[jackrabbit-oak] branch trunk updated: OAK-9829: always add filter on FV field to prevent plugin failures (#613)

[fortino]

This is an automated email from the ASF dual-hosted git repository.

fortino pushed a commit to branch trunk
in repository https://gitbox.apache.org/repos/asf/jackrabbit-oak.git


The following commit(s) were added to refs/heads/trunk by this push:
     new 057a0de357 OAK-9829: always add filter on FV field to prevent plugin 
failures (#613)
057a0de357 is described below

commit 057a0de357003bac0f6d1e9515280b1d847b6b37
Author: Fabrizio Fortino <fabrizio.fort...@gmail.com>
AuthorDate: Sat Jul 9 07:59:08 2022 +0200

    OAK-9829: always add filter on FV field to prevent plugin failures (#613)
---
 .../oak/plugins/index/elastic/query/ElasticRequestHandler.java   | 9 ++++++---
 .../oak/plugins/index/elastic/ElasticSimilarQueryTest.java       | 9 ++++++++-
 2 files changed, 14 insertions(+), 4 deletions(-)

diff --git 
a/oak-search-elastic/src/main/java/org/apache/jackrabbit/oak/plugins/index/elastic/query/ElasticRequestHandler.java
 
b/oak-search-elastic/src/main/java/org/apache/jackrabbit/oak/plugins/index/elastic/query/ElasticRequestHandler.java
index 4f026ef314..41cee6ba11 100644
--- 
a/oak-search-elastic/src/main/java/org/apache/jackrabbit/oak/plugins/index/elastic/query/ElasticRequestHandler.java
+++ 
b/oak-search-elastic/src/main/java/org/apache/jackrabbit/oak/plugins/index/elastic/query/ElasticRequestHandler.java
@@ -317,9 +317,10 @@ public class ElasticRequestHandler {
                     continue;
                 }
 
+                String similarityPropFieldName = 
FieldNames.createSimilarityFieldName(pd.name);
                 String knnQuery = "{" +
                         "  \"elastiknn_nearest_neighbors\": {" +
-                        "    \"field\": \"" + 
FieldNames.createSimilarityFieldName(pd.name) + "\"," +
+                        "    \"field\": \"" + similarityPropFieldName + "\"," +
                         "    \"model\": \"" + 
pd.getSimilaritySearchParameters().getQueryModel() + "\"," +
                         "    \"similarity\": \"" + 
pd.getSimilaritySearchParameters().getQueryTimeSimilarityFunction() + "\"," +
                         "    \"candidates\": " + 
pd.getSimilaritySearchParameters().getCandidates() + "," +
@@ -332,8 +333,10 @@ public class ElasticRequestHandler {
                         "  }" +
                         "}";
 
-                query.should(s -> s
-                        .wrapper(w -> 
w.query(Base64.getEncoder().encodeToString(knnQuery.getBytes(StandardCharsets.UTF_8))))
+                query
+                        .filter(fb -> fb.exists(ef -> 
ef.field(similarityPropFieldName)))
+                        .should(s -> s
+                                .wrapper(w -> 
w.query(Base64.getEncoder().encodeToString(knnQuery.getBytes(StandardCharsets.UTF_8))))
                 );
             }
         }
diff --git 
a/oak-search-elastic/src/test/java/org/apache/jackrabbit/oak/plugins/index/elastic/ElasticSimilarQueryTest.java
 
b/oak-search-elastic/src/test/java/org/apache/jackrabbit/oak/plugins/index/elastic/ElasticSimilarQueryTest.java
index 0b7868254a..eaaa34d3b9 100644
--- 
a/oak-search-elastic/src/test/java/org/apache/jackrabbit/oak/plugins/index/elastic/ElasticSimilarQueryTest.java
+++ 
b/oak-search-elastic/src/test/java/org/apache/jackrabbit/oak/plugins/index/elastic/ElasticSimilarQueryTest.java
@@ -33,6 +33,7 @@ import java.io.File;
 import java.io.FileInputStream;
 import java.net.URI;
 import java.nio.charset.Charset;
+import java.nio.file.Files;
 import java.util.ArrayList;
 import java.util.Arrays;
 import java.util.Collection;
@@ -302,7 +303,7 @@ public class ElasticSimilarQueryTest extends 
ElasticAbstractQueryTest {
         File file = new File(uri);
 
         Collection<String> children = new LinkedList<>();
-        for (String line : IOUtils.readLines(new FileInputStream(file), 
Charset.defaultCharset())) {
+        for (String line : 
IOUtils.readLines(Files.newInputStream(file.toPath()), 
Charset.defaultCharset())) {
             String[] split = line.split(",");
             List<Double> values = 
Arrays.stream(split).skip(1).map(Double::parseDouble).collect(Collectors.toList());
             byte[] bytes = toByteArray(values);
@@ -315,6 +316,12 @@ public class ElasticSimilarQueryTest extends 
ElasticAbstractQueryTest {
             child.setProperty("fv", blob, Type.BINARY);
             children.add(child.getPath());
         }
+
+        // add a node without FV, the plugin cannot handle it directly
+        Tree child = test.addChild("nofv");
+        child.setProperty("nofv", "test");
+        children.add(child.getPath());
+
         root.commit();
 
         // check that similarity changes across different feature vectors