Java流操作總結(jié)

Java流(Stream)操作自Java 8引入蚀浆，通過Stream操作可以簡化代碼編寫，提高代碼執(zhí)行效率搜吧。流整體操作分為創(chuàng)建(Supplier)市俊、轉(zhuǎn)換、約簡和收集幾個操作滤奈。

創(chuàng)建流

創(chuàng)建流最常見的方式是調(diào)用Collection.stream()方法摆昧，將集合數(shù)據(jù)轉(zhuǎn)換為流，也可以通過generate方法創(chuàng)建流蜒程。如下所示：

    public static void main(String[] args) {
        //從Collection中產(chǎn)生流
        ArrayList<String> strings = new ArrayList<>();
        for (int i = 0; i < 20; ++i) {
            strings.add("string" + i);
        }

        Stream<String> stringStream = strings.stream();
        show("stringStream", stringStream);

        //使用generate產(chǎn)生流
        Stream<Integer> integerStream = Stream.generate(() -> (int) (Math.random() * 100));
        show("integerStream", integerStream);
        //使用iterate產(chǎn)生序列
        Stream<Integer> integerStream1 = Stream.iterate(0, n -> n + 1);
        show("integerStream1", integerStream1);

        //使用of方法創(chuàng)建流
        Stream<String> stringStream1 = Stream.of("Bob", "Mark", "Alice", "Edd");
        show("stringStream1", stringStream1);

        //正則表達(dá)式產(chǎn)生流
        String sentences = "I had a dog whose name was Bingo. B, I, N, G, O, Bingo was his name, Oh!";
        Stream<String> words = Pattern.compile("\\PL+").splitAsStream(sentences);
        show("PatterStream", words);
    }

    public static<T> void show(String title, Stream<T> stream) {
        final int size = 10;
        List<T> array = stream.limit(size).collect(Collectors.toList());
        System.out.println(title + ": ");
        for (int i = 0; i < array.size(); ++i) {
            if (i > 0) {
                System.out.print(", ");
            }
            if (i < array.size()) {
                System.out.print(array.get(i));
            } else {
                System.out.print("...");
            }
        }
        System.out.println();
    }

流的轉(zhuǎn)換

流的轉(zhuǎn)換即將一個給定的流通過流的方法轉(zhuǎn)換為另一個流绅你，常見的方法有filter，map昭躺，distinct, sorted忌锯，reverse等方法。

filter方法保留滿足filter方法中的條件表達(dá)式的元素领炫；
map方法可以對流中的元素進(jìn)行一個轉(zhuǎn)換操作偶垮；
distinct方法將流中重復(fù)的元素去除
sorted按給定的比較方法進(jìn)行排序

String sentences = "I had a dog whose name was Bingo. B, I, N, G, O, Bingo was his name, Oh!";
//保留長度大于2的單詞
Stream<String> words1 = Pattern.compile("\\PL+").splitAsStream(sentences).filter(s -> s.length() > 2);
show("PatterStream1", words1);

//將單詞轉(zhuǎn)換為大寫
Stream<String> words2 = Pattern.compile("\\PL+").splitAsStream(sentences).map(String::toUpperCase);
show("PatterStream2", words2);

//去除重復(fù)的單詞
Stream<String> words3 = Pattern.compile("\\PL+").splitAsStream(sentences).distinct();
show("PatterStream3", words3);

//按長度對單詞進(jìn)行排序
Stream<String> words4 = Pattern.compile("\\PL+").splitAsStream(sentences).sorted(Comparator.comparing(String::length));
show("PatterStream4", words4);

 //按長度對單詞進(jìn)行排序后反序
 Stream<String> words5 = Pattern.compile("\\PL+").splitAsStream(sentences).sorted(Comparator.comparing(String::length).reversed());
 show("PatterStream4", words5);

流的約減（Reduce）

流經(jīng)過處理后，提供了一系列方法從流中抽取數(shù)據(jù)。抽取數(shù)據(jù)是一種終結(jié)操作针史，常用的方法如下：

count方法：獲取流中滿足條件的元素個數(shù)
min晶伦、max方法：按照給定的比較方法獲取最小最大值，注意該方法返回的是一個Optional<T>的包裝類啄枕，這是因?yàn)榱骺赡転榭铡?/li>
find查找方法：包括findFirst婚陪，findAny等。
match方法：根據(jù)條件查找是否有匹配條件的元素频祝，有anyMatch泌参，allMatch和noneMatch，返回的是boolean類型常空。
消費(fèi)流：通過方法消費(fèi)流中的元素沽一，例如使用forEach迭代，傳遞方法處理流中的每個元素漓糙。
轉(zhuǎn)換為Collection或數(shù)組：通過collect方法傳遞Collectors類提供的toCollection方法可以轉(zhuǎn)換為Collection或者調(diào)用toArray方法轉(zhuǎn)換為數(shù)組铣缠。
拼接：使用Collectors.join方法可以拼接流.
統(tǒng)計，使用IntSummaryStatistics昆禽、DoubleSummaryStatistics和LongSummaryStatistics方法構(gòu)造統(tǒng)計對象蝗蛙，獲取總數(shù)，平均值醉鳖，最大值捡硅，最小值。需要注意調(diào)用Collectors.summarizing{Int,Double, Long}方法需要傳遞一個lambda表達(dá)式盗棵，確定統(tǒng)計的內(nèi)容壮韭。

//獲取單詞的數(shù)量
long wordsCount = Pattern.compile("\\PL+").splitAsStream(sentences).filter(s -> s.length() > 2).count();
System.out.println("Number of words length greater than 2: " + wordsCount);

//獲取最長的單詞
Optional<String> longestWord = Pattern.compile("\\PL+").splitAsStream(sentences).max(Comparator.comparing(String::length));
System.out.println("longestWord: " + longestWord.orElse("null"));

//獲取最短的單詞
Optional<String> shortestWord = Pattern.compile("\\PL+").splitAsStream(sentences).min(Comparator.comparing(String::length));
System.out.println("shortestWord: " + shortestWord.orElse("null"));

//獲取第一個以h開頭的單詞
Optional<String> firstWordsStartWithH = Pattern.compile("\\PL+").splitAsStream(sentences)
                .filter(s -> s.toLowerCase().startsWith("h")).findFirst();
System.out.println("firstWordsStartWithH: " + firstWordsStartWithH.orElse("Not Found"));

//獲取任何一個以h開頭的單詞
Optional<String> anyWordsStartWithH = Pattern.compile("\\PL+").splitAsStream(sentences)
                .filter(s -> s.toLowerCase().startsWith("h")).findAny();
System.out.println("anyWordsStartWithH: " + anyWordsStartWithH.orElse("Not Found"));

//查找其中是否有was這個單詞
boolean hasWas = Pattern.compile("\\PL+").splitAsStream(sentences).anyMatch("was"::equals);
System.out.println("hasWas: " + hasWas);

//查找其中全部是was這個單詞
boolean allWas = Pattern.compile("\\PL+").splitAsStream(sentences).allMatch("was"::equals);
System.out.println("allWas: " + allWas);

//使用forEach方法打印全部單詞的大寫
Pattern.compile("\\PL+").splitAsStream(sentences).map(String::toUpperCase).forEach(System.out::print);

//將流轉(zhuǎn)為Collection
List<String> wordsList = Pattern.compile("\\PL+").splitAsStream(sentences).collect(Collectors.toList());
System.out.println("wordsList: " + wordsList);

//將流轉(zhuǎn)為數(shù)組
String[] wordsArray = Pattern.compile("\\PL+").splitAsStream(sentences).toArray(String[]::new);
System.out.println("wordsArray: " + Arrays.toString(wordsArray));

//將流拼接為字符串
String wordsToSentences = Pattern.compile("\\PL+").splitAsStream(sentences).collect(Collectors.joining(", "));
System.out.println("wordsToSentences: " + wordsToSentences);

//數(shù)據(jù)統(tǒng)計：統(tǒng)計單詞的長度
IntSummaryStatistics lengthSummary = Pattern.compile("\\PL+").splitAsStream(sentences).
                collect(Collectors.summarizingInt(String::length));
System.out.println("Longest word length: " + lengthSummary.getMax());
System.out.println("Shortest word length: " + lengthSummary.getMin());
System.out.println("Average word length: " + lengthSummary.getAverage());

收集到Map：使用Collectors.toMap轉(zhuǎn)到Map數(shù)據(jù)中。
分組：使用groupingBy進(jìn)行分組

    public static class User {
        private String name;
        private int id;
        //1=male, 0 = female, 2= unknown
        private int sex = 2;

        public User(int id, String name, int sex) {
            this.id = id;
            this.name = name;
            if (sex == 0 || sex == 1) {
                this.sex = sex;
            }
        }

        public int getId() {
            return id;
        }

        public String getName() {
            return name;
        }

        public int getSex() {
            return sex;
        }

        @Override
        public String toString() {
            return "User{id = " + id + ", name = " + name + ", sex = " + sex + "}";
        }
    }

    //省略部分代碼...

    public static Stream<User> userStream() {
        return Stream.of(
                new User(1, "Box", 1),
                new User(2, "Succi", 0),
                new User(3, "Lily", 0),
                new User(4, "Sara", 0),
                new User(5, "Mark", 1)
        );
    }
    

    //main方法
    public static void main(String[] args) {
        //...
        //映射為id -> User格式
        Map<Integer, User> userMap = userStream().collect(Collectors.toMap(User::getId, Function.identity()));
        for (Integer key:userMap.keySet()) {
            System.out.println("id " + key + ": " + userMap.get(key));
        }
        
        //映射為id -> User.name格式
        Map<Integer, String> idToNameMap = userStream().collect(Collectors.toMap(User::getId, User::getName));
        System.out.println(idToNameMap);
        
        //按性別分組
        Map<Integer, Set<User>> sexToUserMap = userStream().collect(Collectors.groupingBy(User::getSex, Collectors.toSet()));
        for (Integer key:sexToUserMap.keySet()) {
            Set<User> usersOfSex = sexToUserMap.get(key);
            System.out.println("Sex " + key + ": " + usersOfSex);
        }
    }