说明
在实际编程中，经常会遇到数组或列表去掉重复项，以保持成员唯一性。各个语言的实现方式均不尽相同。针对数组去重，Java实现方式有多种，比如新建数组来存储非重复项，或者在原有基础上删除掉重复的项，也可以利用数据结构Set或ArrayList来达到去重复。以下18种方式都可以实现，但每一种方法都不尽相同，有的很简单，几行代码搞定，有的则稍复杂，需要10来行代码。通过对比不同的实现，我们可以从中发现Java语言里一些不同的东西。
方式
// 1. 双循环遍历全部成员，将当前项目与左边项逐个进行对比，如果值相同且下标相同表示唯一，其他则认为是重复项进行忽略。static int[] unique1(int arr[]) { int newArr[] = new int[arr.length]; int x = 0; for (int i = 0, l = arr.length; i < l; i++) { for (int j = 0; j <= i; j++) { if (arr[i] == arr[j]) { if (i == j) { newArr[x] = arr[i]; x++; } break; } } } int result[] = Arrays.copyOf(newArr, x); return result;}// 2. 先将数组转换为List，利用List的indexOf方法查找下标，当下标匹配时表示唯一，添加到新列表中static Integer[] unique2(Integer arr[]) { int x = 0; List list = new ArrayList<>(Arrays.asList(arr)); int l = list.size(); for (int i = 0; i < l; i++) { if (list.indexOf(arr[i]) == i) { list.add(arr[i]); x++; } } // 返回取出的非重复项 Integer[] result = new Integer[x]; return list.subList(list.size() - x, list.size()).toArray(result);}// 3. 在原有列表上移除重复项目。自后往前遍历，逐个与前面项比较，如果值相同且下标相同，则移除当前项。static Integer[] unique3(Integer arr[]) { List list = new ArrayList<>(Arrays.asList(arr)); int l = list.size(); while (l-- > 0) { int i = l; while (i-- > 0) { if (list.get(l).equals(list.get(i))) { list.remove(l); break; } } } return list.toArray(new Integer[list.size()]);}// 4. 在原有列表上移除重复项目。自前往后遍历，逐个与前面项比较，如果值相同且下标相同，则移除前面项。static Integer[] unique4(Integer arr[]) { List list = new ArrayList<>(Arrays.asList(arr)); int l = list.size(); for (int i = 1; i < l; i++) { for (int j = 0; j < i; j++) { if (list.get(i).equals(list.get(j))) { list.remove(i); i--; l--; break; } } } return list.toArray(new Integer[list.size()]);}// 5. 在原有列表上移除重复项目。自前往后遍历，逐个与后面项比较，如果值相同且下标相同，则移除当前项。static Integer[] unique5(Integer arr[]) { List list = new ArrayList<>(Arrays.asList(arr)); int l = list.size(); for (int i = 0; i < l; i++) { for (int j = i + 1; j < l; j++) { if (list.get(i).equals(list.get(j))) { list.remove(j); i--; l--; break; } } } return list.toArray(new Integer[list.size()]);}// 6. 利用hashMap属性唯一性来实现去重复。static Integer[] unique6(Integer arr[]) { Map map = new HashMap(); for (Integer item : arr) { if (map.containsKey(item)) { continue; } map.put(item, item); } List list = new ArrayList<>(map.values()); return list.toArray(new Integer[list.size()]);}// 7. 利用filter表达式，即把不符合条件的过滤掉。需要借助外部列表存储不重复项。static List unique7newArr = new ArrayList<>();static boolean unique7contains(Integer item) { if (unique7newArr.indexOf(item) < 0) { unique7newArr.add(item); return true; } return false;}static Integer[] unique7(Integer arr[]) { List list = new ArrayList<>(Arrays.asList(arr)); return list.stream().filter(UniqueArray::unique7contains).collect(Collectors.toList()) .toArray(new Integer[UniqueArray.unique7newArr.size()]);}// 8. 利用hashSet数据结构直接去重复项。无序非同步。static Integer[] unique8(Integer arr[]) { System.out.print("covert to steam first then to set: "); Arrays.asList(arr).stream().collect(Collectors.toSet()).forEach(System.out::print); System.out.println("directly convert to set:"); Set set = new HashSet<>(Arrays.asList(arr)); return new ArrayList<>(set).toArray(new Integer[set.size()]);}// 9. 利用LinkedHashSet数据结构直接去重复项。有序链表。static Integer[] unique9(Integer arr[]) { Set linkedHashSet = new LinkedHashSet<>(Arrays.asList(arr)); return new ArrayList<>(linkedHashSet).toArray(new Integer[linkedHashSet.size()]);}// 10. 利用TreeSet数据结构直接去重复项。自然排序和定制排序。static Integer[] unique10(Integer arr[]) { Set treeSet = new TreeSet<>(Arrays.asList(arr)).descendingSet(); return new ArrayList<>(treeSet).toArray(new Integer[treeSet.size()]);}// 11. 提前排序，从后向前遍历，将当前项与前一项对比，如果重复则移除当前项static Integer[] unique11(Integer arr[]) { List list = new ArrayList<>(Arrays.asList(arr)); Collections.sort(list); for (int l = list.size() - 1; l > 0; l--) { if (list.get(l).equals(list.get(l - 1))) { list.remove(l); } } return new ArrayList<>(list).toArray(new Integer[list.size()]);}// 12. 提前排序，自前往后遍历，将当前项与后一项对比，如果重复则移除当前项static Integer[] unique12(Integer arr[]) { List list = new ArrayList<>(Arrays.asList(arr)); Collections.sort(list, Collections.reverseOrder()); int l = list.size() - 1; for (int i = 0; i < l; i++) { if (list.get(i).equals(list.get(i + 1))) { list.remove(i); i--; l--; } } return new ArrayList<>(list).toArray(new Integer[list.size()]);}// 13. 转为stream，利用distinct方法去重复static Integer[] unique13(Integer arr[]) { List list = new ArrayList<>(Arrays.asList(arr)); list = list.stream().distinct().collect(Collectors.toList()); return new ArrayList<>(list).toArray(new Integer[list.size()]);}// 14. 双循环自右往左逐个与左侧项对比，如遇相同则跳过当前项，下一项为当前项，继续逐个与左侧项对比static Integer[] unique14(Integer arr[]) { int len = arr.length; Integer[] result = new Integer[len]; int x = len; for (int i = len - 1; i >= 0; i--) { for (int j = i - 1; j >= 0; j--) { if (arr[i].equals(arr[j])) { i--; j = i; } } // 非重复项的为唯一，追加到新数组 result[--x] = arr[i]; } return Arrays.copyOfRange(result, x, len);}// 15. 利用Interator来遍历List，如果不在新列表中则添加static Integer[] unique15(Integer arr[]) { List list = new ArrayList<>(Arrays.asList(arr)); List result = new ArrayList<>(); Iterator it = list.iterator(); while (it.hasNext()) { Integer item = it.next(); if (!result.contains(item)) { result.add(item); } } return new ArrayList<>(result).toArray(new Integer[result.size()]);}// 16. 利用递归调用来去重复。递归自后往前逐个调用，当长度为1时终止。// 当后一项与前任一项相同说明有重复，则删除当前项。相当于利用自我调用来替换循环static Integer[] uniqueRecursion1(Integer arr[], int len, List result) { int last = len - 1; Integer lastItem = arr[last]; int l = last; boolean isRepeat = false; if (len <= 1) { result.add(0, lastItem); return new ArrayList<>(result).toArray(new Integer[result.size()]); } while (l-- > 0) { if (lastItem.equals(arr[l])) { isRepeat = true; break; } } // 如果不重复表示唯一，则添加到新数组中 if (!isRepeat) { result.add(0, lastItem); } return uniqueRecursion1(arr, len - 1, result);}// 17. 利用递归调用来去重复的另外一种方式。递归自后往前逐个调用，当长度为1时终止。// 与上一个递归不同，这里将不重复的项目作为结果拼接起来static List uniqueRecursion2(List arr, int len) { if (len <= 1) { System.out.println("last arr:" + arr); return arr; } int last = len - 1; int l = last - 1; boolean isRepeat = false; Integer lastItem = arr.get(last); while (l >= 0) { if (lastItem.equals(arr.get(l))) { isRepeat = true; break; } l--; } // 如果不重复则添加到临时列表，最后将全部结果拼接 List result = new ArrayList<>(); arr.remove(last); if (!isRepeat) { result.add(lastItem); } return Stream.concat(uniqueRecursion2(arr, len - 1).stream(), result.stream()).collect(Collectors.toList());}// 18. 双重循环，将左侧项逐个与当前项比较。如果遇到值想等则比较下标，下标相同则追加到新数组。这里与第1个方案稍微的不同。static Integer[] unique18(Integer arr[]) { Integer newArr[] = new Integer[arr.length]; int x = 0; for (int i = 0; i < arr.length; i++) { int j = 0;; for (; j < i; j++) { if (arr[i].equals(arr[j])) { break; } } if (i == j) { newArr[x] = arr[i]; x++; } } return Arrays.copyOf(newArr, x);}执行测试
import java.util.*;import java.util.stream.Collectors;import java.util.stream.Stream;public class UniqueArray { // 此处省略代码...public static void main(final String args[]) { int arr1[] = { 1, 3, -1, 1, 2, 2, 4, 2, 2, -1 }; int[] result; long startTime; int arr1[] = { 1, 3, -1, 1, 2, 2, 4, 2, 2, -1 }; int[] result; long startTime; // 1. System.out.println("unique1 start:" + Arrays.toString(arr1)); startTime = System.currentTimeMillis(); result = UniqueArray.unique1(arr1); System.out.println("unique1 result:" + Arrays.toString(result)); System.out.println("ime:" + (System.currentTimeMillis() - startTime) + " ms."); // 2. Integer arr2[] = { 1, 3, -1, 1, 2, 2, 4, 2, 2, -1 }; System.out.println("unique2 start:" + Arrays.toString(arr2)); startTime = System.currentTimeMillis(); Integer result2[] = UniqueArray.unique2(arr2); System.out.println("unique2 result:" + Arrays.toString(result2)); System.out.println("ime:" + (System.currentTimeMillis() - startTime) + " ms."); // 3. Integer arr3[] = { 1, 3, -1, 1, 2, 2, 4, 2, 2, -1 }; System.out.println("unique3 start:" + Arrays.toString(arr2)); startTime = System.currentTimeMillis(); Integer result3[] = UniqueArray.unique3(arr3); System.out.println("unique3 result:" + Arrays.toString(result3)); System.out.println("ime:" + (System.currentTimeMillis() - startTime) + " ms."); // 4. Integer arr4[] = { 1, 3, -1, 1, 2, 2, 4, 2, 2, -1 }; System.out.println("unique4 start:" + Arrays.toString(arr4)); startTime = System.currentTimeMillis(); Integer result4[] = UniqueArray.unique4(arr4); System.out.println("unique4 result:" + Arrays.toString(result4)); System.out.println("ime:" + (System.currentTimeMillis() - startTime) + " ms."); // 5. Integer arr5[] = { 1, 3, -1, 1, 2, 2, 4, 2, 2, -1 }; System.out.println("unique5 start:" + Arrays.toString(arr5)); startTime = System.currentTimeMillis(); Integer result5[] = UniqueArray.unique5(arr5); System.out.println("unique5 result:" + Arrays.toString(result5)); System.out.println("ime:" + (System.currentTimeMillis() - startTime) + " ms."); // 6. Integer arr6[] = { 1, 3, -1, 1, 2, 2, 4, 2, 2, -1 }; System.out.println("unique6 start:" + Arrays.toString(arr6)); startTime = System.currentTimeMillis(); Integer result6[] = UniqueArray.unique6(arr6); System.out.println("unique6 result:" + Arrays.toString(result6)); System.out.println("ime:" + (System.currentTimeMillis() - startTime) + " ms."); // 7. Integer arr7[] = { 1, 3, -1, 1, 2, 2, 4, 2, 2, -1 }; System.out.println("unique7 start:" + Arrays.toString(arr7)); startTime = System.currentTimeMillis(); Integer result7[] = UniqueArray.unique7(arr7); System.out.println("unique7 result:" + Arrays.toString(result7)); System.out.println("ime:" + (System.currentTimeMillis() - startTime) + " ms."); // 8. Integer arr8[] = { 1, 3, -1, 1, 2, 2, 4, 2, 2, -1 }; System.out.println("unique8 start:" + Arrays.toString(arr8)); startTime = System.currentTimeMillis(); Integer result8[] = UniqueArray.unique8(arr8); System.out.println("unique8 result:" + Arrays.toString(result8)); System.out.println("ime:" + (System.currentTimeMillis() - startTime) + " ms."); // 9. Integer arr9[] = { 1, 3, -1, 1, 2, 2, 4, 2, 2, -1 }; System.out.println("unique9 start:" + Arrays.toString(arr9)); startTime = System.currentTimeMillis(); Integer result9[] = UniqueArray.unique9(arr9); System.out.println("unique9 result:" + Arrays.toString(result9)); System.out.println("ime:" + (System.currentTimeMillis() - startTime) + " ms."); // 10. Integer arr10[] = { 1, 3, -1, 1, 2, 2, 4, 2, 2, -1 }; System.out.println("unique10 start:" + Arrays.toString(arr10)); startTime = System.currentTimeMillis(); Integer result10[] = UniqueArray.unique10(arr10); System.out.println("unique10 result:" + Arrays.toString(result10)); System.out.println("ime:" + (System.currentTimeMillis() - startTime) + " ms."); // 11. Integer arr11[] = { 1, 3, -1, 1, 2, 2, 4, 2, 2, -1 }; System.out.println("unique11 start:" + Arrays.toString(arr11)); startTime = System.currentTimeMillis(); Integer result11[] = UniqueArray.unique11(arr11); System.out.println("unique11 result:" + Arrays.toString(result11)); System.out.println("ime:" + (System.currentTimeMillis() - startTime) + " ms."); // 12. Integer arr12[] = { 1, 3, -1, 1, 2, 2, 4, 2, 2, -1 }; System.out.println("unique12 start:" + Arrays.toString(arr12)); startTime = System.currentTimeMillis(); Integer result12[] = UniqueArray.unique12(arr12); System.out.println("unique12 result:" + Arrays.toString(result12)); System.out.println("ime:" + (System.currentTimeMillis() - startTime) + " ms."); // 13. Integer arr13[] = { 1, 3, -1, 1, 2, 2, 4, 2, 2, -1 }; System.out.println("unique13 start:" + Arrays.toString(arr13)); startTime = System.currentTimeMillis(); Integer result13[] = UniqueArray.unique13(arr13); System.out.println("unique13 result:" + Arrays.toString(result13)); System.out.println("ime:" + (System.currentTimeMillis() - startTime) + " ms."); // 14. Integer arr14[] = { 1, 3, -1, 1, 2, 2, 4, 2, 2, -1 }; System.out.println("unique14 start:" + Arrays.toString(arr14)); startTime = System.currentTimeMillis(); Integer result14[] = UniqueArray.unique14(arr14); System.out.println("unique14 result:" + Arrays.toString(result14)); System.out.println("ime:" + (System.currentTimeMillis() - startTime) + " ms."); // 15. Integer arr15[] = { 1, 3, -1, 1, 2, 2, 4, 2, 2, -1 }; System.out.println("unique15 start:" + Arrays.toString(arr15)); startTime = System.currentTimeMillis(); Integer result15[] = UniqueArray.unique15(arr15); System.out.println("unique15 result:" + Arrays.toString(result15)); System.out.println("ime:" + (System.currentTimeMillis() - startTime) + " ms."); // 16. Integer arr16[] = { 1, 3, -1, 1, 2, 2, 4, 2, 2, -1 }; System.out.println("uniqueRecursion1 start:" + Arrays.toString(arr16)); startTime = System.currentTimeMillis(); Integer result16[] = UniqueArray.uniqueRecursion1(arr16, arr16.length, new ArrayList<>()); System.out.println("uniqueRecursion1 result:" + Arrays.toString(result16)); System.out.println("ime:" + (System.currentTimeMillis() - startTime) + " ms."); // 17. Integer arr17[] = { 1, 3, -1, 1, 2, 2, 4, 2, 2, -1 }; System.out.println("uniqueRecursion2 start:" + Arrays.toString(arr17)); startTime = System.currentTimeMillis(); List result17 = UniqueArray.uniqueRecursion2(new ArrayList<>(Arrays.asList(arr17)), arr17.length); System.out.println("uniqueRecursion2 result:" + result17); System.out.println("ime:" + (System.currentTimeMillis() - startTime) + " ms."); // 18. Integer arr18[] = { 1, 3, -1, 1, 2, 2, 4, 2, 2, -1 }; System.out.println("unique18 start:" + Arrays.toString(arr18)); startTime = System.currentTimeMillis(); Integer result18[] = UniqueArray.unique18(arr18); System.out.println("unique18 result:" + Arrays.toString(result18)); System.out.println("ime:" + (System.currentTimeMillis() - startTime) + " ms.");打印结果
MacBook-Pro:unique jarry$ java --version java 10.0.1 2018-04-17 Java(TM) SE Runtime Environment 18.3 (build 10.0.1+10) Java HotSpot(TM) 64-Bit Server VM 18.3 (build 10.0.1+10, mixed mode)MacBook-Pro:unique jarry$ javac UniqueArray.java MacBook-Pro:unique jarry$ java UniqueArrayunique1 start:[1, 3, -1, 1, 2, 2, 4, 2, 2, -1]unique1 result:[1, 3, -1, 2, 4]time:1 ms.unique2 start:[1, 3, -1, 1, 2, 2, 4, 2, 2, -1]unique2 result:[1, 3, -1, 2, 4]// ...其他省略...etctime:0 ms.uniqueRecursion2 start:[1, 3, -1, 1, 2, 2, 4, 2, 2, -1]last arr:[1]uniqueRecursion2 result:[1, 3, -1, 2, 4]time:1 ms.unique18 start:[1, 3, -1, 1, 2, 2, 4, 2, 2, -1]unique18 result:[1, 3, -1, 2, 4]思考
通过去重复的多种实现，其实是让我们更了解Java语言，还可以让我们根据场景采用不用方式，毕竟业务场景千差万别。你常用哪几种方式？你还有其他的实现方式吗？
扫一扫
1262
被折叠的 条评论
为什么被折叠?
到【灌水乐园】发言
