tempbottle / spring-data-solr

Spring Data - Solr integration

Spring Data Solr

The primary goal of the Spring Data project is to make it easier to build Spring-powered applications that use new data access technologies such as non-relational databases, map-reduce frameworks, and cloud based data services.

The Spring Data Solr project provides integration with the Apache Solr search engine

Providing its own extensible MappingSolrConverter as alternative to DocumentObjectBinder Spring Data Solr handles inheritance as well as usage of custom Types such as Point or DateTime.

Getting Help

• Reference Documentation
• API Documentation
• Spring Data Project
• Issues
• Code Analysis
• Questions

If you are new to Spring as well as to Spring Data, look for information about Spring projects.

Quick Start

SolrTemplate

SolrTemplate is the central support class for solr operations.

SolrRepository

A default implementation of SolrRepository, aligning to the generic Repository Interfaces, is provided. Spring can do the Repository implementation for you depending on method names in the interface definition.

The SolrCrudRepository extends PagingAndSortingRepository

   public interface SolrCrudRepository<T, ID extends Serializable> extends SolrRepository<T, ID>, PagingAndSortingRepository<T, ID> {
   } 

The SimpleSolrRepository implementation uses MappingSolrConverter. In order support native solrj mapping via DocumentObjectBinder fields have to be annotated with org.apache.solr.client.solrj.beans.Field. org.springframework.data.solr.core.mapping.Indexed can be used as substitute for Field offering additional attributes to be used eg. for index time boosting.

To enable native solrj mapping use SolrJConverter along with SolrTemplate.

public interface SolrProductRepository extends SolrCrudRepository<Product, String> {
  
  //Derived Query will be "q=popularity:<popularity>&start=<page.number>&rows=<page.size>"
  Page<Product> findByPopularity(Integer popularity, Pageable page);
  
  //Will execute count prior to determine total number of elements
  //Derived Query will be "q=name:<name>*&start=0&rows=<result of count query for q=name:<name>>"
  List<Product> findByNameStartingWith(String name);
  
  //Derived Query will be "q=inStock:true&start=<page.number>&rows=<page.size>"
  Page<Product> findByAvailableTrue(Pageable page);
  
  //Derived Query will be "q=inStock:<inStock>&start=<page.number>&rows=<page.size>"
  @Query("inStock:?0")
  Page<Product> findByAvailableUsingAnnotatedQuery(boolean inStock, Pageable page);
  
  //Will execute count prior to determine total number of elements
  //Derived Query will be "q=inStock:false&start=0&rows=<result of count query for q=inStock:false>&sort=name desc"
  List<Product> findByAvailableFalseOrderByNameDesc();
  
  //Execute faceted search 
  //Query will be "q=name:<name>&facet=true&facet.field=cat&facet.limit=20&start=<page.number>&rows=<page.size>"
  @Query(value = "name:?0")
  @Facet(fields = { "cat" }, limit=20)
  FacetPage<Product> findByNameAndFacetOnCategory(String name, Pageable page);
  
  //Boosting criteria
  //Query will be "q=name:<name>^2 OR description:<description>&start=<page.number>&rows=<page.size>"
  Page<Product> findByNameOrDescription(@Boost(2) String name, String description, Pageable page);
  
  //Highlighting results
  //Query will be "q=name:(<name...>)&hl=true&hl.fl=*"
  @Highlight
  HighlightPage<Product> findByNameIn(Collection<String> name, Pageable page);
  
  //Spatial Search
  //Query will be "q=location:[<bbox.start.latitude>,<bbox.start.longitude> TO <bbox.end.latitude>,<bbox.end.longitude>]"
  Page<Product> findByLocationNear(Box bbox);
  
  //Spatial Search
  //Query will be "q={!geofilt pt=<location.latitude>,<location.longitude> sfield=location d=<distance.value>}"
  Page<Product> findByLocationWithin(Point location, Distance distance);
  
}

Furthermore you may provide a custom implementation for some operations.

public interface SolrProductRepository extends SolrCrudRepository<Product, String>, SolrProductRepositoryCustom {
  
  @Query(fields = { "id", "name", "popularity" })
  Page<Product> findByPopularity(Integer popularity, Pageable page);
  
  List<Product> findByAuthorLike(String author);
  
}

public interface SolrProductRepositoryCustom {
  
  Page<Product> findProductsByCustomImplementation(String value, Pageable page)
  
}

public class SolrProductRepositoryImpl implements SolrProductRepositoryCustom {
  
  private SolrOperations solrTemplate;
  
  @Override
  public Page<Product> findProductsByCustomImplementation(String value, Pageable page) {
    Query query = new SimpleQuery(new SimpleStringCriteria("name:"+value)).setPageRequest(page);
    return solrTemplate.queryForPage(query, Product.class);
  }
  
  @Autowired
  public void setOperations(SolrOperations operations) {
    this.operations = operations;
  }
  
}

Go on and use it as shown below:

@Configuration
@EnableSolrRepositories(basePackages = { "com.acme.sorl" }), multicoreSupport = true)
public class SolrContext {
  
  private @Resource Environment env;

  @Bean
  public SolrServer solrServer() throws MalformedURLException, IllegalStateException {
    return new HttpSolrServer(env.getRequiredProperty("solr.host"));
  }

}

@Service
public class ProductService {
  
  private SolrProductRepository repository;

  @Autowired
  public ProductService(SolrProductRepository repository) {
    this.repository = repository;
  }
  
  public void doSomething() {
    repository.deleteAll();
    
    Product product = new Product("spring-data-solr");
    product.setAuthor("Christoph Strobl");
    product.setCategory("search");
    repository.save(product);
    
    Product singleProduct = repository.findById("spring-data-solr");
    List<Product> productList = repository.findByAuthorLike("Chr");
  }
  
}

XML Namespace

You can set up repository scanning via xml configuration, which will happily create your repositories.

<?xml version="1.0" encoding="UTF-8"?>
<beans xmlns="http://www.springframework.org/schema/beans"
  xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
  xmlns:solr="http://www.springframework.org/schema/data/solr"
  xsi:schemaLocation="http://www.springframework.org/schema/data/solr http://www.springframework.org/schema/data/solr/spring-solr.xsd
    http://www.springframework.org/schema/beans http://www.springframework.org/schema/beans/spring-beans.xsd">
  
  <solr:repositories base-package="com.acme.repository" multicoreSupport="true" />
  <solr:solr-server id="solrServer" url="http://localhost:8983/solr" />
  
</beans>

Automatic Schema Population

Automatic schema population will inspect your domain types whenever the applications context is refreshed and populate new fields to your index based on the properties configuration. This requires solr to run in Schemaless Mode.

Use @Indexed to provide additional details like specific solr types to use.

@Configuration
@EnableSolrRepositories(schemaCreationSupport = true, multicoreSupport = true)
class Config {

  @Bean
  public SolrServer solrServer() {
    return new HttpSolrServer("http://localhost:8983/solr");
  }
}

@Document(coreName="collection1")
class Product {
  
  @Id String id;
  @Indexed(solrType="text_general") String author;
  @Indexed("cat") List<String> category;

}

// curl ../solr/collection1/schema/fields -X POST -H 'Content-type:application/json'
[
  {
    "name":"id",
    "type":"string",
    "stored":true,
    "indexed":true,
    "multiValued":false
  }
  {
    "name":"author",
    "type":"text_general",
    "stored":true,
    "indexed":true,
    "multiValued":false
  }
  {
    "name":"cat",
    "type":"string",
    "stored":true,
    "indexed":true,
    "multiValued":true
  }
]

Maven

RELEASE

<dependency>
  <groupId>org.springframework.data</groupId>
  <artifactId>spring-data-solr</artifactId>
  <version>1.2.0.RELEASE</version>
</dependency>  

Build Snapshot

<dependency>
  <groupId>org.springframework.data</groupId>
  <artifactId>spring-data-solr</artifactId>
  <version>1.3.0.BUILD-SNAPSHOT</version>
</dependency> 

<repository>
  <id>spring-maven-snapshot</id>
  <url>http://repo.spring.io/libs-snapshot</url>
</repository>  

Contributing to Spring Data

Please refer to CONTRIBUTING

Stay in touch

Follow the project team (@stroblchristoph, @SpringData) on Twitter. Releases are announced via our news feed.