build fix and PR feedback

Author: heylissi

README.md

@@ -173,15 +173,14 @@ Subsequence Match Merging
                           (between 1 and 50, default: 6).
 
 Frequency Analysis
-      --frequency-strategy=<{COMPLETE_MATCHES, CONTAINED_MATCHES, SUB_MATCHES, 
-                        WINDOW_OF_MATCHES}>
+      --frequency-strategy=<{COMPLETE_MATCHES, CONTAINED_MATCHES, SUB_MATCHES,
+        WINDOW_OF_MATCHES}>
                         Strategy for frequency Analysis, one of:
-                         COMPLETE_MATCHES, CONTAINED_MATCHES, SUB_MATCHES,
-                         WINDOW_OF_MATCHES (default: null).
+                          COMPLETE_MATCHES, CONTAINED_MATCHES, SUB_MATCHES,
+                          WINDOW_OF_MATCHES (default: null).
       --weighting-strategy=<{PROPORTIONAL, LINEAR, QUADRATIC, SIGMOID}>
                         Strategy for frequency Weighting, one of: PROPORTIONAL,
-                         LINEAR, QUADRATIC, SIGMOID
-                          (default: null).
+                          LINEAR, QUADRATIC, SIGMOID (default: null).
 Languages:
   c
   cpp

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

@@ -71,7 +71,8 @@ public final double minimalSimilarity() {
 
     /**
      * Computes the average (or symmetric) similarity between the two submissions. The similarity is adjusted based on
-     * whether both submissions contain base code matches.
+     * whether both submissions contain base code matches. Checks if the frequency analysis is used, if it is then a
+     * frequency based similarity score will be returned.
      * @return Average similarity in interval [0, 1]. 0 means zero percent structural similarity, 1 means 100 percent
      * structural similarity.
      */

core/src/main/java/de/jplag/highlightextraction/ContainedMatchesStrategy.java

@@ -52,9 +52,11 @@ public double calculateMatchFrequency(Match match, Map<List<TokenType>, Integer>
         List<List<TokenType>> subSequences = getSubSequences(matchToken, minSubSequenceLength);
         List<Integer> frequencies = new ArrayList<>();
         for (List<TokenType> subsequence : subSequences) {
-            frequencies.add(frequencyMap.getOrDefault(subsequence, 0));
+            Integer subSequenceFrequency = frequencyMap.get(subsequence);
+            if (subSequenceFrequency != null && subSequenceFrequency != 0) {
+                frequencies.add(subSequenceFrequency);
+            }
         }
-        return frequencies.stream().filter(frequency -> frequency > 0).mapToInt(Integer::intValue).average().orElse(0.0);
+        return frequencies.stream().mapToInt(Integer::intValue).average().orElse(0.0);
     }
-
 }

docs/1.-How-to-Use-JPlag.md

@@ -108,15 +108,14 @@ Subsequence Match Merging
                           (between 1 and 50, default: 6).
 
 Frequency Analysis
-      --frequency-strategy=<{COMPLETE_MATCHES,
-        CONTAINED_MATCHES, SUB_MATCHES, WINDOW_OF_MATCHES}>
-                        Strategy for frequency Analysis, one
-                          of: COMPLETE_MATCHES, CONTAINED_MATCHES, SUB_MATCHES,
+      --frequency-strategy=<{COMPLETE_MATCHES, CONTAINED_MATCHES, SUB_MATCHES,
+        WINDOW_OF_MATCHES}>
+                        Strategy for frequency Analysis, one of:
+                          COMPLETE_MATCHES, CONTAINED_MATCHES, SUB_MATCHES,
                           WINDOW_OF_MATCHES (default: null).
       --weighting-strategy=<{PROPORTIONAL, LINEAR, QUADRATIC, SIGMOID}>
-                        Strategy for frequency Weighting, one
-                          of: PROPORTIONAL, LINEAR, QUADRATIC, SIGMOID
-                          (default: null).
+                        Strategy for frequency Weighting, one of: PROPORTIONAL,
+                          LINEAR, QUADRATIC, SIGMOID (default: null).
 Languages:
   c
   cpp