Firebase Admin Java SDK

The Firebase Admin Java SDK allows you to access Firebase services from privileged environments, such as servers or cloud functions, using Java.

Currently, the SDK provides support for:

• Firebase custom authentication
• Firebase Realtime Database access

It is compatible with Google App Engine.

The Firebase Admin Java SDK is distributed via Maven Central. To use it in your project, configure your build tool (like Maven or Gradle) to include the following coordinates:

• Group ID: com.google.firebase
• Artifact ID: firebase-admin

<!-- Maven dependency example -->
<dependency>
  <groupId>com.google.firebase</groupId>
  <artifactId>firebase-admin</artifactId>
  <version>LATEST_VERSION</version>
</dependency>

The Firebase Admin Java SDK supports Java 8 and higher.

For new development, Java 11 and Java 17 are recommended as the best choices. The SDK follows the Oracle Java SE support roadmap.

When using FirebaseMessaging#sendAll(List<Message>) or FirebaseMessaging#sendMulticast(MulticastMessage), the SDK returns a BatchResponse. This object contains the individual results for every message sent in the batch.

You can use BatchResponse to inspect which specific messages succeeded or failed and to get aggregate counts of successes and failures.

// Example of processing a BatchResponse
BatchResponse response = messaging.sendAll(messages);

int successes = response.getSuccessCount();
int failures = response.getFailureCount();

List<SendResponse> individualResponses = response.getResponses();
for (SendResponse res : individualResponses) {
    if (!res.isSuccessful()) {
        // Handle individual message failure
        System.out.println("Message failed: " + res.getException().getMessage());
    }
}

Use Index.fromQueryDefinition(String str) to create an index based on a Firebase Realtime Database query definition string.

Supported strings:

• ".value": Returns a ValueIndex for indexing by node value.
• ".key": Returns a KeyIndex for indexing by node key.
• Any other string: Returns a PathIndex corresponding to the provided path.

Note: Passing ".priority" will throw an IllegalStateException because priority is the default ordering behavior.

// Example of creating an index for a specific path
Index pathIndex = Index.fromQueryDefinition("some/path/to/index");

// Example of creating a value index
Index valueIndex = Index.fromQueryDefinition(".value");

The Index class is an abstract base class used to define how nodes in a Realtime Database snapshot are ordered and compared. It implements Comparator<NamedNode>, allowing you to sort nodes based on specific criteria (like value, key, or a specific path).

Key methods for consumers:

• compare(NamedNode one, NamedNode two, boolean reverse): Compares two nodes. Use the reverse flag to toggle between ascending and descending order.
• indexedValueChanged(Node oldNode, Node newNode): Determines if the indexed value has changed between two nodes, which is useful for tracking updates.
• getQueryDefinition(): Returns the string representation of the query definition this index represents.

The BatchResponse interface provides the following methods to inspect the results of a bulk FCM operation:

• getResponses(): Returns a List<SendResponse> containing the result of each individual message in the batch.
• getSuccessCount(): Returns the total number of messages that were successfully sent.
• getFailureCount(): Returns the total number of messages that failed to send.