Java8使用lambda表达式进行函数式编程可以对集合进行非常方便的操作。一个比较常见的操作是将list转换成map,一般使用Collectors的toMap()方法进行转换。一个比较常见的问题是当list中含有相同元素的时候,如果不指定取哪一个,则会抛出异常。因此,这个指定是必须的。
当然,使用toMap()的另一个重载方法,可以直接指定。这里,我们想讨论的是另一种方法:在进行转map的操作之前,能不能使用distinct()先把list的重复元素过滤掉,然后转map的时候就不用考虑重复元素的问题了。
package example.mystream;
import lombok.AllArgsConstructor;
import lombok.Getter;
import lombok.NoArgsConstructor;
import lombok.ToString;
import java.util.Arrays;
import java.util.List;
import java.util.Map;
import java.util.stream.Collectors;
public class ListToMap {
@AllArgsConstructor
@NoArgsConstructor
@ToString
private static class VideoInfo {
@Getter
String id;
int width;
int height;
}
public static void main(String [] args) {
List<VideoInfo> list = Arrays.asList(new VideoInfo("123", 1, 2),
new VideoInfo("456", 4, 5), new VideoInfo("123", 1, 2));
// preferred: handle duplicated data when toMap()
Map<String, VideoInfo> id2VideoInfo = list.stream().collect(
Collectors.toMap(VideoInfo::getId, x -> x,
(oldValue, newValue) -> newValue)
);
System.out.println("No Duplicated1: ");
id2VideoInfo.forEach((x, y) -> System.out.println("<" + x + ", " + y + ">"));
// handle duplicated data using distinct(), before toMap()
Map<String, VideoInfo> id2VideoInfo2 = list.stream().distinct().collect(
Collectors.toMap(VideoInfo::getId, x -> x)
);
System.out.println("No Duplicated2: ");
id2VideoInfo2.forEach((x, y) -> System.out.println("<" + x + ", " + y + ">"));
}
}
list里总共有三个元素,其中有两个我们认为是重复的。第一种转换是使用toMap()直接指定了对重复key的处理情况,因此可以正常转换成map。而第二种转换是想先对list进行去重,然后再转换成map,结果还是失败了,抛出了IllegalStateException,所以distinct()应该是失败了。
No Duplicated1:
<123, ListToMap.VideoInfo(id=123, width=1, height=2)>
<456, ListToMap.VideoInfo(id=456, width=4, height=5)>
Exception in thread "main" java.lang.IllegalStateException: Duplicate key ListToMap.VideoInfo(id=123, width=1, height=2)
at java.util.stream.Collectors.lambda$throwingMerger$0(Collectors.java:133)
at java.util.HashMap.merge(HashMap.java:1253)
at java.util.stream.Collectors.lambda$toMap$58(Collectors.java:1320)
at java.util.stream.ReduceOps$3ReducingSink.accept(ReduceOps.java:169)
at java.util.stream.DistinctOps$1$2.accept(DistinctOps.java:175)
at java.util.Spliterators$ArraySpliterator.forEachRemaining(Spliterators.java:948)
at java.util.stream.AbstractPipeline.copyInto(AbstractPipeline.java:481)
at java.util.stream.AbstractPipeline.wrapAndCopyInto(AbstractPipeline.java:471)
at java.util.stream.ReduceOps$ReduceOp.evaluateSequential(ReduceOps.java:708)
at java.util.stream.AbstractPipeline.evaluate(AbstractPipeline.java:234)
at java.util.stream.ReferencePipeline.collect(ReferencePipeline.java:499)
at example.mystream.ListToMap.main(ListToMap.java:79)
查看distinct()的API,可以看到如下介绍:
Returns a stream consisting of the distinct elements (according to {@link Object#equals(Object)}) of this stream.
显然,distinct()对对象进行去重时,是根据对象的equals()方法去处理的。如果我们的VideoInfo类不overrride超类Object的equals()方法,就会使用Object的。
但是Object的equals()方法只有在两个对象完全相同时才返回true。而我们想要的效果是只要VideoInfo的id/width/height均相同,就认为两个videoInfo对象是同一个。所以我们比如重写属于videoInfo的equals()方法。相关文章:一次性搞清楚equals和hashCode
我们设计VideoInfo的equals()如下:
@Override
public boolean equals(Object obj) {
if (!(obj instanceof VideoInfo)) {
return false;
}
VideoInfo vi = (VideoInfo) obj;
return this.id.equals(vi.id)
&& this.width == vi.width
&& this.height == vi.height;
}
这样一来,只要两个videoInfo对象的三个属性都相同,这两个对象就相同了。欢天喜地去运行程序,依旧失败!why?
《Effective Java》是本好书,连Java之父James Gosling都说,这是一本连他都需要的Java教程。在这本书中,作者指出,如果重写了一个类的equals()方法,那么就必须一起重写它的hashCode()方法!必须!没有商量的余地!
必须使得重写后的equals()满足如下条件:
因为这是Java的规定,违背这些规定将导致Java程序运行不再正常。
具体更多的细节,建议大家读读原书,必定获益匪浅。强烈推荐!
最终,我按照神书的指导设计VideoInfo的hashCode()方法如下:
@Override
public int hashCode() {
int n = 31;
n = n * 31 + this.id.hashCode();
n = n * 31 + this.height;
n = n * 31 + this.width;
return n;
}
终于,distinct()成功过滤了list中的重复元素,此时使用两种toMap()将list转换成map都是没问题的:
No Duplicated1:
<123, ListToMap.VideoInfo(id=123, width=1, height=2)>
<456, ListToMap.VideoInfo(id=456, width=4, height=5)>
No Duplicated2:
<123, ListToMap.VideoInfo(id=123, width=1, height=2)>
<456, ListToMap.VideoInfo(id=456, width=4, height=5)>
既然说distinct()是调用equals()进行比较的,那按照我的理解,list的3个元素至少需要比较3次吧。那是不是就调用了3次equals()呢?
在equals()中加入一句打印,这样就可以知道了。加后的equals()如下:
@Override
public boolean equals(Object obj) {
if (! (obj instanceof VideoInfo)) {
return false;
}
VideoInfo vi = (VideoInfo) obj;
System.out.println("<===> Invoke equals() ==> " + this.toString() + " vs. " + vi.toString());
return this.id.equals(vi.id) && this.width == vi.width && this.height == vi.height;
}
结果:
No Duplicated1:
<123, ListToMap.VideoInfo(id=123, width=1, height=2)>
<456, ListToMap.VideoInfo(id=456, width=4, height=5)>
<===> Invoke equals() ==> ListToMap.VideoInfo(id=123, width=1, height=2) vs. ListToMap.VideoInfo(id=123, width=1, height=2)
No Duplicated2:
<123, ListToMap.VideoInfo(id=123, width=1, height=2)>
<456, ListToMap.VideoInfo(id=456, width=4, height=5)>
结果发现才调用了一次equals()。为什么不是3次呢?仔细想想,根据hashCode()进行比较,hashCode()相同的情况就一次,就是list的第一个元素和第三个元素(都是VideoInfo(id=123, width=1, height=2))会出现hashCode()相同的情况。
所以我们是不是可以这么猜想:只有当hashCode()返回的hashCode相同的时候,才会调用equals()进行更进一步的判断。如果连hashCode()返回的hashCode都不同,那么可以认为这两个对象一定就是不同的了!
验证猜想:
更改hashCode()如下:
@Override
public int hashCode() {
return 1;
}
这样一来,所有的对象的hashCode()返回值都是相同的。当然,这样搞是符合Java规范的,因为Java只规定equals()相同的对象的hashCode必须相同,但是不同的对象的hashCode未必会不同。
结果:
No Duplicated1:
<123, ListToMap.VideoInfo(id=123, width=1, height=2)>
<456, ListToMap.VideoInfo(id=456, width=4, height=5)>
<===> Invoke equals() ==> ListToMap.VideoInfo(id=456, width=4, height=5) vs. ListToMap.VideoInfo(id=123, width=1, height=2)
<===> Invoke equals() ==> ListToMap.VideoInfo(id=456, width=4, height=5) vs. ListToMap.VideoInfo(id=123, width=1, height=2)
<===> Invoke equals() ==> ListToMap.VideoInfo(id=123, width=1, height=2) vs. ListToMap.VideoInfo(id=123, width=1, height=2)
No Duplicated2:
<123, ListToMap.VideoInfo(id=123, width=1, height=2)>
<456, ListToMap.VideoInfo(id=456, width=4, height=5)>
果然,equals()调用了三次!看来的确只有hashCode相同的时候才会调用equal()进一步判断两个对象究竟是否相同;如果hashCode不相同,两个对象显然不相同。猜想是正确的。
最终参考程序:
package example.mystream;
import lombok.AllArgsConstructor;
import lombok.Getter;
import lombok.NoArgsConstructor;
import lombok.ToString;
import java.util.Arrays;
import java.util.List;
import java.util.Map;
import java.util.stream.Collectors;
public class ListToMap {
@AllArgsConstructor
@NoArgsConstructor
@ToString
private static class VideoInfo {
@Getter
String id;
int width;
int height;
public static void main(String [] args) {
System.out.println(new VideoInfo("123", 1, 2).equals(new VideoInfo("123", 1, 2)));
}
@Override
public boolean equals(Object obj) {
if (!(obj instanceof VideoInfo)) {
return false;
}
VideoInfo vi = (VideoInfo) obj;
return this.id.equals(vi.id)
&& this.width == vi.width
&& this.height == vi.height;
}
/**
* If equals() is override, hashCode() must be override, too.
* 1. if a equals b, they must have the same hashCode;
* 2. if a doesn't equals b, they may have the same hashCode;
* 3. hashCode written in this way can be affected by sequence of the fields;
* 3. 2^5 - 1 = 31. So 31 will be faster when do the multiplication,
* because it can be replaced by bit-shifting: 31 * i = (i << 5) - i.
* @return
*/
@Override
public int hashCode() {
int n = 31;
n = n * 31 + this.id.hashCode();
n = n * 31 + this.height;
n = n * 31 + this.width;
return n;
}
}
public static void main(String [] args) {
List<VideoInfo> list = Arrays.asList(new VideoInfo("123", 1, 2),
new VideoInfo("456", 4, 5), new VideoInfo("123", 1, 2));
// preferred: handle duplicated data when toMap()
Map<String, VideoInfo> id2VideoInfo = list.stream().collect(
Collectors.toMap(VideoInfo::getId, x -> x,
(oldValue, newValue) -> newValue)
);
System.out.println("No Duplicated1: ");
id2VideoInfo.forEach((x, y) -> System.out.println("<" + x + ", " + y + ">"));
// handle duplicated data using distinct(), before toMap()
// Note that distinct() relies on equals() in the object
// if you override equals(), hashCode() must be override together
Map<String, VideoInfo> id2VideoInfo2 = list.stream().distinct().collect(
Collectors.toMap(VideoInfo::getId, x -> x)
);
System.out.println("No Duplicated2: ");
id2VideoInfo2.forEach((x, y) -> System.out.println("<" + x + ", " + y + ">"));
}
}
以上,VideoInfo使我们自己写的类,我们可以往里添加equals()和hashCode()方法。如果VideoInfo是我们引用的依赖中的一个类,我们无权对其进行修改,那么是不是就没办法使用distinct()按照某些元素是否相同,对对象进行自定义的过滤了呢?
在stackoverflow的一个回答上,我们可以找到一个可行的方法:使用wrapper。
假设在一个依赖中(我们无权修改该类),VideoInfo定义如下:
@AllArgsConstructor
@NoArgsConstructor
@ToString
public class VideoInfo {
@Getter
String id;
int width;
int height;
}
使用刚刚的wrapper思路,写程序如下(当然,为了程序的可运行性,还是把VideoInfo放进来了,假设它就是不能修改的,不能为其添加任何方法):
package example.mystream;
import lombok.AllArgsConstructor;
import lombok.Getter;
import lombok.NoArgsConstructor;
import lombok.ToString;
import java.util.Arrays;
import java.util.List;
import java.util.Map;
import java.util.stream.Collectors;
public class DistinctByWrapper {
private static class VideoInfoWrapper {
private final VideoInfo videoInfo;
public VideoInfoWrapper(VideoInfo videoInfo) {
this.videoInfo = videoInfo;
}
public VideoInfo unwrap() {
return videoInfo;
}
@Override
public boolean equals(Object obj) {
if (!(obj instanceof VideoInfo)) {
return false;
}
VideoInfo vi = (VideoInfo) obj;
return videoInfo.id.equals(vi.id)
&& videoInfo.width == vi.width
&& videoInfo.height == vi.height;
}
@Override
public int hashCode() {
int n = 31;
n = n * 31 + videoInfo.id.hashCode();
n = n * 31 + videoInfo.height;
n = n * 31 + videoInfo.width;
return n;
}
}
public static void main(String [] args) {
List<VideoInfo> list = Arrays.asList(new VideoInfo("123", 1, 2),
new VideoInfo("456", 4, 5), new VideoInfo("123", 1, 2));
// VideoInfo --map()--> VideoInfoWrapper ----> distinct(): VideoInfoWrapper --map()--> VideoInfo
Map<String, VideoInfo> id2VideoInfo = list.stream()
.map(VideoInfoWrapper::new).distinct().map(VideoInfoWrapper::unwrap)
.collect(
Collectors.toMap(VideoInfo::getId, x -> x,
(oldValue, newValue) -> newValue)
);
id2VideoInfo.forEach((x, y) -> System.out.println("<" + x + ", " + y + ">"));
}
}
/**
* Assume that VideoInfo is a class that we can't modify
*/
@AllArgsConstructor
@NoArgsConstructor
@ToString
class VideoInfo {
@Getter
String id;
int width;
int height;
}
整个wrapper的思路无非就是构造另一个类VideoInfoWrapper,把hashCode()和equals()添加到wrapper中,这样便可以按照自定义规则对wrapper对象进行自定义的过滤。
我们没法自定义过滤VideoInfo,但是我们可以自定义过滤VideoInfoWrapper啊!
之后要做的,就是将VideoInfo全部转化为VideoInfoWrapper,然后过滤掉某些VideoInfoWrapper,再将剩下的VideoInfoWrapper转回VideoInfo,以此达到过滤VideoInfo的目的。很巧妙!
另一种更精妙的实现方式是自定义一个函数:
private static <T> Predicate<T> distinctByKey(Function<? super T, Object> keyExtractor) {
Map<Object, Boolean> map = new ConcurrentHashMap<>();
return t -> map.putIfAbsent(keyExtractor.apply(t), Boolean.TRUE) == null;
}
(输入元素的类型是T及其父类,keyExtracctor是映射函数,返回Object,整个传入的函数的功能应该是提取key的。distinctByKey函数返回的是Predicate函数,类型为T。)
这个函数传入一个函数(lambda),对传入的对象提取key,然后尝试将key放入concurrentHashMap,如果能放进去,说明此key之前没出现过,函数返回false;如果不能放进去,说明这个key和之前的某个key重复了,函数返回true。
这个函数最终作为filter()函数的入参。根据Java API可知filter(func)过滤的规则为:如果func为true,则过滤,否则不过滤。因此,通过“filter() + 自定义的函数”,凡是重复的key都返回true,并被filter()过滤掉,最终留下的都是不重复的。
最终实现的程序如下
package example.mystream;
import lombok.AllArgsConstructor;
import lombok.Getter;
import lombok.NoArgsConstructor;
import lombok.ToString;
import java.util.Arrays;
import java.util.List;
import java.util.Map;
import java.util.concurrent.ConcurrentHashMap;
import java.util.function.Function;
import java.util.function.Predicate;
import java.util.stream.Collectors;
public class DistinctByFilterAndLambda {
public static void main(String[] args) {
List<VideoInfo> list = Arrays.asList(new VideoInfo("123", 1, 2),
new VideoInfo("456", 4, 5), new VideoInfo("123", 1, 2));
// Get distinct only
Map<String, VideoInfo> id2VideoInfo = list.stream().filter(distinctByKey(vi -> vi.getId())).collect(
Collectors.toMap(VideoInfo::getId, x -> x,
(oldValue, newValue) -> newValue)
);
id2VideoInfo.forEach((x, y) -> System.out.println("<" + x + ", " + y + ">"));
}
/**
* If a key could not be put into ConcurrentHashMap, that means the key is duplicated
* @param keyExtractor a mapping function to produce keys
* @param <T> the type of the input elements
* @return true if key is duplicated; else return false
*/
private static <T> Predicate<T> distinctByKey(Function<? super T, Object> keyExtractor) {
Map<Object, Boolean> map = new ConcurrentHashMap<>();
return t -> map.putIfAbsent(keyExtractor.apply(t), Boolean.TRUE) == null;
}
}
/**
* Assume that VideoInfo is a class that we can't modify
*/
@AllArgsConstructor
@NoArgsConstructor
@ToString
class VideoInfo {
@Getter
String id;
int width;
int height;
}
网站声明:如果转载,请联系本站管理员。否则一切后果自行承担。
添加我为好友,拉您入交流群!
请使用微信扫一扫!