最近一直在看Lamada表达式,在开发中使用后现在真的越看越顺眼,这种函数式编程思维只要初窥门径,简直就停不下来,然后花了点时间做了一个使用总结范例,有不全面或是不对的请给位大佬指出,咱们一起探讨.
import com.google.common.collect.Lists;
import com.google.common.collect.Maps;
import org.junit.Before;
import org.junit.Test;
import java.util.*;
import java.util.stream.Collectors;
import java.util.stream.Stream;
import static com.google.common.primitives.Ints.asList;
import static java.util.stream.Collectors.*;
import static org.junit.Assert.assertEquals;
/**
* @author yihur
* @description jdk 1.8 的lamada表达式的使用范例
* @date 2019/4/23
*/
public class TestJavaStream {
private List> lists;
private List> listCopy;
private Map map;
/**
* 初始化数据
*/
@Before
public void before_map_new_hashMap() {
//用guava的方法创建HashMap,这样初始化的时候需要制定HashMap的大小,一般大小指定为:(长度/0.75)+1
//这样指定大小,创建的好处了以避免HashMap的频繁的扩容导致性能的损耗
map = Maps.newHashMapWithExpectedSize(6);
map.put("go", "wqeqwe");
map.put("python", "eqwe");
map.put("java", "12sdf3");
map.put("c++", "12fsdf3");
map.put("JavaScript", "78797987");
map.put("ruby", "4564");
map.put("php", "4654654");
map.put("c", "465465");
}
/**
* 初始化数据
*/
@Before
public void before_java_junit() {
lists = Lists.newArrayListWithCapacity(8);
Map map = Maps.newHashMapWithExpectedSize(6);
map.put("name", "theyihur");
map.put("age", "18");
map.put("nation", "china");
map.put("type", "1");
lists.add(map);
map = Maps.newHashMapWithExpectedSize(6);
map.put("name", "thezys");
map.put("age", "21");
map.put("nation", "amirica");
map.put("type", "2");
lists.add(map);
map = Maps.newHashMapWithExpectedSize(6);
map.put("name", "thedgd");
map.put("age", "22");
map.put("nation", "jipan");
map.put("type", "3");
lists.add(map);
map = Maps.newHashMapWithExpectedSize(6);
map.put("name", "tom");
map.put("age", "23");
map.put("nation", "louye");
map.put("type", "4");
lists.add(map);
map = Maps.newHashMapWithExpectedSize(6);
map.put("name", "jack");
map.put("age", "22");
map.put("nation", "indian");
map.put("type", "1");
lists.add(map);
map = Maps.newHashMapWithExpectedSize(6);
map.put("name", "fice");
map.put("age", "12");
map.put("nation", "malaixiya");
map.put("type", "3");
lists.add(map);
}
/**
* 初始化数据
*/
@Before
public void before_java_list_copy_junit() {
listCopy = Lists.newArrayListWithCapacity(8);
Map map = Maps.newHashMapWithExpectedSize(6);
map.put("name", "theyihur");
map.put("age", "18");
map.put("nation", "china");
map.put("type", "1");
listCopy.add(map);
map = Maps.newHashMapWithExpectedSize(6);
map.put("name", "thezys");
map.put("age", "123");
map.put("nation", "amirica");
map.put("type", "2");
listCopy.add(map);
map = Maps.newHashMapWithExpectedSize(6);
map.put("name", "thedgd");
map.put("age", "12");
map.put("nation", "jipan");
map.put("type", "3");
listCopy.add(map);
map = Maps.newHashMapWithExpectedSize(6);
map.put("name", "tom");
map.put("age", "1231");
map.put("nation", "louye");
map.put("type", "4");
listCopy.add(map);
map = Maps.newHashMapWithExpectedSize(6);
map.put("name", "jack");
map.put("age", "1231");
map.put("nation", "indian");
map.put("type", "1");
listCopy.add(map);
map = Maps.newHashMapWithExpectedSize(6);
map.put("name", "fice");
map.put("age", "31231");
map.put("nation", "malaixiya");
map.put("type", "3");
listCopy.add(map);
}
/**
* 基于Lamada表达式的list和map的遍历
*/
@Test
public void test_java_stream_foreach() {
//对map的遍历
map.forEach((k, v) -> System.out.println(k + "=" + v));
//list的遍历
lists.forEach(e -> System.out.println(e.toString()));
}
/**
* flatMap扁平化流
*/
@Test
public void test_java_stream_flatMap() {
//拼接去重
List together = Stream.of(asList(1, 2, 3, 4), asList(3, 4, 5, 6, 7))
.flatMap(Collection::stream)
.distinct()
.collect(toList());
System.out.println(Arrays.toString(together.toArray()));
System.out.println("=========================================");
//数组字符分割去重返回list
String[] strArray = {"hello", "world"};
List res = Arrays.stream(strArray)
.map(w -> w.split(""))
.flatMap(Arrays::stream)
.distinct()
.collect(Collectors.toList());
System.out.println(res);
System.out.println("=========================================");
//给定两数组，返回数组对
Integer[] nums2 = {1, 2, 3};
Integer[] nums3 = {3, 4};
List nums2List = Arrays.asList(nums2);
List nums3List = Arrays.asList(nums3);
//使用2个map嵌套过滤,返回一个新的list
List res2 = nums2List.stream().flatMap(i -> nums3List.stream().map(j -> new int[]{i, j})).collect(Collectors.toList());
res2.forEach(e -> System.out.println(Arrays.toString(e)));
System.out.println("=========================================");
//返回总和能被3整除的数对
List res3 = nums2List.stream().flatMap(i -> nums3List.stream().filter(j -> (i + j) % 3 == 0).map(j -> new int[]{i, j})).collect(Collectors.toList());
res3.forEach(e -> System.out.println(Arrays.toString(e)));
}
/**
* 找到list中的最大和最小值
*/
@Test
public void test_java_list_max_min() {
List list = Lists.newArrayList(3, 5, 2, 9, 1);
int maxInt = list.stream()
.max(Integer::compareTo)
.get();
int minInt = list.stream()
.min(Integer::compareTo)
.get();
assertEquals(9, maxInt);
assertEquals(1, minInt);
}
/**
* reduce 对数据计算后返回
*/
@Test
public void test_java_stream_reduce() {
int resultSum = Stream.of(45, 10, 20, 5)
.reduce(0, (accq, element) -> accq + element);
assertEquals(80, resultSum);
int result = Stream.of(1, 2, 3, 4)
.reduce(1, (acc, element) -> acc * element);
assertEquals(24, result);
}
/**
* map计算平方数
*/
@Test
public void test_java_stream_map_count() {
List nums = Arrays.asList(1, 2, 3, 4);
List squareNums = nums.stream().
map(n -> n * n).
collect(Collectors.toList());
System.out.println(Arrays.toString(squareNums.toArray()));
}
/**
* 对list进行排序操作
*/
@Test
public void test_java_list_sort_stream() {
List list = Lists.newArrayList(3, 5, 1, 10, 8);
List sortedList = list.stream()
.sorted(Integer::compareTo)
.collect(Collectors.toList());
assertEquals(sortedList, Lists.newArrayList(1, 3, 5, 8, 10));
}
/**
* 遍历list取相应的值返回一组特定格式的数据
* 返回值可以Collectors.进行定义
*/
@Test
public void test_java_stream_map() {
Set stringSet = lists.stream()
.filter(e -> e.get("name").toString().startsWith("the"))
.map(e -> e.get("nation").toString())
.collect(Collectors.toSet());
System.out.println(Arrays.toString(stringSet.toArray()));
List strings = lists.stream()
.filter(e -> e.get("nation").toString().length() > 6)
.map(e -> e.get("name").toString())
.collect(Collectors.toList());
System.out.println(Arrays.toString(strings.toArray()));
}
/**
* 统计list中某个字段的最大/最小/平均/总和
*/
@Test
public void test_java_stream_collection_group_by() {
IntSummaryStatistics summaryStatistics = lists.stream().mapToInt(e -> Integer.valueOf(e.get("age").toString())).summaryStatistics();
System.out.printf("Max: %d, Min: %d, Ave: %f, Sum: %d, count: %d",
summaryStatistics.getMax(),
summaryStatistics.getMin(),
summaryStatistics.getAverage(),
summaryStatistics.getSum(),
summaryStatistics.getCount());
}
/**
* 数据分组操作
*/
@Test
public void test_group_by_java_type() {
//数据分组
Map>> type = lists.stream().collect(groupingBy(e -> e.get("type")));
System.out.println(type.toString());
//数据分组后计数
Map collect = lists.stream().collect(groupingBy(e -> e.get("type"), counting()));
System.out.println(collect.toString());
//按照类型分组之后 返回带上每组数据的其中一个字段的额list
Map> listMap = lists.stream().collect(groupingBy(e -> e.get("type"), mapping(e_ -> e_.get("name"), toList())));
System.out.println(listMap.toString());
}
/**
* 将list>中的数据取出拼接成字符串
*/
@Test
public void test_group_by_java_list_to_string() {
String name = lists.stream().map(e -> e.get("name").toString()).collect(Collectors.joining(",", "[", "]"));
System.out.println("name:" + name);
}
/**
* list去重操作
*/
@Test
public void test_java_stream_distinct() {
List distinctList = Lists.newArrayList("a", "a", "b", "b", "c", "c");
List afterDistinctList = distinctList.stream().distinct().collect(Collectors.toList());
System.out.println(Arrays.toString(afterDistinctList.toArray()));
}
/**
* parallel()并行化操作,进行操作前会对数据进行分块,然后每块数据开辟线程运算,
* 尽量使用list或数组作为数据源,因为parallel会对数据进行分割,
*
* 注意:只有在数据量大的时候才能体现出并行化的效率优势
*
* @param
* @return void
* @author yihur
* @date 2019/4/29
*/
@Test
public void test_java_stream_parallel() {
int sumSize = Stream.of("Apple", "Banana", "Orange", "Pear")
.parallel()
.map(String::length)
.reduce(Integer::sum)
.get();
assertEquals(21, sumSize);
}
/**
* 并行化的方法生成数组
* 并行化的方法数组求和
*/
@Test
public void test_java_array_parallel() {
double[] values = new double[20];
Arrays.parallelSetAll(values, i -> i);
System.out.println(Arrays.toString(values));
Arrays.parallelPrefix(values, Double::sum);
System.out.println(Arrays.toString(values));
}
/**
* 对于嵌套for循环的lamada表达式的优化
*/
@Test
public void test_list_for_list() {
List strs = Arrays.asList("a", "a", "a", "a", "b");
boolean aa = strs.stream().anyMatch(str -> str.equals("a"));
boolean bb = strs.stream().allMatch(str -> str.equals("a"));
boolean cc = strs.stream().noneMatch(str -> str.equals("a"));
long count = strs.stream().filter(str -> str.equals("a")).count();
System.out.println(aa);// TRUE
System.out.println(bb);// FALSE
System.out.println(cc);// FALSE
System.out.println(count);
//noneMatch 不满足判断条件的数据返回
lists.stream()
.filter(map -> listCopy.stream().noneMatch(map1 -> map.get("age").equals(map1.get("age")) && map.get("name").equals(map1.get("name"))))
.forEach(map -> {
System.out.println(map.toString());
});
//allMatch 所有数据满足全部条件的时候返回数据
System.out.println("==========================================");
lists.stream()
.filter(map -> listCopy.stream().allMatch(map1 -> map.get("age").equals(map1.get("age"))))
.forEach(map -> {
System.out.println(map.toString());
});
//anyMatch 只有满足判断条件的数据返回
System.out.println("==========================================");
lists.stream()
.filter(map -> listCopy.stream().anyMatch(map1 -> map.get("age").equals(map1.get("age"))))
.forEach(map -> {
System.out.println(map.toString());
});
}
}