This page is archived and might not reflect the latest version of the FlutterFire plugins. You can find the latest information on firebase.google.com:
Before talking about how your app reads from and writes to Realtime Database, let's introduce a set of tools you can use to prototype and test Realtime Database functionality: Firebase Emulator Suite. If you're trying out different data models, optimizing your security rules, or working to find the most cost-effective way to interact with the back-end, being able to work locally without deploying live services can be a great idea.
A Realtime Database emulator is part of the Emulator Suite, which enables your app to interact with your emulated database content and config, as well as optionally your emulated project resources (functions, other databases, and security rules).emulator_suite_short
Using the Realtime Database emulator involves just a few steps:
- Adding a line of code to your app's test config to connect to the emulator.
- From the root of your local project directory, running
- Making calls from your app's prototype code using a Realtime Database platform SDK as usual, or using the Realtime Database REST API.
A detailed walkthrough involving Realtime Database and Cloud Functions is available. You should also have a look at the Emulator Suite introduction.
To read or write data from the database, you need an instance of
This document covers the basics of reading and writing Firebase data.
Firebase data is written to a
DatabaseReference and retrieved by
awaiting or listening for events emitted by the reference. Events are emitted
once for the initial state of the data and again anytime the data changes.
By default, read and write access to your database is restricted so only authenticated users can read or write data. To get started without setting up Firebase Authentication, you can configure your rules for public access. This does make your database open to anyone, even people not using your app, so be sure to restrict your database again when you set up authentication.
For basic write operations, you can use
set() to save data to a specified
reference, replacing any existing data at that path. You can set a reference
to the following types:
For instance, you can add a user with
set() as follows:
set() in this way overwrites data at the specified location,
including any child nodes. However, you can still update a child without
rewriting the entire object. If you want to allow users to update their profiles
you could update the username as follows:
update() method accepts a sub-path to nodes, allowing you to update multiple
nodes on the database at once:
To read data at a path and listen for changes, use the
onValue property of
DatabaseReference to listen for
You can use the
DatabaseEvent to read the data at a given path,
as it exists at the time of the event. This event is triggered once when the
listener is attached and again every time the data, including any children,
changes. The event has a
snapshot property containing all data at that
location, including child data. If there is no data, the snapshot's
exists property will be
false and its
value property will be null.
DatabaseEvent is fired every time data is changed at
the specified database reference, including changes to children. To limit the
size of your snapshots, attach only at the highest level needed for watching
changes. For example, attaching a listener to the root of your database is
The following example demonstrates a social blogging application retrieving the details of a post from the database:
The listener receives a
DataSnapshot that contains the data at the specified
location in the database at the time of the event in its
The SDK is designed to manage interactions with database servers whether your app is online or offline.
Generally, you should use the value events techniques described above to read data to get notified of updates to the data from the backend. Those technique reduce your usage and billing, and are optimized to give your users the best experience as they go online and offline.
If you need the data only once, you can use
get() to get a snapshot of the
data from the database. If for any reason
get() is unable to return the
server value, the client will probe the local storage cache and return an error
if the value is still not found.
The following example demonstrates retrieving a user's public-facing username a single time from the database:
Unnecessary use of
get() can increase use of bandwidth and lead to loss
of performance, which can be prevented by using a realtime listener as shown
In some cases you may want the value from the local cache to be returned
immediately, instead of checking for an updated value on the server. In those
cases you can use
once() to get the data from the local disk cache
This is useful for data that only needs to be loaded once and isn't expected to change frequently or require active listening. For instance, the blogging app in the previous examples uses this method to load a user's profile when they begin authoring a new post:
To simultaneously write to specific children of a node without overwriting other
child nodes, use the
This example uses
push() to create a post in the node containing posts for
all users at
/posts/$postid and simultaneously retrieve the key with
key. The key can then be used to create a second entry in the user's
Using these paths, you can perform simultaneous updates to multiple locations in
the JSON tree with a single call to
update(), such as how this example
creates the new post in both locations. Simultaneous updates made this way
are atomic: either all updates succeed or all updates fail.
If you want to know when your data has been committed, you can register
completion callbacks. Both
Futures, to which
you can attach success and error callbacks that are called when the write has
been committed to the database and when the call was unsuccessful.
The simplest way to delete data is to call
remove() on a reference to the
location of that data.
You can also delete by specifying null as the value for another write operation
update(). You can use this technique with
delete multiple children in a single API call.
When working with data that could be corrupted by concurrent modifications,
such as incremental counters, you can use a transaction by passing a
transaction handler to
runTransaction(). A transaction handler takes the
current state of the data as an argument and
returns the new desired state you would like to write. If another client
writes to the location before your new value is successfully written, your
update function is called again with the new current value, and the write is
For instance, in the example social blogging app, you could allow users to star and unstar posts and keep track of how many stars a post has received as follows:
By default, events are raised each time the transaction update function runs,
so you run the function run multiple times, you may see intermediate states.
You can set
false to suppress these intermediate states and
instead wait until the transaction has completed before events are raised:
The result of a transaction is a
TransactionResult, which contains information
such as whether the transaction was committed, and the new snapshot:
If you want to safely cancel a transaction, call
In the above use case we're writing two values to the database: the ID of the user who stars/unstars the post, and the incremented star count. If we already know that user is starring the post, we can use an atomic increment operation instead of a transaction.
This code does not use a transaction operation, so it does not automatically get re-run if there is a conflicting update. However, since the increment operation happens directly on the database server, there is no chance of a conflict.
If you want to detect and reject application-specific conflicts, such as a user starring a post that they already starred before, you should write custom security rules for that use case.
If a client loses its network connection, your app will continue functioning correctly.
Every client connected to a Firebase database maintains its own internal version of any active data. When data is written, it's written to this local version first. The Firebase client then synchronizes that data with the remote database servers and with other clients on a "best-effort" basis.
As a result, all writes to the database trigger local events immediately, before any data is written to the server. This means your app remains responsive regardless of network latency or connectivity.
Once connectivity is reestablished, your app receives the appropriate set of events so that the client syncs with the current server state, without having to write any custom code.
Note: The Firebase Realtime Database web APIs do not persist data offline outside of the session. In order for writes to be persisted to the server, the web page must not be closed before the data is written to the server
We'll talk more about offline behavior in Learn more about online and offline capabilities.