├── README.txt
├── pom.xml
└── src
    ├── main
        └── java
        │   └── com
        │       └── jayway
        │           └── mongodb
        │               ├── AnalyzedDBObject.java
        │               └── MongoDatastore.java
    └── test
        └── java
            └── com
                └── jayway
                    └── mongodb
                        └── MongoAnalysisTest.java


/README.txt:
--------------------------------------------------------------------------------
1 | Shows how to use mongo with lucene analyzers.
2 | 
3 | http://blog.jayway.com/2010/11/14/full-text-search-with-mongodb-and-lucene-analyzers/


--------------------------------------------------------------------------------
/pom.xml:
--------------------------------------------------------------------------------
 1 | <project xmlns="http://maven.apache.org/POM/4.0.0" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
 2 | 	xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/xsd/maven-4.0.0.xsd">
 3 | 	<modelVersion>4.0.0</modelVersion>
 4 | 
 5 | 	<groupId>com.jayway</groupId>
 6 | 	<artifactId>mongo-text-search</artifactId>
 7 | 	<version>1.0-SNAPSHOT</version>
 8 | 	<packaging>jar</packaging>
 9 | 
10 | 	<name>mongo-vs-lucene</name>
11 | 	<url>http://maven.apache.org</url>
12 | 
13 | 	<properties>
14 | 		<project.build.sourceEncoding>UTF-8</project.build.sourceEncoding>
15 | 	</properties>
16 | 
17 | 	<dependencies>
18 | 		<dependency>
19 | 			<groupId>junit</groupId>
20 | 			<artifactId>junit</artifactId>
21 | 			<version>4.8.1</version>
22 | 			<scope>test</scope>
23 | 		</dependency>
24 | 		<dependency>
25 | 			<groupId>org.mongodb</groupId>
26 | 			<artifactId>mongo-java-driver</artifactId>
27 | 			<version>2.3</version>
28 | 		</dependency>
29 | 		<dependency>
30 | 			<groupId>org.apache.lucene</groupId>
31 | 			<artifactId>lucene-core</artifactId>
32 | 			<version>3.0.1</version>
33 | 		</dependency>
34 | 
35 | 	</dependencies>
36 | 	<build>
37 | 		<plugins>
38 | 			<plugin>
39 | 				<artifactId>maven-compiler-plugin</artifactId>
40 | 				<configuration>
41 | 					<compilerVersion>1.5</compilerVersion>
42 | 					<fork>true</fork>
43 | 					<source>1.6</source>
44 | 					<target>1.6</target>
45 | 				</configuration>
46 | 			</plugin>
47 | 		</plugins>
48 | 	</build>
49 | </project>
50 | 


--------------------------------------------------------------------------------
/src/main/java/com/jayway/mongodb/AnalyzedDBObject.java:
--------------------------------------------------------------------------------
 1 | package com.jayway.mongodb;
 2 | 
 3 | import java.io.IOException;
 4 | import java.io.StringReader;
 5 | import java.util.ArrayList;
 6 | import java.util.List;
 7 | 
 8 | import org.apache.lucene.analysis.Analyzer;
 9 | import org.apache.lucene.analysis.TokenStream;
10 | import org.apache.lucene.analysis.tokenattributes.TermAttribute;
11 | 
12 | import com.mongodb.BasicDBObject;
13 | 
14 | public class AnalyzedDBObject extends BasicDBObject {
15 | 
16 | 	private static final long serialVersionUID = -3375115283230844689L;
17 | 
18 | 	public static enum Condition {ALL,IN}
19 | 	
20 | 	private Analyzer analyzer;
21 | 
22 | 	public AnalyzedDBObject(Analyzer analyzer) {
23 | 		this.analyzer = analyzer;
24 | 	}
25 | 
26 | 	public AnalyzedDBObject createQuery(String name,String text) throws IOException {
27 | 		return createQuery(name, text, Condition.ALL);
28 | 	}
29 | 	
30 | 	public AnalyzedDBObject createQuery(String name,String text,Condition condition) throws IOException {
31 | 		List<String> tokens = tokenize(analyzer.tokenStream(name, new StringReader(text)));
32 | 
33 | 		append(name,new BasicDBObject(
34 | 				String.format("$%s",condition.toString().toLowerCase()),
35 | 				tokens.toArray(new String[0])));
36 | 		return this;
37 | 	}
38 | 	
39 | 	public AnalyzedDBObject appendAndAnalyzeFullText(String name, String text)
40 | 			throws IOException {
41 | 		append(name,
42 | 				tokenize(analyzer.tokenStream(name, new StringReader(text))));
43 | 		return this;
44 | 	}
45 | 
46 | 	private List<String> tokenize(TokenStream stream) throws IOException {
47 | 		List<String> tokens = new ArrayList<String>();
48 | 		TermAttribute term = (TermAttribute) stream
49 | 				.addAttribute(TermAttribute.class);
50 | 		while (stream.incrementToken()) {
51 | 			// Not sure if we somehow can use termBuffer() to get a char[]
52 | 			// so we do no have to create a new String for each term
53 | 			tokens.add(term.term());
54 | 		}
55 | 		return tokens;
56 | 	}
57 | 
58 | }
59 | 


--------------------------------------------------------------------------------
/src/main/java/com/jayway/mongodb/MongoDatastore.java:
--------------------------------------------------------------------------------
 1 | package com.jayway.mongodb;
 2 | 
 3 | import java.net.UnknownHostException;
 4 | import java.util.ArrayList;
 5 | import java.util.List;
 6 | 
 7 | import com.mongodb.BasicDBObject;
 8 | import com.mongodb.DB;
 9 | import com.mongodb.DBCollection;
10 | import com.mongodb.DBCursor;
11 | import com.mongodb.DBObject;
12 | import com.mongodb.Mongo;
13 | import com.mongodb.MongoException;
14 | 
15 | public class MongoDatastore {
16 | 
17 | 	DB db;
18 | 	
19 | 	public MongoDatastore() throws UnknownHostException, MongoException {
20 | 		Mongo m = new Mongo( "localhost" , 27017 );
21 | 		db = m.getDB( "mongotextsearch" );
22 | 	}
23 | 	
24 | 	private DBCollection getCollection(String name) {
25 | 		return db.getCollection(name);
26 | 	}
27 | 	
28 | 	public void dropDatabase() {
29 | 		db.dropDatabase();
30 | 	}
31 | 	
32 | 	public void ensureIndex(String col, String field) {
33 | 		getCollection(col).createIndex(new BasicDBObject(field, new Integer(1)));
34 | 	}
35 | 	
36 | 	public void save(DBObject object,String col) {
37 | 		getCollection(col).save(object);
38 | 	}
39 | 	
40 | 	public DBObject get(DBObject query,String col) {
41 | 		return getCollection(col).findOne(query);
42 | 	}
43 | 	
44 | 	public List<DBObject> find(DBObject query,String col) {
45 | 		
46 | 		DBCursor cursor = getCollection(col).find(query);
47 | 		
48 | 		List<DBObject> dbObjects = new ArrayList<DBObject>();
49 | 		while(cursor.hasNext()) {
50 | 			dbObjects.add(cursor.next());
51 | 		}
52 | 		return dbObjects;
53 | 	}
54 | 	
55 | 	public long cnt(String col) {
56 | 		return getCollection(col).count();
57 | 	}
58 | 	
59 | }
60 | 


--------------------------------------------------------------------------------
/src/test/java/com/jayway/mongodb/MongoAnalysisTest.java:
--------------------------------------------------------------------------------
 1 | package com.jayway.mongodb;
 2 | 
 3 | import static junit.framework.Assert.assertEquals;
 4 | import static junit.framework.Assert.assertNull;
 5 | 
 6 | import java.io.IOException;
 7 | import java.net.UnknownHostException;
 8 | 
 9 | import org.apache.lucene.analysis.Analyzer;
10 | import org.apache.lucene.analysis.standard.StandardAnalyzer;
11 | import org.apache.lucene.util.Version;
12 | import org.junit.BeforeClass;
13 | import org.junit.Test;
14 | 
15 | import com.jayway.mongodb.AnalyzedDBObject.Condition;
16 | import com.mongodb.DBObject;
17 | import com.mongodb.MongoException;
18 | 
19 | public class MongoAnalysisTest {
20 | 
21 | 	
22 | 	static final String TEXT = "I Would like to use mongodb for full text search";
23 | 	static final String COLLECTION_NAME = "article";
24 | 	static final String INDEXED_FIELD = "indtext";
25 | 	static final String TEXT_FIELD = "text";
26 | 	static MongoDatastore mongo;
27 | 	
28 | 	static Analyzer analyzer = new StandardAnalyzer(Version.LUCENE_30);
29 | 	
30 | 	@BeforeClass
31 | 	public static void setup() throws UnknownHostException, MongoException {
32 | 		mongo = new MongoDatastore();
33 | 		mongo.dropDatabase();
34 | 	}
35 | 	
36 | 	@Test
37 | 	public void verifySaveAndGetTestSearchWithAll() throws IOException {
38 | 
39 | 		mongo.save(
40 | 				new AnalyzedDBObject(analyzer)
41 | 				.appendAndAnalyzeFullText(INDEXED_FIELD, TEXT)
42 | 				.append(TEXT_FIELD, TEXT),
43 | 				COLLECTION_NAME);
44 | 		
45 | 		DBObject result = mongo.get(new AnalyzedDBObject(analyzer).createQuery(INDEXED_FIELD,TEXT),COLLECTION_NAME);
46 | 		assertEquals(TEXT, result.get(TEXT_FIELD));
47 | 		
48 | 		result = mongo.get(new AnalyzedDBObject(analyzer).createQuery(INDEXED_FIELD,"MonGoDB sEarch woulD TO"),COLLECTION_NAME);
49 | 		assertEquals(TEXT, result.get(TEXT_FIELD));
50 | 		
51 | 		// In this query, only "search" matches
52 | 		result = mongo.get(new AnalyzedDBObject(analyzer).createQuery(INDEXED_FIELD,"MonGoDBs wouldd search "),COLLECTION_NAME);
53 | 		assertNull(result);
54 | 	}
55 | 	
56 | 	@Test
57 | 	public void verifySaveAndGetTestSearchWithAtLeastOne() throws IOException {
58 | 
59 | 		mongo.save(
60 | 				new AnalyzedDBObject(analyzer)
61 | 				.appendAndAnalyzeFullText(INDEXED_FIELD, TEXT)
62 | 				.append(TEXT_FIELD, TEXT),
63 | 				COLLECTION_NAME);
64 | 		
65 | 		DBObject result = mongo.get(new AnalyzedDBObject(analyzer).createQuery(INDEXED_FIELD,TEXT,Condition.IN),COLLECTION_NAME);
66 | 		assertEquals(TEXT, result.get(TEXT_FIELD));
67 | 		
68 | 		result = mongo.get(new AnalyzedDBObject(analyzer).createQuery(INDEXED_FIELD,"MonGoDB sEarch woulD",Condition.IN),COLLECTION_NAME);
69 | 		assertEquals(TEXT, result.get(TEXT_FIELD));
70 | 		
71 | 		// In this query, only "search" matches
72 | 		result = mongo.get(new AnalyzedDBObject(analyzer).createQuery(INDEXED_FIELD,"MonGoDBs wouldd search ",Condition.IN),COLLECTION_NAME);
73 | 		assertEquals(TEXT, result.get(TEXT_FIELD));
74 | 	}
75 | }
76 | 


--------------------------------------------------------------------------------