
}
}
public class AdapterMethodIdiom {
public static void main(String[] args) {
ReversibleArrayList<String> ral = new ReversibleArrayList<String>(Arrays.asList("To be or not to be".split(" ")));
// Grabs the ordinary iterator via iterator():
for(String s : ral)
System.out.print(s + " ");
System.out.println();
// Hand it the Iterable of your choice
for(String s : ral.reversed())
System.out.print(s + " ");
}
} /* Output:
To be or not to be
be to not or be To
*///:~
使用这种方法还可以添加另外两种历遍的方法：
//: holding/MultiIterableClass.java
// Adding several Adapter Methods.
import java.util.*;
public class MultiIterableClass extends IterableClass {
public Iterable<String> reversed() {
return new Iterable<String>() {
public Iterator<String> iterator() {
return new Iterator<String>() {
int current = words.length - 1;
public boolean hasNext() { return current > -1; }
public String next() { return words[current--]; }
public void remove() { // Not implemented
throw new UnsupportedOperationException();
}
};
}
};
}
public Iterable<String> randomized() {
return new Iterable<String>() {
public Iterator<String> iterator() {
List<String> shuffled = new ArrayList<String>(Arrays.asList(words));
Collections.shuffle(shuffled, new Random(47));
return shuffled.iterator();
}
};
}
public static void main(String[] args) {
MultiIterableClass mic = new MultiIterableClass();
for(String s : mic.reversed())
System.out.print(s + " ");
System.out.println();
for(String s : mic.randomized())
System.out.print(s + " ");
System.out.println();
for(String s : mic)
System.out.print(s + " ");
}
} /* Output:
banana-shaped. be to Earth the know we how is that And
is banana-shaped. Earth that how the be And we know to
And that is how we know the Earth to be banana-shaped.
*///:~
我们可以看到random方法没有创建自己的Iterator仅仅是返回了乱序的List。我们可以看到Collections.shuffled方法不影响原数组的顺序，仅仅是打乱shuffle中的引用而已。之所以这样是因为randomized方法中先将Array.asList封装到ArrayList中，如果直接对Array.asList洗牌，那么产生List将会改变原数组中的是顺序，如下：
//: holding/ModifyingArraysAsList.java
import java.util.*;
public class ModifyingArraysAsList {
public static void main(String[] args) {
Random rand = new Random(47);
Integer[] ia = { 1, 2, 3, 4, 5, 6, 7, 8, 9, 10 };
List<Integer> list1 = new ArrayList<Integer>(Arrays.asList(ia));
System.out.println("Before shuffling: " + list1);
Collections.shuffle(list1, rand);
System.out.println("After shuffling: " + list1);
System.out.println("array: " + Arrays.toString(ia));
List<Integer> list2 = Arrays.asList(ia);
System.out.println("Before shuffling: " + list2);
Collections.shuffle(list2, rand);
System.out.println("After shuffling: " + list2);
System.out.println("array: " + Arrays.toString(ia));
}
} /* Output:
Before shuffling: [1, 2, 3, 4, 5, 6, 7, 8, 9,
After shuffling: [4, 6, 3, 1, 8, 7, 2, 5, 10,
array: [1, 2, 3, 4, 5, 6, 7, 8, 9, 10]
Before shuffling: [1, 2, 3, 4, 5, 6, 7, 8, 9,
After shuffling: [9, 1, 6, 3, 7, 2, 5, 10, 4,
array: [9, 1, 6, 3, 7, 2, 5, 10, 4, 8]
*///:~