PR 13757 follow-up: add missing with-discountOverlaps Similarity constructor variants, CHANGES.txt entries (#13845)

lucene/CHANGES.txt

@@ -9,6 +9,11 @@ Bug Fixes
 ---------------------
 (No changes)
 
+API Changes
+---------------------
+
+* GITHUB#13845: Add missing with-discountOverlaps Similarity constructor variants. (Pierre Salagnac, Christine Poerschke, Robert Muir)
+
 ======================== Lucene 9.12.0 =======================
 
 Security Fixes
@@ -47,6 +52,9 @@ API Changes
   the entire segment should be scored. Subclasses that override the method should instead override its replacement.
   (Luca Cavanna)
 
+* GITHUB#13757: For similarities, provide default computeNorm implementation and remove remaining discountOverlaps setters.
+  (Christine Poerschke, Adrien Grand, Robert Muir)
+
 New Features
 ---------------------
 

lucene/core/src/java/org/apache/lucene/search/similarities/Axiomatic.java

@@ -44,13 +44,26 @@ public abstract class Axiomatic extends SimilarityBase {
   protected final int queryLen;
 
   /**
-   * Constructor setting all Axiomatic hyperparameters
+   * Constructor setting all Axiomatic hyperparameters and using default discountOverlaps value.
    *
    * @param s hyperparam for the growth function
    * @param queryLen the query length
    * @param k hyperparam for the primitive weighting function
    */
   public Axiomatic(float s, int queryLen, float k) {
+    this(true, s, queryLen, k);
+  }
+
+  /**
+   * Constructor setting all Axiomatic hyperparameters
+   *
+   * @param discountOverlaps true if overlap tokens should not impact document length for scoring.
+   * @param s hyperparam for the growth function
+   * @param queryLen the query length
+   * @param k hyperparam for the primitive weighting function
+   */
+  public Axiomatic(boolean discountOverlaps, float s, int queryLen, float k) {
+    super(discountOverlaps);
     if (Float.isFinite(s) == false || Float.isNaN(s) || s < 0 || s > 1) {
       throw new IllegalArgumentException("illegal s value: " + s + ", must be between 0 and 1");
     }

lucene/core/src/java/org/apache/lucene/search/similarities/DFISimilarity.java

@@ -46,11 +46,23 @@ public class DFISimilarity extends SimilarityBase {
   private final Independence independence;
 
   /**
-   * Create DFI with the specified divergence from independence measure
+   * Create DFI with the specified divergence from independence measure and using default
+   * discountOverlaps value
    *
    * @param independenceMeasure measure of divergence from independence
    */
   public DFISimilarity(Independence independenceMeasure) {
+    this(independenceMeasure, true);
+  }
+
+  /**
+   * Create DFI with the specified parameters
+   *
+   * @param independenceMeasure measure of divergence from independence
+   * @param discountOverlaps true if overlap tokens should not impact document length for scoring.
+   */
+  public DFISimilarity(Independence independenceMeasure, boolean discountOverlaps) {
+    super(discountOverlaps);
     this.independence = independenceMeasure;
   }
 

lucene/core/src/java/org/apache/lucene/search/similarities/DFRSimilarity.java

@@ -83,7 +83,7 @@ public class DFRSimilarity extends SimilarityBase {
   protected final Normalization normalization;
 
   /**
-   * Creates DFRSimilarity from the three components.
+   * Creates DFRSimilarity from the three components and using default discountOverlaps value.
    *
    * <p>Note that <code>null</code> values are not allowed: if you want no normalization, instead
    * pass {@link NoNormalization}.
@@ -98,7 +98,7 @@ public DFRSimilarity(
   }
 
   /**
-   * Creates DFRSimilarity from the three components.
+   * Creates DFRSimilarity from the three components and with the specified discountOverlaps value.
    *
    * <p>Note that <code>null</code> values are not allowed: if you want no normalization, instead
    * pass {@link NoNormalization}.

lucene/core/src/java/org/apache/lucene/search/similarities/IBSimilarity.java

@@ -76,7 +76,7 @@ public class IBSimilarity extends SimilarityBase {
   protected final Normalization normalization;
 
   /**
-   * Creates IBSimilarity from the three components.
+   * Creates IBSimilarity from the three components and using default discountOverlaps value.
    *
    * <p>Note that <code>null</code> values are not allowed: if you want no normalization, instead
    * pass {@link NoNormalization}.
@@ -86,6 +86,26 @@ public class IBSimilarity extends SimilarityBase {
    * @param normalization term frequency normalization
    */
   public IBSimilarity(Distribution distribution, Lambda lambda, Normalization normalization) {
+    this(distribution, lambda, normalization, true);
+  }
+
+  /**
+   * Creates IBSimilarity from the three components and with the specified discountOverlaps value.
+   *
+   * <p>Note that <code>null</code> values are not allowed: if you want no normalization, instead
+   * pass {@link NoNormalization}.
+   *
+   * @param distribution probabilistic distribution modeling term occurrence
+   * @param lambda distribution's &lambda;<sub>w</sub> parameter
+   * @param normalization term frequency normalization
+   * @param discountOverlaps true if overlap tokens should not impact document length for scoring.
+   */
+  public IBSimilarity(
+      Distribution distribution,
+      Lambda lambda,
+      Normalization normalization,
+      boolean discountOverlaps) {
+    super(discountOverlaps);
     this.distribution = distribution;
     this.lambda = lambda;
     this.normalization = normalization;

lucene/core/src/java/org/apache/lucene/search/similarities/IndriDirichletSimilarity.java

@@ -37,6 +37,13 @@ public class IndriDirichletSimilarity extends LMSimilarity {
   /** The μ parameter. */
   private final float mu;
 
+  /** Instantiates the similarity with the provided parameters. */
+  public IndriDirichletSimilarity(
+      CollectionModel collectionModel, boolean discountOverlaps, float mu) {
+    super(collectionModel, discountOverlaps);
+    this.mu = mu;
+  }
+
   /** Instantiates the similarity with the provided μ parameter. */
   public IndriDirichletSimilarity(CollectionModel collectionModel, float mu) {
     super(collectionModel);

lucene/core/src/java/org/apache/lucene/search/similarities/LMDirichletSimilarity.java

@@ -39,7 +39,13 @@ public class LMDirichletSimilarity extends LMSimilarity {
 
   /** Instantiates the similarity with the provided μ parameter. */
   public LMDirichletSimilarity(CollectionModel collectionModel, float mu) {
-    super(collectionModel);
+    this(collectionModel, true, mu);
+  }
+
+  /** Instantiates the similarity with the provided parameters. */
+  public LMDirichletSimilarity(
+      CollectionModel collectionModel, boolean discountOverlaps, float mu) {
+    super(collectionModel, discountOverlaps);
     if (Float.isFinite(mu) == false || mu < 0) {
       throw new IllegalArgumentException(
           "illegal mu value: " + mu + ", must be a non-negative finite value");

lucene/core/src/java/org/apache/lucene/search/similarities/LMJelinekMercerSimilarity.java

@@ -43,7 +43,13 @@ public class LMJelinekMercerSimilarity extends LMSimilarity {
 
   /** Instantiates with the specified collectionModel and λ parameter. */
   public LMJelinekMercerSimilarity(CollectionModel collectionModel, float lambda) {
-    super(collectionModel);
+    this(collectionModel, true, lambda);
+  }
+
+  /** Instantiates with the specified collectionModel and parameters. */
+  public LMJelinekMercerSimilarity(
+      CollectionModel collectionModel, boolean discountOverlaps, float lambda) {
+    super(collectionModel, discountOverlaps);
     if (Float.isNaN(lambda) || lambda <= 0 || lambda > 1) {
       throw new IllegalArgumentException("lambda must be in the range (0 .. 1]");
     }

lucene/core/src/java/org/apache/lucene/search/similarities/LMSimilarity.java

@@ -43,6 +43,12 @@ public abstract class LMSimilarity extends SimilarityBase {
 
   /** Creates a new instance with the specified collection language model. */
   public LMSimilarity(CollectionModel collectionModel) {
+    this(collectionModel, true);
+  }
+
+  /** Creates a new instance with the specified collection language model and discountOverlaps. */
+  public LMSimilarity(CollectionModel collectionModel, boolean discountOverlaps) {
+    super(discountOverlaps);
     this.collectionModel = collectionModel;
   }