[OAK-9596] Full text search using Lucene index for binary content

• Type: Task
• Status: Open
• Priority: Major
• Resolution: Unresolved
• Component/s: indexing, lucene

Language

Java

• Assignee: Unassigned
• Reporter: Ankush Nagapure
• Created: 06/Oct/21 17:48
• Updated: 08/Nov/21 14:05

Description

I am trying out jackrabbit oak with lucene in a file node store. The index definition record is created successfully but it seems the index record is not created. lucene index creation code snippets:

 

public void initRepository() {
 LuceneIndexProvider provider = new LuceneIndexProvider();
 Jcr jcr = new Jcr(nodeStore)
 .withAsyncIndexing("async",3)
 .with(new LuceneIndexEditorProvider())
 .with((QueryIndexProvider) provider)
 .with((Observer) provider)
 .withAsyncIndexing("async",3);
 repository = jcr.createRepository();
 log.info("Repository initialized");
 }
public void createLuceneIndex() throws RepositoryException {
 Session session = createAdminSession();
 Node lucene = JcrUtils.getOrCreateByPath("/oak:index/lucene", "oak:Unstructured",
 "oak:QueryIndexDefinition", session, false);
 lucene.setProperty("compatVersion", 2);
 lucene.setProperty("type", "lucene");
 lucene.setProperty("async", "async");
 Node rules = lucene.addNode("indexRules", "nt:unstructured");
 Node allProps = rules.addNode("nt:base")
 .addNode("properties", "nt:unstructured")
 .addNode("allProps", "oak:Unstructured");
 allProps.setProperty(Property.JCR_DATA, ".*");
 allProps.setProperty("isRegexp", true);
 allProps.setProperty("nodeScopeIndex", true);
 session.save();
 session.logout();
 log.info("Lucene index created");
 }

 

After creating Lucene index, I have uploaded test.doc file in node store using below code:

 

log.info("Setting the JCR content for file name: test.doc, under path: " + folderNode.getPath());
final Binary binary = new BinaryImpl(fileBytes);
final Node content = folderNode.addNode(Property.JCR_CONTENT, NodeType.NT_RESOURCE);
content.setProperty(Property.JCR_DATA, binary);
//JCR session save code here

 

test.doc file contents:

HelloWorld, Test file contents for Full text search using Lucene index.

 

I have used below query to fetch result:

 

final Query query = queryManager.createQuery(
         "select * from [nt:base] where contains(*, 'HelloWorld')",
         Query.JCR_SQL2);
final QueryResult result = query.execute();
final NodeIterator nodeIter = result.getNodes();
log.info("Number of nodes: " + nodeIter.getSize());

But this query is not returning node where file contents are stored. I am getting below result:

Number of nodes: 0

Logs:

"Traversal query (query without index): select * from [nt:base] where contains(*,'HelloWorld'); consider creating an index"

Could you please let me know where I am making mistake or how to create proper index and do full text search on binary contents.