Skip to content
This repository


Subversion checkout URL

You can clone with HTTPS or Subversion.

Download ZIP

Extensible BSON encoder/decoder library written in Java with pluggable Java-to-BSON type mappings.

branch: master

Update BsonDocumentTest

I've found a potential bug in the default BsonDocument implementation
DefaultDocument. The default equals method (which is delegated to a
LinkedHashMap if my memory serves me right) compares arrays by reference.

I've added a test case for this which fails like it should. When I'll have time
I'll write a proper hashCode/equals for DefaultDocument.
latest commit e17522c375
Kohányi Róbert authored
Octocat-spinner-32 src Update BsonDocumentTest March 02, 2013
Octocat-spinner-32 .gitignore Initial commit. January 22, 2012
Octocat-spinner-32 LICENSE.txt Update POM (plugin versions, checkstyle rules). June 03, 2012
Octocat-spinner-32 Fix a minor typo in the Maven instructions May 14, 2012
Octocat-spinner-32 pom.xml Update pom.xml December 04, 2012


ebson is an extensible BSON encoder/decoder library written in Java. The library is extensible in the sense that the mappings between Java and BSON types are configurable and the logic to serialize custom Java types is pluggable. Its single dependency is the Guava libraries by Google.


Released under the permissive MIT License.


Kohányi Róbert.


Add the library as a dependency in your project's pom.xml like this.


Releases and snapshots are deployed to Sonatype's OSS repository (and synced to the Central Maven Repository from there). To download JARs from Sonatype's repository include the following repository tag inside your Maven installation's settings.xml or your project's pom.xml.



As the project is managed with Maven you simply clone it and issue mvn install or mvn package inside the clone's directory.

git clone git://
cd ebson/
mvn package
# and/or
mvn install



// create documents to serialize
BsonDocument document = BsonDocuments.of("key", new Date());

// grab a little-endian byte buffer
ByteBuffer buffer = ByteBuffer.allocate(32).order(ByteOrder.LITTLE_ENDIAN);

// use the documents utility class to write the document into the buffer
BsonDocuments.writeTo(buffer, document);

// use the serialized data


// given the previous buffer
BsonDocument newDocument = BsonDocuments.readFrom(buffer);

// prints true


// to use joda-time's date-time instead of java's date supply
// a predicate (to test whether an input class is compatible with
// date-time or not) for the appropriate bson type
BsonObject.UTC_DATE_TIME.predicate(new Predicate<Class<?>>() {
  @Override public boolean apply(Class<?> input) {
    return input == null ? false : DateTime.class.isAssignableFrom(input);

// register a writer with the same bson type which is
// able to serialize date-times into byte buffers
BsonObject.UTC_DATE_TIME.writer(new BsonWriter() {
  @Override public void writeTo(ByteBuffer buffer, Object reference) {
    buffer.putLong(((DateTime) reference).getMillis());

// finally register a reader to do all this ass backwards
BsonObject.UTC_DATE_TIME.reader(new BsonReader() {
  @Override public Object readFrom(ByteBuffer buffer) {
    return new DateTime(buffer.getLong());
Something went wrong with that request. Please try again.