Map集合、HashMap集合、LinkedHashMap集合、Hashtable集合、Collections工具类和模拟斗地主洗牌和发牌

1、Map集合概述和特点
  * A:Map接口概述
     * 查看API可以知道:
            * 将键映射到值的对象
            * 一个映射不能包含重复的键
            * 每个键最多只能映射到一个值
  * B:Map接口和Collection接口的不同
        * Map是双列的,Collection是单列的
        * Map的键唯一,Collection的子体系Set是唯一的
        * Map集合的数据结构值针对键有效,跟值无关;Collection集合的数据结构是针对元素有效

2、Map集合的功能概述
  * A:Map集合的功能概述
      * a:添加功能
          * V put(K key,V value):添加元素。
              * 如果键是第一次存储,就直接存储元素,返回null
              * 如果键不是第一次存在,就用值把以前的值替换掉,返回以前的值
      * b:删除功能
          * void clear():移除所有的键值对元素
          * V remove(Object key):根据键删除键值对元素,并把值返回
      * c:判断功能
          * boolean containsKey(Object key):判断集合是否包含指定的键
          * boolean containsValue(Object value):判断集合是否包含指定的值
          * boolean isEmpty():判断集合是否为空
      * d:获取功能
          * Set<Map.Entry<K,V>> entrySet():
          * V get(Object key):根据键获取值
          * Set<K> keySet():获取集合中所有键的集合
          * Collection<V> values():获取集合中所有值的集合
      * e:长度功能
          * int size():返回集合中的键值对的个数

3、Map集合的遍历之键找值
  * A:键找值思路:
      * 获取所有键的集合
      * 遍历键的集合,获取到每一个键
      * 根据键找值
  * B:案例演示
      * Map集合的遍历之键找值

HashMap<String, Integer> hm = new HashMap<>();
            hm.put("张三", 23);
            hm.put("李四", 24);
            hm.put("王五", 25);
            hm.put("赵六", 26);
            
            /*Set<String> keySet = hm.keySet();            //获取集合中所有的键
            Iterator<String> it = keySet.iterator();    //获取迭代器
            while(it.hasNext()) {                        //判断单列集合中是否有元素
                String key = it.next();                    //获取集合中的每一个元素,其实就是双列集合中的键
                Integer value = hm.get(key);            //根据键获取值
                System.out.println(key + "=" + value);    //打印键值对
            }*/
            
            for(String key : hm.keySet()) {                //增强for循环迭代双列集合第一种方式
                System.out.println(key + "=" + hm.get(key));
            }

4、Map集合的遍历之键值对对象找键和值
  * A:键值对对象找键和值思路:
      * 获取所有键值对对象的集合
      * 遍历键值对对象的集合,获取到每一个键值对对象
      * 根据键值对对象找键和值
  * B:案例演示
      * Map集合的遍历之键值对对象找键和值
    
            HashMap<String, Integer> hm = new HashMap<>();
            hm.put("张三", 23);
            hm.put("李四", 24);
            hm.put("王五", 25);
            hm.put("赵六", 26);
            /*Set<Map.Entry<String, Integer>> entrySet = hm.entrySet();    //获取所有的键值对象的集合
            Iterator<Entry<String, Integer>> it = entrySet.iterator();//获取迭代器
            while(it.hasNext()) {
                Entry<String, Integer> en = it.next();                //获取键值对对象
                String key = en.getKey();                                //根据键值对对象获取键
                Integer value = en.getValue();                            //根据键值对对象获取值
                System.out.println(key + "=" + value);
            }*/
            
            for(Entry<String,Integer> en : hm.entrySet()) {
                System.out.println(en.getKey() + "=" + en.getValue());
            }

5、LinkedHashMap的概述和使用
  * A:案例演示
      * LinkedHashMap的特点
          * 底层是链表实现的可以保证怎么存就怎么取

6、统计字符串中每个字符出现的次数
  * A:案例演示
      * 需求:统计字符串中每个字符出现的次数
            String str = "aaaabbbcccccccccc";
            char[] arr = str.toCharArray();                        //将字符串转换成字符数组
            HashMap<Character, Integer> hm = new HashMap<>();    //创建双列集合存储键和值
            
            for(char c : arr) {                                    //遍历字符数组
                /*if(!hm.containsKey(c)) {                        //如果不包含这个键
                    hm.put(c, 1);                                //就将键和值为1添加
                }else {                                            //如果包含这个键
                    hm.put(c, hm.get(c) + 1);                    //就将键和值再加1添加进来
                }*/
                
                //hm.put(c, !hm.containsKey(c) ? 1 : hm.get(c) + 1);
                Integer i = !hm.containsKey(c) ? hm.put(c, 1) : hm.put(c, hm.get(c) + 1);
            }
            
            for (Character key : hm.keySet()) {                    //遍历双列集合
                System.out.println(key + "=" + hm.get(key));
            }

7、集合嵌套之HashMap嵌套HashMap
  * A:案例演示
      * 集合嵌套之HashMap嵌套HashMap

 public class Demo {

     public static void main(String[] args) {
HashMap<Student,String> hm88=new HashMap<>();
hm88.put(new Student("张三",23),"北京");
hm88.put(new Student("李四",24),"北京");
hm88.put(new Student("王五",25),"上海");
hm88.put(new Student("赵六",26),"广州"); HashMap<Student,String> hm99=new HashMap<>();
hm99.put(new Student("唐僧",1023),"北京");
hm99.put(new Student("孙悟空",1024),"北京");
hm99.put(new Student("猪八戒",1025),"上海");
hm99.put(new Student("沙和尚",1026),"广州"); HashMap<HashMap<Student,String>,String> hm=new HashMap<>();
hm.put(hm88,"第88期基础班");
hm.put(hm99,"第99期基础班"); //遍历双列集合
for(HashMap<Student,String> h:hm.keySet()){
String value=hm.get(h);
for(Student key:h.keySet()){
String value2=h.get(key); System.out.println(key+"="+value2+"="+value);
}
} } }

8、HashMap和Hashtable的区别
  * A:面试题
      * HashMap和Hashtable的区别
          * Hashtable是JDK1.0版本出现的,是线程安全的,效率低,HashMap是JDK1.2版本出现的,是线程不安全的,效率高
          * Hashtable不可以存储null键和null值,HashMap可以存储null键和null值

9、Collections工具类的概述和常见方法讲解
  * A:Collections类概述
      * 针对集合操作 的工具类
  * B:Collections成员方法
          public static <T> void sort(List<T> list)

          public static <T> int binarySearch(List<?> list,T key)


          public static <T> T max(Collection<?> coll)


          public static void reverse(List<?> list)


          public static void shuffle(List<?> list)

10、模拟斗地主洗牌和发牌
  * A:案例演示
      * 模拟斗地主洗牌和发牌,牌没有排序

//买一副扑克
            String[] num = {"A","2","3","4","5","6","7","8","9","10","J","Q","K"};
            String[] color = {"方片","梅花","红桃","黑桃"};
            ArrayList<String> poker = new ArrayList<>();
            
            for(String s1 : color) {
                for(String s2 : num) {
                    poker.add(s1.concat(s2));
                }
            }
            
            poker.add("小王");
            poker.add("大王");
            //洗牌
            Collections.shuffle(poker);
            //发牌
            ArrayList<String> gaojin = new ArrayList<>();
            ArrayList<String> longwu = new ArrayList<>();
            ArrayList<String> me = new ArrayList<>();
            ArrayList<String> dipai = new ArrayList<>();
            
            for(int i = 0; i < poker.size(); i++) {
                if(i >= poker.size() - 3) {
                    dipai.add(poker.get(i));
                }else if(i % 3 == 0) {
                    gaojin.add(poker.get(i));
                }else if(i % 3 == 1) {
                    longwu.add(poker.get(i));
                }else {
                    me.add(poker.get(i));
                }
            }
            
            //看牌
            
            System.out.println(gaojin);
            System.out.println(longwu);
            System.out.println(me);
            System.out.println(dipai);

11、模拟斗地主洗牌和发牌并对牌进行排序的代码实现
  * A:案例演示
      * 模拟斗地主洗牌和发牌并对牌进行排序的代码实现

//买一副牌
            String[] num = {"3","4","5","6","7","8","9","10","J","Q","K","A","2"};
            String[] color = {"方片","梅花","红桃","黑桃"};
            HashMap<Integer, String> hm = new HashMap<>();            //存储索引和扑克牌
            ArrayList<Integer> list = new ArrayList<>();            //存储索引
            int index = 0;                                            //索引的开始值
            for(String s1 : num) {
                for(String s2 : color) {
                    hm.put(index, s2.concat(s1));                    //将索引和扑克牌添加到HashMap中
                    list.add(index);                                //将索引添加到ArrayList集合中
                    index++;
                }
            }
            hm.put(index, "小王");
            list.add(index);
            index++;
            hm.put(index, "大王");
            list.add(index);
            //洗牌
            Collections.shuffle(list);
            //发牌
            TreeSet<Integer> gaojin = new TreeSet<>();
            TreeSet<Integer> longwu = new TreeSet<>();
            TreeSet<Integer> me = new TreeSet<>();
            TreeSet<Integer> dipai = new TreeSet<>();
            
            for(int i = 0; i < list.size(); i++) {
                if(i >= list.size() - 3) {
                    dipai.add(list.get(i));                         //将list集合中的索引添加到TreeSet集合中会自动排序
                }else if(i % 3 == 0) {
                    gaojin.add(list.get(i));
                }else if(i % 3 == 1) {
                    longwu.add(list.get(i));
                }else {
                    me.add(list.get(i));
                }
            }
            
            //看牌
            lookPoker("高进", gaojin, hm);
            lookPoker("龙五", longwu, hm);
            lookPoker("冯佳", me, hm);
            lookPoker("底牌", dipai, hm);
            
        }
        
        public static void lookPoker(String name,TreeSet<Integer> ts,HashMap<Integer, String> hm) {
            System.out.print(name + "的牌是:");
            for (Integer index : ts) {
                System.out.print(hm.get(index) + " ");
            }
            
            System.out.println();
        }

12、泛型固定下边界
  * ? super E

上一篇:专业的web打印插件


下一篇:Java用代码演示String类中的以下方法的用法