Rename num_chunks to size

Author: kderusso

x-pack/plugin/inference/src/main/java/org/elasticsearch/xpack/inference/rank/textsimilarity/ChunkScorerConfig.java

@@ -20,12 +20,12 @@
 
 public class ChunkScorerConfig implements Writeable {
 
-    public final Integer numChunks;
+    public final Integer size;
     private final String inferenceText;
     private final ChunkingSettings chunkingSettings;
 
     public static final int DEFAULT_CHUNK_SIZE = 300;
-    public static final int DEFAULT_NUM_CHUNKS = 1;
+    public static final int DEFAULT_SIZE = 1;
 
     public static ChunkingSettings createChunkingSettings(Integer chunkSize) {
         int chunkSizeOrDefault = chunkSize != null ? chunkSize : DEFAULT_CHUNK_SIZE;
@@ -48,31 +48,31 @@ public static ChunkingSettings chunkingSettingsFromMap(Map<String, Object> map)
     }
 
     public ChunkScorerConfig(StreamInput in) throws IOException {
-        this.numChunks = in.readOptionalVInt();
+        this.size = in.readOptionalVInt();
         this.inferenceText = in.readString();
         Map<String, Object> chunkingSettingsMap = in.readGenericMap();
         this.chunkingSettings = ChunkingSettingsBuilder.fromMap(chunkingSettingsMap);
     }
 
-    public ChunkScorerConfig(Integer numChunks, ChunkingSettings chunkingSettings) {
-        this(numChunks, null, chunkingSettings);
+    public ChunkScorerConfig(Integer size, ChunkingSettings chunkingSettings) {
+        this(size, null, chunkingSettings);
     }
 
-    public ChunkScorerConfig(Integer numChunks, String inferenceText, ChunkingSettings chunkingSettings) {
-        this.numChunks = numChunks;
+    public ChunkScorerConfig(Integer size, String inferenceText, ChunkingSettings chunkingSettings) {
+        this.size = size;
         this.inferenceText = inferenceText;
         this.chunkingSettings = chunkingSettings;
     }
 
     @Override
     public void writeTo(StreamOutput out) throws IOException {
-        out.writeOptionalVInt(numChunks);
+        out.writeOptionalVInt(size);
         out.writeString(inferenceText);
         out.writeGenericMap(chunkingSettings.asMap());
     }
 
-    public Integer numChunks() {
-        return numChunks;
+    public Integer size() {
+        return size;
     }
 
     public String inferenceText() {
@@ -88,13 +88,13 @@ public boolean equals(Object o) {
         if (this == o) return true;
         if (o == null || getClass() != o.getClass()) return false;
         ChunkScorerConfig that = (ChunkScorerConfig) o;
-        return Objects.equals(numChunks, that.numChunks)
+        return Objects.equals(size, that.size)
             && Objects.equals(inferenceText, that.inferenceText)
             && Objects.equals(chunkingSettings, that.chunkingSettings);
     }
 
     @Override
     public int hashCode() {
-        return Objects.hash(numChunks, inferenceText, chunkingSettings);
+        return Objects.hash(size, inferenceText, chunkingSettings);
     }
 }

x-pack/plugin/inference/src/main/java/org/elasticsearch/xpack/inference/rank/textsimilarity/TextSimilarityRankBuilder.java

@@ -198,7 +198,7 @@ public RankFeaturePhaseRankCoordinatorContext buildRankFeaturePhaseCoordinatorCo
             minScore,
             failuresAllowed,
             chunkScorerConfig != null
-                ? new ChunkScorerConfig(chunkScorerConfig.numChunks, inferenceText, chunkScorerConfig.chunkingSettings())
+                ? new ChunkScorerConfig(chunkScorerConfig.size, inferenceText, chunkScorerConfig.chunkingSettings())
                 : null
         );
     }

x-pack/plugin/inference/src/main/java/org/elasticsearch/xpack/inference/rank/textsimilarity/TextSimilarityRankRetrieverBuilder.java

@@ -53,7 +53,7 @@ public class TextSimilarityRankRetrieverBuilder extends CompoundRetrieverBuilder
     public static final ParseField FIELD_FIELD = new ParseField("field");
     public static final ParseField FAILURES_ALLOWED_FIELD = new ParseField("allow_rerank_failures");
     public static final ParseField CHUNK_RESCORER_FIELD = new ParseField("chunk_rescorer");
-    public static final ParseField NUM_CHUNKS_FIELD = new ParseField("num_chunks");
+    public static final ParseField CHUNK_SIZE_FIELD = new ParseField("size");
     public static final ParseField CHUNKING_SETTINGS_FIELD = new ParseField("chunking_settings");
 
     public static final ConstructingObjectParser<TextSimilarityRankRetrieverBuilder, RetrieverParserContext> PARSER =
@@ -79,11 +79,11 @@ public class TextSimilarityRankRetrieverBuilder extends CompoundRetrieverBuilder
 
     private static final ConstructingObjectParser<ChunkScorerConfig, RetrieverParserContext> CHUNK_SCORER_PARSER =
         new ConstructingObjectParser<>(CHUNK_RESCORER_FIELD.getPreferredName(), true, args -> {
-            Integer numChunks = (Integer) args[0];
+            Integer size = (Integer) args[0];
             @SuppressWarnings("unchecked")
             Map<String, Object> chunkingSettingsMap = (Map<String, Object>) args[1];
             ChunkingSettings chunkingSettings = ChunkScorerConfig.chunkingSettingsFromMap(chunkingSettingsMap);
-            return new ChunkScorerConfig(numChunks, chunkingSettings);
+            return new ChunkScorerConfig(size, chunkingSettings);
         });
 
     static {
@@ -99,7 +99,7 @@ public class TextSimilarityRankRetrieverBuilder extends CompoundRetrieverBuilder
         PARSER.declareBoolean(optionalConstructorArg(), FAILURES_ALLOWED_FIELD);
         PARSER.declareObject(optionalConstructorArg(), CHUNK_SCORER_PARSER, CHUNK_RESCORER_FIELD);
         if (RERANK_RESCORE_CHUNKS.isEnabled()) {
-            CHUNK_SCORER_PARSER.declareInt(optionalConstructorArg(), NUM_CHUNKS_FIELD);
+            CHUNK_SCORER_PARSER.declareInt(optionalConstructorArg(), CHUNK_SIZE_FIELD);
             CHUNK_SCORER_PARSER.declareObjectOrNull(optionalConstructorArg(), (p, c) -> p.map(), null, CHUNKING_SETTINGS_FIELD);
         }
 
@@ -156,8 +156,8 @@ public TextSimilarityRankRetrieverBuilder(
         if (retrieverSource.size() != 1) {
             throw new IllegalArgumentException("[" + getName() + "] retriever should have exactly one inner retriever");
         }
-        if (chunkScorerConfig != null && chunkScorerConfig.numChunks() != null && chunkScorerConfig.numChunks() < 1) {
-            throw new IllegalArgumentException("num_chunks must be greater than 0, was: " + chunkScorerConfig.numChunks());
+        if (chunkScorerConfig != null && chunkScorerConfig.size() != null && chunkScorerConfig.size() < 1) {
+            throw new IllegalArgumentException("size must be greater than 0, was: " + chunkScorerConfig.size());
         }
         this.inferenceId = inferenceId;
         this.inferenceText = inferenceText;
@@ -220,7 +220,7 @@ protected SearchSourceBuilder finalizeSourceBuilder(SearchSourceBuilder sourceBu
                 minScore,
                 failuresAllowed,
                 chunkScorerConfig != null
-                    ? new ChunkScorerConfig(chunkScorerConfig.numChunks, inferenceText, chunkScorerConfig.chunkingSettings())
+                    ? new ChunkScorerConfig(chunkScorerConfig.size, inferenceText, chunkScorerConfig.chunkingSettings())
                     : null
             )
         );
@@ -252,8 +252,8 @@ protected void doToXContent(XContentBuilder builder, Params params) throws IOExc
         }
         if (chunkScorerConfig != null) {
             builder.startObject(CHUNK_RESCORER_FIELD.getPreferredName());
-            if (chunkScorerConfig.numChunks() != null) {
-                builder.field(NUM_CHUNKS_FIELD.getPreferredName(), chunkScorerConfig.numChunks());
+            if (chunkScorerConfig.size() != null) {
+                builder.field(CHUNK_SIZE_FIELD.getPreferredName(), chunkScorerConfig.size());
             }
             if (chunkScorerConfig.chunkingSettings() != null) {
                 builder.field(CHUNKING_SETTINGS_FIELD.getPreferredName(), chunkScorerConfig.chunkingSettings().asMap());

x-pack/plugin/inference/src/main/java/org/elasticsearch/xpack/inference/rank/textsimilarity/TextSimilarityRerankingRankFeaturePhaseRankShardContext.java

@@ -23,7 +23,7 @@
 import java.io.IOException;
 import java.util.List;
 
-import static org.elasticsearch.xpack.inference.rank.textsimilarity.ChunkScorerConfig.DEFAULT_NUM_CHUNKS;
+import static org.elasticsearch.xpack.inference.rank.textsimilarity.ChunkScorerConfig.DEFAULT_SIZE;
 
 public class TextSimilarityRerankingRankFeaturePhaseRankShardContext extends RerankingRankFeaturePhaseRankShardContext {
 
@@ -47,7 +47,7 @@ public RankShardResult doBuildRankFeatureShardResult(SearchHits hits, int shardI
             DocumentField docField = hit.field(field);
             if (docField != null) {
                 if (chunkScorerConfig != null) {
-                    int numChunks = chunkScorerConfig.numChunks() != null ? chunkScorerConfig.numChunks() : DEFAULT_NUM_CHUNKS;
+                    int size = chunkScorerConfig.size() != null ? chunkScorerConfig.size() : DEFAULT_SIZE;
                     List<Chunker.ChunkOffset> chunkOffsets = chunker.chunk(docField.getValue().toString(), chunkingSettings);
                     List<String> chunks = chunkOffsets.stream()
                         .map(offset -> { return docField.getValue().toString().substring(offset.start(), offset.end()); })
@@ -59,9 +59,9 @@ public RankShardResult doBuildRankFeatureShardResult(SearchHits hits, int shardI
                         List<MemoryIndexChunkScorer.ScoredChunk> scoredChunks = scorer.scoreChunks(
                             chunks,
                             chunkScorerConfig.inferenceText(),
-                            numChunks
+                            size
                         );
-                        bestChunks = scoredChunks.stream().map(MemoryIndexChunkScorer.ScoredChunk::content).limit(numChunks).toList();
+                        bestChunks = scoredChunks.stream().map(MemoryIndexChunkScorer.ScoredChunk::content).limit(size).toList();
                     } catch (IOException e) {
                         throw new IllegalStateException("Could not generate chunks for input to reranker", e);
                     }

x-pack/plugin/inference/src/yamlRestTest/resources/rest-api-spec/test/inference/70_text_similarity_rank_retriever.yml

@@ -522,7 +522,7 @@ setup:
       reason: rescore_chunks introduced in 9.2.0
 
   - do:
-      catch: /num_chunks must be greater than 0/
+      catch: /size must be greater than 0/
       search:
         index: test-index
         body:
@@ -539,7 +539,7 @@ setup:
               inference_text: "How often does the moon hide the sun?"
               field: inference_text_field
               chunk_rescorer:
-                num_chunks: 0
+                size: 0
           size: 10
 
   - match: { status: 400 }
@@ -570,7 +570,7 @@ setup:
               inference_text: "How often does the moon hide the sun?"
               field: text
               chunk_rescorer:
-                num_chunks: 2
+                size: 2
           size: 10
 
   - match: { hits.total.value: 2 }
@@ -638,7 +638,7 @@ setup:
               inference_text: "how often does the moon hide the sun?"
               field: semantic_text_field
               chunk_rescorer:
-                num_chunks: 2
+                size: 2
           size: 10
 
   - match: { hits.total.value: 2 }