SOLR-15531: Cross-collection join fixed to ignore documents that do n…

…ot contain value in from field
MarcusSorealheis · Jul 28, 2021 · e7e80b1 · e7e80b1
1 parent 5e8857c
commit e7e80b1
Show file tree

Hide file tree

Showing 3 changed files with 15 additions and 3 deletions.
diff --git a/solr/CHANGES.txt b/solr/CHANGES.txt
@@ -394,6 +394,8 @@ Bug Fixes
 
 * SOLR-15531: SignificantTerms streaming function should not fail on empty collections (Benedikt Arnold via Mike Drob)
 
+* SOLR-15531: Cross-collection join fixed to ignore documents that do not contain value in "from" field (hossman)
+
 Other Changes
 ---------------------
 (No changes)

diff --git a/solr/core/src/java/org/apache/solr/search/join/CrossCollectionJoinQuery.java b/solr/core/src/java/org/apache/solr/search/join/CrossCollectionJoinQuery.java
@@ -292,7 +292,9 @@ private DocSet getDocSet() throws IOException {
           }
 
           Object value = tuple.get(fromField);
-          collector.collect(value);
+          if (null != value) {
+            collector.collect(value);
+          }
         }
       } catch (IOException e) {
         throw new SolrException(SolrException.ErrorCode.SERVER_ERROR, e);

diff --git a/solr/core/src/test/org/apache/solr/search/join/CrossCollectionJoinQueryTest.java b/solr/core/src/test/org/apache/solr/search/join/CrossCollectionJoinQueryTest.java
@@ -20,6 +20,7 @@
 import java.io.IOException;
 import java.util.ArrayList;
 import java.util.Collection;
+import java.util.Collections;
 import java.util.List;
 import java.util.Locale;
 
@@ -68,8 +69,6 @@ public static void setupIndexes(boolean routeByKey) throws IOException, SolrServ
 
     buildIndexes(routeByKey);
 
-    assertResultCount("products", "*:*", NUM_PRODUCTS, true);
-    assertResultCount("parts", "*:*", NUM_PRODUCTS * 10 / 4, true);
   }
 
   private static void clearCollection(String collection) throws IOException, SolrServerException {
@@ -104,11 +103,20 @@ private static void buildIndexes(boolean routeByKey) throws IOException, SolrSer
       }
     }
 
+    // some extra docs in each collection (not counded in NUM_PRODUCTS) that should drop out of the joins because they don't have the join key
+    productDocs.add(new SolrInputDocument("id", buildId(NUM_PRODUCTS+10, String.valueOf(NUM_PRODUCTS+10), routeByKey), "size_s", "M"));
+    partDocs.add(new SolrInputDocument("id", buildId(NUM_PRODUCTS+10, String.valueOf(NUM_PRODUCTS+10), routeByKey)));
+
+    Collections.shuffle(productDocs, random());
+    Collections.shuffle(partDocs, random());
+
     indexDocs("products", productDocs);
     cluster.getSolrClient().commit("products");
+    assertResultCount("products", "*:*", 1 + NUM_PRODUCTS, true);
 
     indexDocs("parts", partDocs);
     cluster.getSolrClient().commit("parts");
+    assertResultCount("parts", "*:*", 1 + (NUM_PRODUCTS * 10 / 4), true);
   }
 
   private static String buildId(int productId, String id, boolean routeByKey) {