flutter_mongodb_realm

Unofficial Flutter plugin for using MongoDB Realm services on Android, iOS and web.

Getting started

For using it on your app:

import 'package:flutter_mongodb_realm/flutter_mongo_realm.dart';

For API reference check here

The minimum requirements are:

  • Android 5.0 (API 21)
  • iOS 13.0

Setup

Doesn't require any setup besides adding as a dependency Web integration automatically!

Supported Features

Database support (MongoDB Atlas):

  • Insert
  • Find
  • Delete
  • Update
  • Watch (also to specified IDs or with Filter)
  • Aggregate [X]

Auth Providers:

  • Email/Password
  • Anonymously
  • Google [X]
  • Facebook [X]
  • JWT
  • Custom Authentication Function
  • Apple ID [X]

Authentication:

  • Auth Listener
  • Reset Password
  • Login/Sign in/Logout
  • Link User Identities X
  • Access token or Refresh token

Functions

  • Calling a Realm function

Note: Other more features will come in the future :)

Note: [X] = Not implemented on Web

Usage

Initialization

Inorder to connect a RealmApp to flutter add your main function:

main() async{
  // ADD THESE 2 LINES
  WidgetsFlutterBinding.ensureInitialized();
  await RealmApp.init(<your_app_id>);
  
  // other configurations..
  
  runApp(MyApp());
}

In order to use the client define:

  final client = MongoRealmClient();

Authentication

In order to use authentication-related operations you have to use RealmApp object:

final Realm app = RealmApp();

You can retrieve the current logged user by:

final user = app.currentUser;

// for more details (like email,birthday) use his 'profile' attribute
// For example:
final userEmail = user.profile.email;

Login

You can log in using the following providers:

  • Anonymous
CoreRealmUser? mongoUser = await app.login(Credentials.anonymous());
  • Email\Password
CoreRealmUser? mongoUser = await app.login(
  Credentials.emailPassword(username: <email_address>, password: <password>));
  • Facebook

In order to login with Facebook import the required flutter's package and configure in the native side as their instructions.

usage:

CoreRealmUser? mongoUser = await app.login(
  Credentials.facebook(<access_token>));
  • Google

Inorder to make Google login works, please follow the following instructions use the following:

  1. Create project in Firebase
  2. Add a web app
  3. Go to Google Cloud Platorm -> API & services -> Credentials [https://console.cloud.google.com/apis/credentials?project=YOUR_PROJECT_ID]
  4. use the client ID of WEB
  5. use this client ID and client secret on MongoDB's as Google Authentication
  6. create SHA-1 for the app and add it to Firebase project you created
  7. Take the serverAuthCode after login, this will work if all set correctly

    This has to be done in order to get the auth code need by the Mongo Realm SDK

Calling on Flutter:

 GoogleSignIn _googleSignIn = GoogleSignIn(
    scopes: [
      'email',
    ],
    clientId: '[YOUR_APP_ID].apps.googleusercontent.com'
);

var account = await _googleSignIn.signIn();
var serverAuthCode = account?.serverAuthCode;

CoreRealmUser? mongoUser = await app.login(GoogleCredential2(serverAuthCode!));
  • Custom JWT
CoreRealmUser? mongoUser = await app.login(Credentials.jwt(<token>);
  • Custom (Auth) Function
MongoDocument payload = MongoDocument({
  "username": "bob"
})
CoreRealmUser? mongoUser = await app.login(Credentials.function(payload);
  • Apple ID
  1. Add a dependency that implementing signing in with apple, such as apple_sign_in, or sign_in_with_apple

  2. In your Flutter app, retrieve Identify token from the login results

// taken from the example project using the apple_sign_in plugin
final appleResult = await AppleSignIn.performRequests([
      AppleIdRequest(requestedScopes: [Scope.email, Scope.fullName])
    ]);

    if (appleResult.error != null) {
      // handle errors from Apple here
    }

    var idToken =  String.fromCharCodes(appleResult.credential.identityToken);
  1. Use the token in the login function with the Credentials.apple auth provider
CoreRealmUser? mongoUser = app.login(Credentials.apple(idToken));

NOTE: In any case , if mongoUser != null the login was successful.

Register

Register a user with Email\Password

CoreRealmUser? mongoUser = await app.registerWithEmail(
    email: <email_address>, password: <password>);

Logout

await app.currentUser.logout();   // Logout the current user

Reset Password

You can send an email to reset user password: (email must be linked to an existing account)

await app.sendResetPasswordEmail(<YOUR_DESIRED_EMAIL>);

Auth Listener

You can be notified when the auth state changes, such as login/logout..

/// for using as smart navigation according to user logged in or not
StreamBuilder _authBuilder(BuildContext context) {
  return StreamBuilder(
    stream: app.authListener(),
    builder: (context, AsyncSnapshot snapshot) {
      switch (snapshot.connectionState) {
        case ConnectionState.none:
        case ConnectionState.waiting:
          // show loading indicator
          return Scaffold(body: Center(child: CircularProgressIndicator()));
        case ConnectionState.active:
          // log error to console
          if (snapshot.error != null) {
            return Container(
              alignment: Alignment.center,
              child: Text(snapshot.error.toString()),
            );
          }

          // redirect to the proper page
          return snapshot.hasData ? HomeScreen() : LoginScreen();

          default:
            return LoginScreen();
        }
      },
  );
}

The app can then link the new identity with the current user.

CoreRealmUser? mongoUser = app.login(Credentials.anonymous());
user.linkCredentials(Credentials.emailPassword("email", "password"));

Database

To get access to a collection:

  final collection = client.getDatabase(<database_name>).getCollection(<collection_name>);

Insert


collection.insertOne(MongoDocument({
  // key-value pairs describing the document
}));

// OR

List<MongoDocument> documents = ...;
collection.insertMany(documents);

Find

filtering can be used with QueryOperator class for more robust code


// fetch all document in the collection
var docs = collection.find();

// fetch all document in the collection applying some filter
var docs = collection.find(
  filter: {
    "year": QueryOperator.gt(2010)..lte(2014), // == 'year'>2010 && 'year'<=2014
});

// optional: can also add find options (projection/limit/sort)
var docs = await collection.find(
  filter: {
    "year": QueryOperator.gt(2010)..lte(2014),
  },
  options: RemoteFindOptions(
    projection: {
      "title": ProjectionValue.INCLUDE,
      "rated": ProjectionValue.INCLUDE,
      "year": ProjectionValue.INCLUDE,
    },
    limit: 70,
    sort: {
      "year": OrderValue.ASCENDING,
    }
  ),
);


// the same as above, but just get the first matched one
var document = await collection.findOne();

var document = await collection.findOne(
  filter: {
    "year": QueryOperator.gt(2010)..lte(2014), // == 'year'>2010 && 'year'<=2014
});


// fetch the number of document in the collection
int collectionSize = await collection.count();

// count the number of document that apply to some filter
int size = await collection.count({
    "age": 25,
});

Delete

filtering can be used with QueryOperator class for more robust code

// fetch all documents in the collection
var deletedDocs = await collection.deleteMany({});

// fetch the first document in the collection applying some filter
var deletedDocs = await collection.deleteOne({"age": 24});

// fetch all document in the collection applying some filter
var deletedDocs = 
  await collection.deleteMany({"age": QueryOperator.gte(24)});

Update

Updating the only first matched document:

await collection.updateOne(
  // adding a filter (optional)
  filter:{
    "_id": ObjectId('601204a6a80d3fbab2e3a73f'),
  },

  // adding an update operation (as matched the MongoDB SDK ones)
  update: UpdateSelector.set({
    "age": 26,
  });

);

Updating the only all matched documents:

  // adding a filter (optional)
  filter:{
    "name": "adam",
  },

  // adding an update operation (as matched the MongoDB SDK ones)
  update: UpdateSelector.set({
    "quantity": 670,
  });

  // removing 'apples' from 'favs' array
  update: UpdateOperator.pull({
     "favs": QueryOperator.in$(['apples'])
  });
  
  // adding 'tomatoes' & 'onions into 'favs' array
  update: UpdateOperator.push({
     "favs": ArrayModifier.each(['tomatoes','onions'])
  });
);

Watch

First, Get the stream to subscribed to any document change in the collection

final stream = collection.watch();

// (optional) can watch only specified documents by their ids:
// 1. if they defined as ObjectId type
final stream2 = myCollection.watch(ids: ["5eca2d9fff448a4cbf8f6627"]);

// 2. if they defined as String type (`asObjectIds` is true by default)
final stream3 = myCollection.watch(ids: ["22", "8"], asObjectIds: false);

OR get the stream to subscribed to a part of the collection applying filter on the listened documents

final streamFilter = collection.watchWithFilter({
  "age": QueryOperator.lte(26)
});

Afterwards, set a listener to a change in the collection

stream.listen((data) {
  // data contains JSON string of the document that was changed
  var fullDocument = MongoDocument.parse(data);
  
  // Do other stuff...
});

Aggregation

define Pipeline Stages for aggregation, i.e:

List<PipelineStage> pipeline = [
  PipelineStage.addFields({
    "totalHomework": AggregateOperator.sum("homework"),
    "totalQuiz": AggregateOperator.sum("quiz"),
  }),
  PipelineStage.addFields({
    "totalScore": AggregateOperator.add(
        ["totalHomework", "totalQuiz", "extraCredit"]),
  }),
];

And then set the pipelines stages to the aggregate function:

 var list = await collection.aggregate(pipeline);

Another usages (not all stages are supported):

List<PipelineStage> pipeline = [
  PipelineStage.skip(2),
  PipelineStage.match({"status": "A"}),
  PipelineStage.group(
    "cust_id",
    accumulators: {"total": AggregateOperator.sum("amount")},
  ),
];

List<PipelineStage> pipeline = [
  PipelineStage.sample(2),
];

// can also RAW typed one
List<PipelineStage> pipeline = [

  PipelineStage.raw({
        //... some expression according to the MongoDB API
  }),
  PipelineStage.raw({
        //... some expression according to the MongoDB API
  }),
  ...
];

Functions

for calling a defined Realm function "sum" with argument 3&4

var result = await client.callFunction("sum", args: [3, 4])

You can also add a timeout (in ms), i.e 60 seconds:

var result = await client.callFunction("sum", args: [3, 4], requestTimeout: 60000)

Donate

If you found this project helpful or you learned something from the source code and want to thank me, consider buying me a cup of :coffee:

Note: flutter_mongo_realm is not directly and/or indirectly associated/affiliated with MongoDBTM , Flutter or Google LLC.