Indexing Framework

Unless you’re searching for model entities using database queries (not recommended in most cases), each asset in Liferay DXP must be indexed in the search engine. The indexing code is specific to each asset, as the asset’s developers know what fields to index and what filters to apply to the search query.

In past versions of Liferay DXP, when your asset required indexing, you would implement a new Indexer by extending<T>. Liferay DXP 7.1 introduces a new pattern that relies on composition instead of inheritance. If you want to use the old approach, feel free to extend BaseIndexer. It’s still supported.

Search and Indexing Overview

Liferay’s original Search API was built around the Lucene search and indexing library. To this day, familiarity with Lucene will jump-start your understanding of Liferay’s Search API. However, starting with the 7.0 version of Liferay DXP, the Search API is being reworked, so that the parts closely tied to Lucene are becoming more generic. Elasticsearch support was added (in addition to Solr), and most of the newer searching and indexing APIs aim to leverage/map Elasticsearch APIs. This means that in many cases (for example the Query types) there is a one-to-one mapping between the Liferay and Elasticsearch APIs.

In addition to the Elasticsearch centered APIs, Liferay’s Search Infrastructure includes additional APIs serving these purposes:

  • Indexed documents include the fields needed by Liferay DXP (e.g., entryClassName, entryClassPK, assetTagNames, assetCategories, companyId, groupId, staging status).
  • It ensures the scope of returned search results is appropriate by applying the right filters to search requests.
  • It provides permission checking and hit summaries to display in the search portlet.

To understand how search and indexing code makes your custom models seamlessly searchable, you must know how to influence each portion of the search and indexing cycle.


Model entities store data fields in the database. For example, Guestbooks store a name field. During the cycle’s Indexing step, you prepare the model entity to be searchable by defining the model’s fields that are sent to the search engine, later used during a search.

To control how model entity fields are indexed in search engine documents,

ModelDocumentContributor classes specify which database fields are indexed in the model entity’s search engine document. This class’s contribute method is called each time the add and update methods in the entity’s service layer are called.

ModelIndexerWriterContributor classes configure the re-indexing and batch re-indexing behavior for the model entity. This class’s method is called when a re-index is triggered from the Search administrative application found in Control Panel → Configuration → Search, or when a CRUD operation is made on the entity, if the modelIndexed method is implemented in your contributor.

DocumentContributor classes (without any component property or type parameter) contribute certain fields to every index document, regardless of what base entity is being indexed. For example, the GroupedModelDocumentContributor contributes the GROUP_ID and SCOPE_GROUP_ID fields for all documents with a backing entity that’s also a GroupedModel.


Searches start with a user entering keywords into a search bar. The entered keywords are processed by the back-end search infrastructure, transformed into a query the search engine can understand, and used to match against each search document’s fields.

To control the way model entity documents are searched,

KeywordQueryContributor classes contribute clauses to the ongoing search query. This is called at search time and ensures that all the fields you indexed are also the ones searched. For example, if you index fields with the Site locale appended to them (title_en_us, for example), you want the search query to include the same locale when the document is searched. If the search query contains another locale (like title_es_ES) or searches the plain field (title), inaccurate results are returned.

To contribute query clauses to every search, regardless of what base entity is being searched, implement a KeywordQueryContributor class registered without an component property. For example, see AlwaysPresentFieldsKeywordQueryContributor.

ModelPreFilterContributors control how search results are filtered before they’re returned from the search engine. For example, adding the workflow status to the query ensures that an entity in the trash isn’t returned in the search results.

To contribute Filter clauses to every search, regardless of what base entity is being searched, implement a QueryPreFilterContributor. QueryPreFilterContributor is constructed only once under the root filter during a search. For example, see AssetCategoryTitlesKeywordQueryContributor.

What’s the difference between QueryPreFilterContributor and ModelPreFilterContributor? QueryPreFilterContributor is constructed only once under the root filter during a search, while ModelPreFilterContributor is constructed once per model entity, and added under each specific entity’s subfilter.

Returning Results

When a model entity’s indexed search document is obtained during a search request, it’s converted into a summary of the model entity.

To control the result summaries for model entity documents,

ModelSummaryContributor classes get the Summary object created for each search document, so you can manipulate it by adding specific fields or setting the length of the displayed content.

ModelVisibilityContributor classes control the visibility of model entities that can be attached to other asset types (for example, File Entries can be attached to Wiki Pages), in the search context.

One important step must occur to make sure the above classes are discovered by the search framework.

Search Service Registration

To register model entities with Liferay’s search framework,

SearchRegistrars use the search framework’s registry to define certain things about your model entity’s ModelSearchDefinition: which fields are used by default to retrieve documents from the search engine, and which optional search services are registered for your entity. Registration occurs as soon as the Component is activated (during portal startup or deployment of the bundle).

Mapping the Composite Search and Indexing Framework to Indexer/BaseIndexer Code

If you’re used to the old way of indexing custom entities (extending BaseIndexer, the abstract implementation of Indexer), the table below provides a quick overview about how the methods of the Indexer interface were decomposed into several new classes and methods.

Indexer/BaseIndexer methodComposite Indexer EquivalentExample
Class ConstructorSearchRegistrarGuestbookSearchRegistrar
getSummary/createSummary/doGetSummaryModelSummaryContributorGuestbookModelSummaryContributor changeGuestbook view_search.jsp
Indexer.delete/doDeleteNo changeMBMessageLocalServiceImpl.deleteDiscussionMessages

In addition, you can index ExpandoBridge attributes. This was previously accomplished in BaseIndexer’s getBaseModelDocument. Now you implement an ExpandoBridgeRetriever. See DLFileEntryExpandoBridgeRetriever for an example implementation.

Permissions Aware Searching and Indexing

In previous versions of Liferay DXP, search was only permissions aware (indexed with the entity’s permissions and searched with those permissions intact) if the application developer specified this line in the Indexer class’s constructor:


Now, search is permissions aware by default if the new permissions approach, as described in these tutorials, is implemented for an application.

Annotating Service Methods to Trigger Indexing

Having entities translated into database entities and search engine documents means that there’s a possibility for a state mismatch between the database and search engine. For example, when a Blogs entry is added, updated or removed from the database, corresponding changes must be made in the search engine. To do this, intervention must be made into the service layer. For Service Builder entities, this occurs in the LocalServiceImpl classes. There’s an annotation that simplifies this: @Indexable. It takes a type property that can have two values: REINDEX or DELETE. Commonly, a deleteEntity method in the service layer is annotated like this:

@Indexable(type = IndexableType.DELETE)
@SystemEvent(type = SystemEventConstants.TYPE_DELETE)
public BlogsEntry deleteEntry(BlogsEntry entry) throws PortalException {

The @Indexable annotation is executed by Liferay’s Spring infrastructure, so if you have a method with that annotation, you must call it using a service instance variable with the spring wrapped logic. The reference is available by default in Service Builder generated *LocalServiceImpl classes because of this declaration in the parent *LocalServiceBaseImpl:

@BeanReference(type = BlogsEntryLocalService.class)
    protected BlogsEntryLocalService blogsEntryLocalService;

This means that in the *LocalServiceImpl, you must not call


The annotation won’t be executed and you’ll be left with a state mismatch between the search engine document and the database column. Instead follow the pattern in Liferay DXP’s code, using the service instance variable to call service methods:


For step-by-step instructions on indexing model entities, visit the Search and Indexing section of the Developing a Web Application tutorials.

« Elasticsearch LoggingIntroduction to Asset Framework »
¿Fue útil este artículo?
Usuarios a los que les pareció útil: 1 de 1