[SOLR-17726] Fix CloudMLTQParser to support copyField in qf

solr/core/src/java/org/apache/solr/search/mlt/CloudMLTQParser.java

@@ -67,7 +67,8 @@ protected Query likeDoc(MoreLikeThis moreLikeThis, SolrDocument doc) throws IOEx
     Map<String, Collection<Object>> filteredDocument = new HashMap<>();
 
     for (String field : moreLikeThis.getFieldNames()) {
-      Collection<Object> fieldValues = doc.getFieldValues(field);
+      Collection<Object> fieldValues = getFieldOrCopyFieldValues(doc, field);
+
       if (fieldValues != null) {
         Collection<Object> values = new ArrayList<>();
         for (Object val : fieldValues) {
@@ -110,4 +111,22 @@ private SolrDocument getDocument(String id) {
 
     return (SolrDocument) response.get("doc");
   }
+
+  private Collection<Object> getFieldOrCopyFieldValues(SolrDocument doc, String field) {
+    Collection<Object> fieldValues = doc.getFieldValues(field);
+    if (fieldValues != null) return fieldValues;
+
+    // Fields created using copyField are not included in documents returned by RealTime Get.
+    // So if a copyField destination is used in the MLT query (qf), we need to get the values
+    // from its source field instead. If there are multiple source fields, their values must be
+    // combined.
+    Collection<Object> combinedValues = new ArrayList<>();
+    for (String fieldSource : req.getSchema().getCopySources(field)) {
+      Collection<Object> sourceValues = doc.getFieldValues(fieldSource);
+      if (sourceValues != null) {
+        combinedValues.addAll(sourceValues);
+      }
+    }
+    return combinedValues.isEmpty() ? null : combinedValues;
+  }
 }

solr/core/src/test-files/solr/configsets/cloud-dynamic/conf/schema.xml

@@ -230,6 +230,18 @@
   <field name="payload" type="sortable_binary" indexed="false"
          stored="true" multiValued="false"/>
 
+  <!-- to test copyField in MLT qf -->
+  <field name="copyfield_dest" type="nametext" indexed="true" stored="true" multiValued="true"/>
+  <copyField source="copyfield_source" dest="copyfield_dest" />
+
+  <field name="copyfield_dest_not_stored" type="nametext" indexed="true" stored="false" multiValued="true"/>
+  <copyField source="copyfield_source" dest="copyfield_dest_not_stored" />
+
+  <field name="copyfield_source_multiple" type="string" indexed="true" stored="true" multiValued="true"/>
+  <field name="copyfield_dest_multiple" type="nametext" indexed="true" stored="true" multiValued="true"/>
+  <copyField source="copyfield_source" dest="copyfield_dest_multiple" />
+  <copyField source="copyfield_source_multiple" dest="copyfield_dest_multiple" />
+
   <!-- for versioning -->
   <field name="_version_" type="long" indexed="true" stored="true"/>
   <!-- points to the root document of a block of nested documents -->

solr/core/src/test/org/apache/solr/search/mlt/CloudMLTQParserTest.java

@@ -50,6 +50,9 @@ static void indexDocs()
     String id = "id";
     String FIELD1 = "lowerfilt_u";
     String FIELD2 = "lowerfilt1_u";
+    String FIELD3 = "copyfield_source";
+    String FIELD4 = "copyfield_source_not_stored";
+    String FIELD5 = "copyfield_source_multiple";
 
     new UpdateRequest()
         .add(sdoc(id, "1", FIELD1, "toyota"))
@@ -119,6 +122,13 @@ static void indexDocs()
                 "The slim red fox jumped over the lazy brown dogs.",
                 FIELD2,
                 "yellow white black"))
+        .add(
+            sdoc(
+                id, "33", FIELD3, "hard rock", FIELD4, "hard rock", FIELD5, "instrumental version"))
+        .add(
+            sdoc(
+                id, "34", FIELD3, "hard rock", FIELD4, "hard rock", FIELD5, "instrumental version"))
+        .add(sdoc(id, "35", FIELD3, "pop rock", FIELD4, "pop rock", FIELD5, "full version"))
         .commit(client, COLLECTION);
   }
 
@@ -339,4 +349,84 @@ public void testInvalidSourceDocument() {
                 .getSolrClient()
                 .query(COLLECTION, new SolrQuery("{!mlt qf=lowerfilt_u}999999")));
   }
+
+  @Test
+  public void testUsesACopyFieldInQf_shouldUseTheSourceFieldAndReturnResults() throws Exception {
+    // Verifies that when a copyField destination is used in the qf parameter, the MLT query
+    // correctly
+    // retrieves values from the source field(s) and returns relevant results.
+    QueryResponse queryResponse =
+        cluster
+            .getSolrClient()
+            .query(COLLECTION, new SolrQuery("{!mlt qf=copyfield_dest mindf=0 mintf=1}33"));
+    SolrDocumentList solrDocuments = queryResponse.getResults();
+    int[] expectedIds = new int[] {34, 35};
+    int[] actualIds = new int[solrDocuments.size()];
+    int i = 0;
+    for (SolrDocument solrDocument : solrDocuments) {
+      actualIds[i++] = Integer.parseInt(String.valueOf(solrDocument.getFieldValue("id")));
+    }
+
+    Arrays.sort(actualIds);
+    Arrays.sort(expectedIds);
+    assertArrayEquals(expectedIds, actualIds);
+  }
+
+  @Test
+  public void testCopyFieldSourceMissing_shouldReturnNoResults() throws Exception {
+    // Ensures that no results are returned when the copyField source field is missing in the source
+    // document.
+    QueryResponse queryResponse =
+        cluster
+            .getSolrClient()
+            .query(COLLECTION, new SolrQuery("{!mlt qf=copyfield_dest mindf=0 mintf=1}30"));
+    SolrDocumentList solrDocuments = queryResponse.getResults();
+    assertEquals("Expected no results if source field is missing", 0, solrDocuments.size());
+  }
+
+  @Test
+  public void testCopyFieldDestinationNotStored_shouldReturnResults() throws Exception {
+    // Even if the copyField destination field used in the MLT query (qf) is NOT stored, documents
+    // can still be returned, as long as its source field contains the text used
+    // to build the similarity query.
+    QueryResponse queryResponse =
+        cluster
+            .getSolrClient()
+            .query(
+                COLLECTION, new SolrQuery("{!mlt qf=copyfield_dest_not_stored mindf=0 mintf=1}33"));
+    SolrDocumentList solrDocuments = queryResponse.getResults();
+    int[] expectedIds = new int[] {34, 35};
+    int[] actualIds = new int[solrDocuments.size()];
+    int i = 0;
+    for (SolrDocument solrDocument : solrDocuments) {
+      actualIds[i++] = Integer.parseInt(String.valueOf(solrDocument.getFieldValue("id")));
+    }
+
+    Arrays.sort(actualIds);
+    Arrays.sort(expectedIds);
+    assertArrayEquals(expectedIds, actualIds);
+  }
+
+  @Test
+  public void testCopyFieldDestinationMultiple_shouldReturnResults() throws Exception {
+    // Validates that when multiple source fields map to a single copyField destination, their
+    // values are
+    // correctly combined to construct the MLT query, and appropriate results are returned.
+    QueryResponse queryResponse =
+        cluster
+            .getSolrClient()
+            .query(
+                COLLECTION, new SolrQuery("{!mlt qf=copyfield_dest_multiple mindf=0 mintf=1}33"));
+    SolrDocumentList solrDocuments = queryResponse.getResults();
+    int[] expectedIds = new int[] {34, 35};
+    int[] actualIds = new int[solrDocuments.size()];
+    int i = 0;
+    for (SolrDocument solrDocument : solrDocuments) {
+      actualIds[i++] = Integer.parseInt(String.valueOf(solrDocument.getFieldValue("id")));
+    }
+
+    Arrays.sort(actualIds);
+    Arrays.sort(expectedIds);
+    assertArrayEquals(expectedIds, actualIds);
+  }
 }