Spring Data Couchbase 2.0 is a rewrite of the original Spring Data Couchbase 1.4.x connector. It is based on the Couchbase Java 2.2 SDK and makes heavy use of the new query language N1QL (which was introduced in Couchbase Server 4.0) to offer more features for Spring Data users.

The first Milestone has been released last august, then a Release Candidate followed, and since then additional features (and bugfixes) were implemented a GA release can now be unleashed on the public.

Let's take a quick tour of what changed (with a ⭐ to ⭐⭐⭐ notation of how awesome and significant we think each feature is ?):

New Features in Spring Data Couchbase 2.0

The main differences between the 1.x generation of Spring Data Couchbase and its 2.x version are:

  • Configuration elements are closer to the Couchbase reality: Environment, Cluster, Bucket (potentially allowing you to create CouchbaseTemplates that each connect to a different bucket, or even to different clusters!)
  • Backing custom repository methods is not always done with views anymore, it is (by default) now done via N1QL, which is much more flexible and requires less server-side maintenance.
  • Custom methods using views have been modified a little to better stick to the Spring Data philosophy. This reduces flexibility a little bit, but the implementations are generated from the method name (through “query derivation”).
  • You can now do geospatial querying of your data (or multidimensional querying if you go beyond 3 dimensions) with views.

Of course you can still access the lower level API by using the CouchbaseTemplate rather than the CouchbaseRepository interface, and you can even access the underlying Bucket from the SDK.

Repository methods through N1QL


The big new feature in Couchbase 4.0 is N1QL, a SQL extension that works on JSON documents (so it added JSON-related specificities to SQL).

This is especially great for the Repository pattern and query derivation in Spring Data, because the vast majority of query derivation keywords can be easily translated to N1QL.

N1QL is now the default backing Couchbase feature for Repository methods. You can also elect to use the @Query interface if you want to be explicit on the query executed.

Repository methods through Views


One big change in this version is that now, Repository Queries (aka custom repository methods) that are based on views are more in line with the Spring Data philosophy. They also have to be annotated explicitly with @View(viewName="something").

This means that nothing Couchbase-specific should leak into your repository interface. Instead, what you can do is use query derivation mechanisms for most of the queries.

Query derivation is also possible to a small extent, with a few keywords being accepted in a view-backed method.

Using reduce function from Views

Another new thing that wasn't previously supported is the execution of the reduce function if you have one. Now, in order to execute it, you simply set the reduce flag to true in the @View annotation.

You could also prefix your method with “count” instead of “find” if that is meaningful to you (ie. you actually use the “count” reduce function).

Note that the reduce function in Couchbase can be something else than the preexisting _count one, and could even return something else than a long like a JsonObject, like for built-in _stats.

Similarly, adding the variation “topX” or “firstX” in a method name will result in an additional limit being set on the request (eg. findFirst5ByLastName will limit the list to 5 results).

Configuring consistency, Read Your Own Writes


One thing that comes up often when using asynchronously populated secondary indexes like views and GSI (the new secondary index engine backing N1QL), is the need to immediately read the modifications from your previous write operations.

This implies that the view/N1QL shouldn't answer as long as the data is still in the process of being indexed, so this sacrifices some performance in favor of consistency.

The opposite (and current default for Spring Data Couchbase) is to favor performance by accepting stale data to be returned.

We added global semantics for configuring all queries (view-based or N1QL-based) that are constructed by the framework through query derivation, by providing a small abstraction around the concept of Consistency.

This is done by overriding the AbstractCouchbaseConfiguration's getDefaultConsistency() method. Consistency is an enum that lets you choose between READ_YOUR_OWN_WRITES, STRONGLY_CONSISTENT, UPDATE_AFTER and EVENTUALLY_CONSISTENT. Please refer to the official documentation for more information on how they work exactly and what their impact is at query time.

You can also do that in XML by using the consistency attribute on the tag.

Since GA, CRUD methods in repositories now also take the default configured consistency into account.

Changing the type information field in stored JSON

Some users have reported issues with Spring Data and the Couchbase Mobile side of things, with the Sync Gateway rejecting documents containing fields prefixed by an underscore.

This is problematic for Spring Data, since by default it stores the type information in a _class field 🙁

The solution is to allow, through the configuration, to modify the name of that type information field. You can do so by overriding the typeKey() method in AbstractCouchbaseConfiguration. For instance, you can use the constant MappingCouchbaseConverter.TYPEKEY_SYNCGATEWAY_COMPATIBLE (which is “javaClass“).

This field is the one used by generated N1QL queries to filter only documents corresponding to the repository's entity.

Support for Pageable/PageRequest in N1QL derived queries


Using N1QL, for queries that are generated through query derivation, Pageable and Sort parameters are now supported.

  • Support for PagingAndSortingRepository based on N1QL.
  • Adds two findAll methods that rely on N1QL for paging and/or sorting. Uses the default configured consistency.

GeoSpatial and Multi-Dimensional querying using Spatial Views


Query Couchbase using coordinates! Provided your entity has a Point (or x and y) location, you can find it using:

  • a bounding box: findByLocationWithin(Box area)
  • a circle: findByLocationWithin(Circle area), findByLocationWithin(Point center, Distance radius).
  • a polygon: findByLocationWithin(Polygon area), findByLocationWithin(Point[] polygon).
  • a distance findByLocationNear(Point near, Distance maxDistance).

Circle and polygon-like queries are performed fast as bounding box approximations on the server then the false positives are eliminated by the framework before presenting results.

You can take advantage of the multidimensional aspect of Couchbase Spatial Views to add extra dimensions to your queries (e.g. stores that open late at night within a city…).

Note: if you want to reuse annotations, you can do that too (works for @View and @Query too):

Inline N1QL @Query now have SpEL support


Inline queries can use SpEL notation to:

  1. ensure that correct selection and filtering is applied to the statement in order to construct and return entities: use #{#n1ql.selectEntity} to generate a SELECT ... FROM ... clause, and #{#n1ql.filter} in the WHERE clause to limit query to the correct entity.
  2. compute values or retrieve data from external SpEL value provides configured in the Spring context.

Repository “main” indexes creation can be automatically triggered


⚠️ IMPORTANT: this is considered as an aid during development/testing and discouraged in production

In order to make sure that N1QL indexing of the entities in a given repository is activated in a dev or pre-production environment, one can annotate it with @N1qlPrimaryIndexed (which enables bucket-wide freeform querying) and @N1qlSecondaryIndexed (which will index only the documents corresponding to the entity type, similarly to the WHERE clause produced by SpEL #{#n1ql.filter}).

Also, the backing view for CRUD operation can be automatically created by annotating the repository with @ViewIndexed (you'll need to provide the design document name, which should correspond to the entity's simple class name with a lowercase first letter).

This feature must additionally be opted-in by redefining the indexManager bean in the AbstractCouchbaseConfiguration.

Simple return types (primitives and String) are now supported when using a single-row projection


This is especially targeted at inline N1QL queries with aggregation functions like COUNT(*), AVG(field), etc… The query must return a single row with a single projection.

Support named parameters in N1QL inline queries


Use either named parameters or positional parameter, but not both. Syntax for named parameters is $paramName, requiring that each method parameter be annotated with @Param("paramName").

Other Features

Other features include:

  • Fix bean naming so that all beans created by Spring Data Couchbase are prefixed with “couchbase“, in order to avoid clashes with other stores.
  • Changing the base class for all repositories is now supported (following the process documented in the common Spring Data documentation)
  • In case indexes are stale, deleted documents are eliminated from find methods in the CouchbaseTemplate
  • Expiry can be set on a @Document, as a long + timeUnit

A few bugfixes and improvements over RC1 have also been implemented.



Documentation has also been improved, adding Couchbase-oriented examples on how to add the implementation of a custom method to a repository, how to change the base class of all repositories, how to deal with SpEL in inline queries, …

A note about Spring Cache

The Spring Cache support has been moved out of the Spring Data repository. It is still there and we plan on improving on it. You can for now find it on a Couchbase repository on github but it should soon reintegrate the official Spring family of projects.

Getting Spring Data Couchbase

You can add the following to your project's pom.xml to get this GA Release (in the dependencies section:

We hope you enjoy this release and all the new features it brings to the table. Next step will be to re-attach to the Hopper release train with a version 2.1 expected before Summer.

Posted by Simon Basle, Software Engineer, Pivotal

Simon Basl_ is a Paris-based Software Engineer working in the Spring team at Pivotal. Previously, he worked in the Couchbase Java SDK team. His interests span software design aspects (OOP, design patterns, software architecture), rich clients, what lies beyond code (continuous integration, (D)VCS, best practices), and reactive programming. He is also an editor for the French version of InfoQ.com.

Leave a reply