Merge pull request #2050 from jplag/reverse-distribution

Reverse distribution

core/src/main/java/de/jplag/JPlagResult.java

@@ -1,5 +1,6 @@
 package de.jplag;
 
+import java.util.Arrays;
 import java.util.Comparator;
 import java.util.List;
 import java.util.function.ToDoubleFunction;
@@ -98,8 +99,8 @@ public JPlagOptions getOptions() {
 
     /**
      * For the {@link SimilarityMetric} JPlag was run with, this returns the similarity distribution of detected matches in
-     * a 10-element array. Each entry represents the absolute frequency of matches whose similarity lies within the
-     * respective interval. Intervals: 0: [0% - 10%), 1: [10% - 20%), 2: [20% - 30%), ..., 9: [90% - 100%]
+     * a 100-element array. Each entry represents the absolute frequency of matches whose similarity lies within the
+     * respective interval. Intervals: 0: [0% - 1%), 1: [1% - 2%), 2: [2% - 3%), ..., 99: [99% - 100%]
      * @return the similarity distribution array.
      */
     public int[] getSimilarityDistribution() {
@@ -108,9 +109,9 @@ public int[] getSimilarityDistribution() {
 
     /**
      * For the {@link SimilarityMetric#MAX} that is built in to every {@link JPlagComparison}, this returns the similarity
-     * distribution of detected matches in a 10-element array. Each entry represents the absolute frequency of matches whose
-     * similarity lies within the respective interval. Intervals: 0: [0% - 10%), 1: [10% - 20%), 2: [20% - 30%), ..., 9:
-     * [90% - 100%]
+     * distribution of detected matches in a 100-element array. Each entry represents the absolute frequency of matches
+     * whose similarity lies within the respective interval. Intervals: 0: [0% - 1%), 1: [1% - 20%), 2: [2% - 3%), ..., 99:
+     * [99% - 100%]
      * @return the similarity distribution array. When JPlag was run with the {@link SimilarityMetric#MAX}, this will return
      * the same distribution as {@link JPlagResult#getSimilarityDistribution()}
      */
@@ -122,6 +123,16 @@ public List<ClusteringResult<Submission>> getClusteringResult() {
         return this.clusteringResult;
     }
 
+    /**
+     * Calculates the distribution of all comparisons. The distribution is boxed to a 100-Element Array, index with ranges:
+     * 0: [0%, 1%), 1: [1%, 2%), ..., 99: [99%, 100%]
+     * @param similarityMetric Metric to use
+     * @return the similarity distribution
+     */
+    public List<Integer> calculateDistributionFor(ToDoubleFunction<JPlagComparison> similarityMetric) {
+        return Arrays.stream(calculateDistributionFor(this.comparisons, similarityMetric)).boxed().toList();
+    }
+
     @Override
     public String toString() {
         return String.format("JPlagResult { comparisons: %d, duration: %d ms, language: %s, submissions: %d }", getAllComparisons().size(),

core/src/main/java/de/jplag/reporting/reportobject/mapper/MetricMapper.java

@@ -1,8 +1,6 @@
 package de.jplag.reporting.reportobject.mapper;
 
-import java.util.ArrayList;
-import java.util.Arrays;
-import java.util.Collections;
+import java.util.HashMap;
 import java.util.List;
 import java.util.Map;
 import java.util.function.Function;
@@ -29,8 +27,11 @@ public MetricMapper(Function<Submission, String> submissionToIdFunction) {
      * @return Map with key as name of metric and value as distribution
      */
     public static Map<String, List<Integer>> getDistributions(JPlagResult result) {
-        return Map.of(SimilarityMetric.AVG.name(), convertDistribution(result.getSimilarityDistribution()), SimilarityMetric.MAX.name(),
-                convertDistribution(result.getMaxSimilarityDistribution()));
+        Map<String, List<Integer>> distributions = new HashMap<>();
+        for (SimilarityMetric metric : SimilarityMetric.values()) {
+            distributions.put(metric.name(), result.calculateDistributionFor(metric));
+        }
+        return distributions;
     }
 
     /**
@@ -48,10 +49,4 @@ public List<TopComparison> getTopComparisons(JPlagResult result) {
     private Map<String, Double> getComparisonMetricMap(JPlagComparison comparison) {
         return Map.of(SimilarityMetric.AVG.name(), comparison.similarity(), SimilarityMetric.MAX.name(), comparison.maximalSimilarity());
     }
-
-    private static List<Integer> convertDistribution(int[] array) {
-        List<Integer> list = new ArrayList<>(Arrays.stream(array).boxed().toList());
-        Collections.reverse(list);
-        return list;
-    }
 }

core/src/test/java/de/jplag/reporting/reportobject/mapper/MetricMapperTest.java

@@ -5,7 +5,7 @@
 import static org.mockito.Mockito.mock;
 
 import java.util.ArrayList;
-import java.util.Collections;
+import java.util.Arrays;
 import java.util.List;
 import java.util.Map;
 
@@ -16,6 +16,7 @@
 import de.jplag.JPlagResult;
 import de.jplag.Submission;
 import de.jplag.options.JPlagOptions;
+import de.jplag.options.SimilarityMetric;
 import de.jplag.reporting.reportobject.model.TopComparison;
 
 public class MetricMapperTest {
@@ -39,7 +40,8 @@ public void test_getDistributions() {
         Map<String, List<Integer>> result = MetricMapper.getDistributions(jPlagResult);
 
         // then
-        Assertions.assertEquals(Map.of("AVG", EXPECTED_AVG_DISTRIBUTION, "MAX", EXPECTED_MAX_DISTRIBUTION), result);
+        Assertions.assertEquals(EXPECTED_AVG_DISTRIBUTION, result.get("AVG"));
+        Assertions.assertEquals(EXPECTED_MAX_DISTRIBUTION, result.get("MAX"));
     }
 
     @Test
@@ -58,9 +60,8 @@ public void test_getTopComparisons() {
     }
 
     private int[] distribution(List<Integer> expectedDistribution) {
-        var reversedDistribution = new ArrayList<>(expectedDistribution);
-        Collections.reverse(reversedDistribution);
-        return reversedDistribution.stream().mapToInt(Integer::intValue).toArray();
+        var distribution = new ArrayList<>(expectedDistribution);
+        return distribution.stream().mapToInt(Integer::intValue).toArray();
     }
 
     private CreateSubmission submission(String name) {
@@ -73,8 +74,8 @@ private Comparison comparison(CreateSubmission submission1, CreateSubmission sub
 
     private JPlagResult createJPlagResult(int[] avgDistribution, int[] maxDistribution, Comparison... createComparisonsDto) {
         JPlagResult jPlagResult = mock(JPlagResult.class);
-        doReturn(avgDistribution).when(jPlagResult).getSimilarityDistribution();
-        doReturn(maxDistribution).when(jPlagResult).getMaxSimilarityDistribution();
+        doReturn(Arrays.stream(avgDistribution).boxed().toList()).when(jPlagResult).calculateDistributionFor(SimilarityMetric.AVG);
+        doReturn(Arrays.stream(maxDistribution).boxed().toList()).when(jPlagResult).calculateDistributionFor(SimilarityMetric.MAX);
 
         JPlagOptions options = mock(JPlagOptions.class);
         doReturn(createComparisonsDto.length).when(options).maximumNumberOfComparisons();

report-viewer/src/model/Distribution.ts

@@ -11,8 +11,9 @@ export class Distribution {
   public splitIntoBuckets(bucketCount: BucketOptions): number[] {
     const bucketArray = new Array<number>(bucketCount).fill(0)
     const divisor = 100 / bucketCount
+    const reversedDistribution = Array.from(this._distribution).reverse()
     for (let i = 99; i >= 0; i--) {
-      bucketArray[Math.floor(i / divisor)] += this._distribution[i]
+      bucketArray[Math.floor(i / divisor)] += reversedDistribution[i]
     }
     return bucketArray
   }

report-viewer/tests/unit/model/Distribution.test.ts

@@ -2,10 +2,10 @@ import { describe, expect, it } from 'vitest'
 import { Distribution } from '@/model/Distribution'
 
 const distributionData = [
+  0, 7, 15, 42, 109, 225, 470, 869, 1442, 2052, 3025, 4056, 5091, 6130, 7023, 7292, 7445, 7177,
+  6343, 5373, 4309, 3163, 2244, 1544, 923, 493, 273, 168, 61, 31, 8, 12, 2, 1, 0, 1, 2, 0, 0, 0, 0,
   0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
-  0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 2, 1, 0,
-  1, 2, 12, 8, 31, 61, 168, 273, 493, 923, 1544, 2244, 3163, 4309, 5373, 6343, 7177, 7445, 7292,
-  7023, 6130, 5091, 4056, 3025, 2052, 1442, 869, 470, 225, 109, 42, 15, 7, 0
+  0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0
 ]
 const distribution = new Distribution(distributionData)
 
@@ -14,7 +14,8 @@ describe('Distribution', () => {
     expect(distribution.splitIntoBuckets(10)).toEqual([0, 0, 0, 0, 0, 0, 26, 13209, 58955, 5231])
   }),
     it('get in 100 Buckets', () => {
-      expect(distribution.splitIntoBuckets(100)).toEqual(distributionData)
+      const reversedOriginal = Array.from(distributionData).reverse()
+      expect(distribution.splitIntoBuckets(100)).toEqual(reversedOriginal)
     }),
     it('get in 25 Buckets', () => {
       expect(distribution.splitIntoBuckets(25)).toEqual([