Getting Started

Installation

Follow the installation instructions below to install Realm JavaScript via npm, or see the source on GitHub.

• React Native
• Node.js

cd ios && pod install && cd ..

const Realm = require('realm');

class <project-name> extends Component {
  constructor(props) {
    super(props);
    this.state = { realm: null };
  }

  componentWillMount() {
    Realm.open({
      schema: [{name: 'Dog', properties: {name: 'string'}}]
    }).then(realm => {
      realm.write(() => {
        realm.create('Dog', {name: 'Rex'});
      });
      this.setState({ realm });
    });
  }

  render() {
    const info = this.state.realm
      ? 'Number of dogs in this Realm: ' + this.state.realm.objects('Dog').length
      : 'Loading...';

    return (
      <View style={styles.container}>
        <Text style={styles.welcome}>
          {info}
        </Text>
      </View>
    );
  }
}

npm install --save realm

var Realm = require('realm');

Introduction

Realm JavaScript enables you to efficiently write your app’s model layer in a safe, persisted and fast way. It’s designed to work with React Native and Node.js.

Here’s a quick example:

const Realm = require('realm');

// Define your models and their properties
const CarSchema = {
  name: 'Car',
  properties: {
    make:  'string',
    model: 'string',
    miles: {type: 'int', default: 0},
  }
};
const PersonSchema = {
  name: 'Person',
  properties: {
    name:     'string',
    birthday: 'date',
    cars:     'Car[]',
    picture:  'data?' // optional property
  }
};

Realm.open({schema: [CarSchema, PersonSchema]})
  .then(realm => {
    // Create Realm objects and write to local storage
    realm.write(() => {
      const myCar = realm.create('Car', {
        make: 'Honda',
        model: 'Civic',
        miles: 1000,
      });
      myCar.miles += 20; // Update a property value
    });

    // Query Realm for all cars with a high mileage
    const cars = realm.objects('Car').filtered('miles > 1000');

    // Will return a Results object with our 1 car
    cars.length // => 1

    // Add another car
    realm.write(() => {
      const myCar = realm.create('Car', {
        make: 'Ford',
        model: 'Focus',
        miles: 2000,
      });
    });

    // Query results are updated in realtime
    cars.length // => 2
  })
  .catch(error => {
    console.log(error);
  });

Note that if you are going to use Realm for server-side/node, there is additional information to be found in the Realm Object Server documentation.

Realm Studio

Realm Studio is our premiere developer tool, built so you can easily manage the Realm Database and Realm Platform. With Realm Studio, you can open and edit local and synced Realms, and administer any Realm Object Server instance. It supports Mac, Windows and Linux.

Realms

Opening Realms

Opening a Realm is simply performed by calling the static open method on the Realm class. Pass a configuration object. We’ve seen this used already in examples with a configuration object that includes the schema key:

// Get the default Realm with support for our objects
Realm.open({schema: [Car, Person]})
  .then(realm => {
    // ...use the realm instance here
  })
  .catch(error => {
    // Handle the error here if something went wrong
  });

For full details about the configuration object, see the API Reference for configuration. Some of the more common keys for the object, beyond schema, include:

• path: specify a path to another Realm
• migration: a migration function
• sync: a sync object, to open a Realm synchronized with the Realm Object Server
• inMemory: the Realm will be opened in-memory, and objects are not persisted; once the last Realm instance is closed, all objects vanish
• deleteRealmIfMigrationNeeded: delete the Realm if migration is needed; this is useful under development since the data model might change often

The Default Realm

You may have noticed in all previous examples that the path argument has been omitted. In this case the default Realm path is used. You can access and change the default Realm path using the Realm.defaultPath global property.

Opening a Synchronized Realm

Other Realms

It’s sometimes useful to have multiple Realms persisted at different locations. For example, you may want to bundle some data with your application in a Realm file, in addition to your main Realm. You can do this by specifying the path argument when initializing your realm. All paths are relative to the writable documents directory for your application:

// Open a realm at another path
Realm.open({
  path: 'anotherRealm.realm',
  schema: [CarSchema]
}).then(/* ... */);

Schema Version

Another option available when opening a Realm is the schemaVersion property. When omitted, the schemaVersion property defaults to 0. You are required to specify the schemaVersion when initializing an existing Realm with a schema that contains objects that differ from their previous specification. If the schema was updated and the schemaVersion was not, an exception will be thrown.

const PersonSchema = {
  name: 'Person',
  properties: {
    name: 'string'
  }
};

// schemaVersion defaults to 0
Realm.open({schema: [PersonSchema]});

If you then later do something like this:

const UpdatedPersonSchema = {
  // The schema name is the same, so previous `Person` object
  // in the Realm will be updated
  name: 'Person',
  properties: {
    name: 'string',
    dog:  'Dog'     // new property
  }
};

// this will throw because the schema has changed
// and `schemaVersion` is not specified
Realm.open({schema: [UpdatedPersonSchema]});

// this will succeed and update the Realm to the new schema
Realm.open({schema: [UpdatedPersonSchema], schemaVersion: 1});

If you wish to retrieve the current schema version of a Realm, you may do so with the Realm.schemaVersion method.

const currentVersion = Realm.schemaVersion(Realm.defaultPath);

Synchronously Opening Realms

You can create a realm instance by simply invoking the constructor and passing a configuration object to it. This is typically not recommended as it blocks and could potentially be a time consuming operation, especially if there are migrations to run or if the realm is synchronized and you don’t want to risk modifying data before it’s been completely downloaded.

If you still want to do this, the pattern is simple:

const realm = new Realm({schema: [PersonSchema]});

// You can now access the realm instance.
realm.write(/* ... */);

Models

Realm data models are defined by the schema information passed into a Realm during initialization. The schema for an object consists of the object’s name and a set of properties. Each property has a name and is described by either a string containing the property’s type, or an object with name, type, objectType, optional, default, and indexed fields.

const Realm = require('realm');

const CarSchema = {
  name: 'Car',
  properties: {
    make:  'string',
    model: 'string',
    miles: {type: 'int', default: 0},
  }
};
const PersonSchema = {
  name: 'Person',
  properties: {
    name:     'string',
    birthday: 'date',
    cars:     'Car[]'
    picture:  'data?', // optional property
  }
};

// Initialize a Realm with Car and Person models
Realm.open({schema: [CarSchema, PersonSchema]})
  .then(realm => {
    // ... use the realm instance to read and modify data
  })

To avoid size limitations and a performance impact, it is best not to store large blobs (such as image and video files) directly in Realm. Instead, save the file to a file store (such as S3) and keep only the location of the file and any relevant metadata in Realm.

Classes

If you want to use ES2015 classes (and maybe want to inherit existing features), you just need to define the schema on the constructor:

class Person {
  get fullName() {
    return this.firstName + ' ' + this.lastName;
  }
}

Person.schema = {
  name: 'Person',
  properties: {
    firstName: 'string',
    lastName: 'string'
  }
};

You can now pass the class itself to the schema property of the open configuration:

Realm.open({schema: [Person]})
  .then( /* ... */ );

You access properties as always:

realm.write(() => {
  const john = realm.create('Person', {
    firstName: 'John',
    lastName: 'Smith'
  });
  john.lastName = 'Peterson';
  console.log(john.fullName); // -> 'John Peterson'
});

Supported types

Realm supports the following basic types: bool, int, float, double, string, data, and date.

• bool properties map to JavaScript boolean values
• int, float, and double properties map to JavaScript number values. Internally ‘int’ and ‘double’ are stored as 64 bits while float is stored with 32 bits.
• string properties map to string
• data properties map to ArrayBuffer
• date properties map to Date

When specifying basic properties as a shorthand you may specify only the type rather than having to specify a dictionary with a single entry:

const CarSchema = {
  name: 'Car',
  properties: {
    // The following property types are equivalent
    make:   {type: 'string'},
    model: 'string',
  }
}

Optional Properties

By default, basic types are non-optional and do not support storing null or undefined. Properties can be made optional by specifying the optional designator in your property definition, or with the shorthand syntax, by appending a ? to the type name:

const PersonSchema = {
  name: 'Person',
  properties: {
    realName:    'string', // required property
    displayName: 'string?', // optional property
    birthday:    {type: 'date', optional: true}, // optional property
  }
};

let realm = new Realm({schema: [PersonSchema, CarSchema]});

realm.write(() => {
  // optional properties can be set to null or undefined at creation
  let charlie = realm.create('Person', {
    realName: 'Charlie',
    displayName: null, // could also be omitted entirely
    birthday: new Date(1995, 11, 25),
  });

  // optional properties can be set to `null`, `undefined`,
  // or to a new non-null value
  charlie.birthday = undefined;
  charlie.displayName = 'Charles';

  // Setting a non-optional property to null will throw `TypeError`
  // charlie.realName = null;
});

List Properties

In addition to storing individual values, properties can also be declared as a list of any of the supported basic types. This is done by appending [] to the type name:

const PersonSchema = {
  name: 'Person',
  properties: {
    name: 'string',
    testScores: 'double?[]'
  }
};

let realm = new Realm({schema: [PersonSchema, CarSchema]});

realm.write(() => {
  let charlie = realm.create('Person', {
    name: 'Charlie',
    testScores: [100.0]
  });

  // Charlie had an excused absense for the second test and was allowed to skip it
  charlie.testScores.push(null);

  // And then he didn't do so well on the third test
  charlie.testScores.push(70.0);
});

When accessing list properties a List object is returned. List has methods very similar to a regular JavaScript array. The big difference is that any changes made to a List are automatically persisted to the underlying Realm, and as such they can only be modified in write transactions. Additionally, Lists belong to the underlying object they were acquired from - you can only get List instances by accessing a property from an owning object and they cannot be manually created.

While the values in a list property can be made optional, the list property itself cannot be. Specifying a list property as optional using the longhand syntax (values: {type: 'int[]', optional: true}) will make the values within the list optional.

Relationships

To-One Relationships

For to-one relationships you specify the name property of the object schema you are referencing as the property’s type:

const PersonSchema = {
  name: 'Person',
  properties: {
    // The following property definitions are equivalent
    car: {type: 'Car'},
    van: 'Car',
  }
};

When using object properties you need to make sure all referenced types are present in the schema used to open the Realm:

// CarSchema is needed since PersonSchema contains properties of type 'Car'
Realm.open({schema: [CarSchema, PersonSchema]})
  .then(/* ... */);

When accessing object properties, you can access nested properties using normal property syntax:

realm.write(() => {
  const nameString = person.car.name;
  person.car.miles = 1100;

  // create a new Car by setting the property to an object
  // with all of the required fields
  person.van = {make: 'Ford', model: 'Transit'};

  // set both properties to the same car instance
  person.car = person.van;
});

Object properties in Realm are always optional, do not have to be explicitly specified as such, and cannot be made required.

To-Many Relationships

Just as with basic properties, you can also have a list of objects to form a to-many relationship. This is done by either appending [] to the name of the target object schema, or by setting the property type to list and specifying the objectType:

const PersonSchema = {
  name: 'Person',
  properties: {
    // The following property definitions are equivalent
    cars: {type: 'list', objectType: 'Car'},
    vans: 'Car[]'
  }
}

let carList = person.cars;

// Add new cars to the list
realm.write(() => {
  carList.push({make: 'Honda', model: 'Accord', miles: 100});
  carList.push({make: 'Toyota', model: 'Prius', miles: 200});
});

let secondCar = carList[1].model;  // access using an array index

Unlike other lists and to-one relationships, to-many relationships cannot be made optional.

Inverse Relationships

Links are unidirectional. So if a to-many property Person.dogs links to Dog instances and a to-one property Dog.owner links to Person, these links are independent from one another. Appending a Dog to a Person instance’s dogs property doesn’t automatically set the dog’s owner property to this Person. Because manually synchronizing pairs of relationships is error prone, complex and duplicates information, Realm provides linking objects properties to represent these inverse relationships.

With linking objects properties, you can obtain all objects that link to a given object from a specific property. For example, a Dog object can have a property named owners that contains all of the Person objects that have this exact Dog object in their dogs property. This is done by making the owners property of type linkingObjects and then specifying the relationship that it has with the Person object.

const PersonSchema = {
  name: 'Person',
  properties: {
    dogs: 'Dog[]'
  }
}

const DogSchema = {
  name:'Dog',
  properties: {
    // No shorthand syntax for linkingObjects properties
    owners: {type: 'linkingObjects', objectType: 'Person', property: 'dogs'}
  }
}

A linkingObjects property can point either to a List property (to-many relationship) or an Object property (to-one relationship):

const ShipSchema = {
  name: 'Ship',
  properties: {
    captain: 'Captain'
  }
}

const CaptainSchema = {
  name: 'Captain',
  properties: {
    ships: {type: 'linkingObjects', objectType: 'Ship', property: 'captain'}
  }
}

When accessing linkingObjects properties, a Results object is returned, so further querying and sorting are fully supported. linkingObject properties belong to the object they were acquired from and can not be set or manipulated directly. They are updated automatically when a transaction is committed.

Accessing linkingObjects without a schema: if you have opened a Realm file without specifying a schema, e.g. in a Realm Functions callback, you can get a linkingObjects property by calling linkingObjects(objectType, property) on an Object instance:

let captain = realm.objectForPrimaryKey('Captain', 1);
let ships = captain.linkingObjects('Ship', 'captain');

Linking objects properties cannot be made optional.

Default Property Values

Default property values can be specified by setting the default designator in the property definition. To use a default value, leave the property unspecified during object creation.

const CarSchema = {
  name: 'Car',
  properties: {
    make:  {type: 'string'},
    model: {type: 'string'},
    drive: {type: 'string', default: 'fwd'},
    miles: {type: 'int',    default: 0}
  }
};

realm.write(() => {
  // Since `miles` is left out it defaults to `0`, and since
  // `drive` is specified, it overrides the default value
  realm.create('Car', {make: 'Honda', model: 'Accord', drive: 'awd'});
});

Indexed Properties

You can add an indexed designator to a property definition to cause that property to be indexed. This is supported for int, string, bool, and date property types:

var BookSchema = {
  name: 'Book',
  properties: {
    name: { type: 'string', indexed: true },
    price: 'float'
  }
};

Indexing a property will greatly speed up queries where the property is compared for equality at the cost of slower insertions.

Primary Keys

You can specify the primaryKey property in an object model for string and int properties. Declaring a primary key allows objects to be looked up and updated efficiently and enforces uniqueness for each value. Once an object with a primary key has been added to a Realm the primary key cannot be changed.

const BookSchema = {
  name: 'Book',
  primaryKey: 'id',
  properties: {
    id:    'int',    // primary key
    title: 'string',
    price: 'float'
  }
};

Primary key properties are automatically indexed.

Writes

Changes to objects in a Realm—creating, updating and deleting—must take place within a write() transaction block. Note that write transactions have a non-negligible overhead; you should try to minimize the number of write blocks within your code.

Creating Objects

Objects are created using the create method:

try {
  realm.write(() => {
    realm.create('Car', {make: 'Honda', model: 'Accord', drive: 'awd'});
  });
} catch (e) {
  console.log("Error on creation");
}

Note that any exceptions thrown in write() will cancel the transaction. The try/catch block won’t be shown in all examples, but it’s good practice.

Nested Objects

If an object has object properties, values for those properties can be created recursively by specifying JSON values for each child property:

realm.write(() => {
  realm.create('Person', {
    name: 'Joe',
    // nested objects are created recursively
    car: {make: 'Honda', model: 'Accord', drive: 'awd'},
  });
});

Updating Objects

Typed Updates

You can update any object by setting its properties within a write transaction.

realm.write(() => {
  car.miles = 1100;
});

Creating and Updating Objects With Primary Keys

If your model class includes a primary key, you can have Realm intelligently update or add objects based off of their primary key values. This is done by passing true as the third argument to the create method:

realm.write(() => {
  // Create a book object
  realm.create('Book', {id: 1, title: 'Recipes', price: 35});

  // Update book with new price keyed off the id
  realm.create('Book', {id: 1, price: 55}, true);
});

In the example above, since an object already exists with the id value of 1 and we have passed in true for the third argument, the price property is updated rather than trying to create a new object. Since the title property is omitted the object retains the original value for this property. Note that when creating or updating objects with primary key properties the primary key must be specified.

Deleting Objects

Objects can be deleted by calling the delete method within a write transaction.

realm.write(() => {
  // Create a book object
  let book = realm.create('Book', {id: 1, title: 'Recipes', price: 35});

  // Delete the book
  realm.delete(book);

  // Delete multiple books by passing in a `Results`, `List`,
  // or JavaScript `Array`
  let allBooks = realm.objects('Book');
  realm.delete(allBooks); // Deletes all books
});

Queries

Queries allow you to get objects of a single type from a Realm, with the option of filtering and sorting those results. All queries (including queries and property access) are lazy in Realm. Data is only read when objects and properties are accessed. This allows you to represent large sets of data in a performant way.

When performing queries you are returned a Results object. Results are simply a view of your data and are not mutable.

The most basic method for retrieving objects from a Realm is using the objects method on a Realm to get all objects of a given type:

let dogs = realm.objects('Dog'); // retrieves all Dogs from the Realm

Filtering

You can get a filtered Results by calling the filtered method on a List or a Results with a query string.

For example, the following would change our earlier example to retrieve all dogs with the color tan and names beginning with ‘B’:

let dogs = realm.objects('Dog');
let tanDogs = dogs.filtered('color = "tan" AND name BEGINSWITH "B"');

Complex filters that also involve sorting and/or limiting should be handled within the same filtered call for better performance. For example, to find all dogs with the color tan, names beginning with ‘B’, sorting in descending order, and limiting the results to 5:

let dogs = realm.objects('Dog');
let onlyFiveTanDogs = dogs.filtered('color = "tan" AND name BEGINSWITH "B" SORT(name DESC) LIMIT(5)')

The query language supported by Realm is inspired by Apple’s NSPredicate. Let’s briefly summarize the supported features and syntax:

• The comparison operands can be property names or constants. At least one of the operands must be a property name. Special constants are false, true, and null.
• Timestamps can be specified in the formats YYYY-MM-DD@HH:MM:SS:NANOSECONDS and YYYY-MM-DDTHH:MM:SS:NANOSECONDS where nanoseconds can be omitted.
• The comparison operators =/==, <=, <, >=, >, and !=/<> are supported for int, float, double, and Date property types, e.g. age = 45.
• The comparison operators =/== and !=/<> are supported for boolean (bool) properties.
• For string and data (ArrayBuffer) properties, the = (and ==), != (and <>), BEGINSWITH, CONTAINS, and ENDSWITH operators are supported, e.g. name CONTAINS 'Ja'.
• Wildcard comparison is possible for strings with the LIKE operator, e.g. name LIKE '*an?' to match “Jane”, “Dan”, “Shane”, etc.
• Case-insensitive comparisons for strings using [c], e.g. , CONTAINS[c] 'Ja'. Note that only characters “A-Z” and “a-z” will be ignored for case.
• Realm supports the following compound operators: AND/&&, OR/ ||, and NOT/!, e.g. name BEGINSWITH 'J' AND age >= 32.
• The aggregate expressions @count/@size, @min, @max, @sum and @avg are supported on list properties, e.g. employees.@count > 5 to find list of employees with more than 5 elements.
• The aggregate expression @count/@size on strings and binary properties, e.g. name.@size = 5 to find all with a name of 5 letters.
• A key path can follow list properties relationships, e.g. child.age >= 13 and cars.@avg.milage > 1000.
• A key path can also follow linking objects (backlinks), e.g. parents.age > 25 and parents.@count == 2.
• The $ operator can be used to substitute arguments, e.g. child.age >= $0 (see the example below).
• Sorting and find distinct values are possible with functions SORT and DISTINCT, e.g. age > 20 SORT(name ASC, age DESC) DISTINCT(name).
  • The ordering for sorting can be one of the following case insensitive literals: ASC, ASCENDING, DESC, DESCENDING.
  • Any number of properties can appear inside the brackets in a comma separated list.
  • Any number of sort/distinct conditions can be indicated, they will be applied in the specified order.
  • Sort or distinct cannot operate independently, these conditions must be attached to at least one query filter.

An non-trivial example on how to query a Realm is:

const Realm = require('realm');

const CarSchema = {
  name: 'Car',
  properties: {
    make:  'string',
    model: 'string',
    miles: {type: 'int', default: 0},
  }
};

const PersonSchema = {
  name: 'Person',
  properties: {
    name:     'string',
    cars:     {type: 'list', objectType: 'Car'},
  }
};

// Initialize a Realm with Car and Person models
Realm.open({schema: [CarSchema, PersonSchema]})
    .then(realm => {

        // Add persons and their cars
        realm.write(() => {
            let john = realm.create('Person', {name: 'John', cars: []});
            john.cars.push({make: 'Honda',  model: 'Accord', miles: 1500});
            john.cars.push({make: 'Toyota', model: 'Prius',  miles: 2780});

            let joan = realm.create('Person', {name: 'Joan', cars: []});
            joan.cars.push({make: 'Skoda', model: 'Octavia', miles: 1120});
            joan.cars.push({make: 'Ford',  model: 'Fiesta',  miles: 95});
            joan.cars.push({make: 'VW',    model: 'Golf',    miles: 1270});

            let jill = realm.create('Person', {name: 'Jill', cars: []});

            let jack = realm.create('Person', {name: 'Jack', cars: []});
            jack.cars.push({make: 'Porche', model: '911',    miles: 965});
        });

        // Find car owners
        let carOwners = realm.objects('Person').filtered('cars.@size > 0');
        console.log('Car owners')
        for (let p of carOwners) {
            console.log(`  ${p.name}`);
        }

        // Find who has been driver longer than average
        let average = realm.objects('Car').avg('miles');
        let longerThanAverage = realm.objects('Person').filtered('cars.@sum.miles > $0', average);
        console.log(`Longer than average (${average})`)
        for (let p of longerThanAverage) {
            console.log(`  ${p.name}: ${p.cars.sum('miles')}`);
        }

        realm.close();
});

The output of the code snippet is:

Car owners
  John
  Joan
  Jack
Longer than average (1288.3333333333333)
  John: 4280
  Joan: 2485

Sorting

Results allows you to specify a sort criteria and order based on a single or multiple properties. For example, the following call sorts the returned cars from the example above numerically by miles:

let hondas = realm.objects('Car').filtered('make = "Honda"');

// Sort Hondas by mileage
let sortedHondas = hondas.sorted('miles');

// Sort in descending order instead
sortedHondas = hondas.sorted('miles', true);

// Sort by price in descending order and then miles in ascending
sortedHondas = hondas.sorted([['price', true], ['miles', false]]);

Results can also be sorted on the values of objects linked to by the objects you’re sorting:

let people = realm.objects('Person');

// Sort people by the milage of their cars
let sortedPeople = people.sorted('car.miles');

Lists of basic types can be sorted by their values by calling sorted() without specifying a property:

let person = realm.objects('Person')[0];
let sortedTestScores = person.testScores.sorted();

Auto-Updating Results

Results instances are live, auto-updating views into the underlying data, which means results never have to be re-fetched. Modifying objects that affect the query will be reflected in the results immediately. The one exception to this is when using for...in or for...of, which will always iterate over the objects which matched the query when the iteration is started, even if some of them are deleted or modified to be excluded by the filter during the iteration.

let hondas = realm.objects('Car').filtered('make = "Honda"');
// hondas.length == 0

realm.write(() => {
  realm.create('Car', {make: 'Honda', model: 'RSX'});
});
// hondas.length == 1

This applies to all Results instances, included those returned by the objects, filtered, and sorted methods.

This property of Results not only keeps Realm fast and efficient, it allows your code to be simpler and more reactive. For example, if your view relies on the results of a query, you can store the Results in a property and access it without having to make sure to refresh its data prior to each access.

You can subscribe to notifications to know when Realm data is updated, indicating when your app’s UI should be refreshed for example, without having to re-fetch your Results.

Limiting Results

Most other database technologies provide the ability to ‘paginate’ results from queries (such as the ‘LIMIT’ keyword in SQLite). This is often done out of necessity to avoid reading too much from disk, or pulling too many results into memory at once.

Since queries in Realm are lazy, performing this sort of paginating behavior isn’t necessary at all, as Realm will only load objects from the results of the query once they are explicitly accessed.

If for UI-related or other implementation reasons you require a specific subset of objects from a query, it’s as simple as taking the Results object, and reading out only the objects you need.

let cars = realm.objects('Car');

// get first 5 Car objects
let firstCars = cars.slice(0, 5);

Migrations

When working with a database your data model will most likely change over time. For example, suppose we have the following Person model:

const PersonSchema = {
  name: 'Person',
  properties: {
    firstName: 'string',
    lastName: 'string',
    age: 'int'
  }
}

We want to update the data model to require a name property, rather than separate first and last names. To do this, we simply change the schema to the following:

const PersonSchema = {
  name: 'Person',
  properties: {
    name: 'string',
    age: 'int'
  }
}

At this point if you had saved any data with the previous model version there will be a mismatch between the new code and the old data Realm has stored on disk. When this occurs, an exception will be thrown when you try to open the existing Realm with the new schema unless you run a migration.

Performing a Migration

You define a migration and the associated schema version by updating the schemaVersion and defining an optional migration function. Your migration function provides any logic needed to convert data models from previous schemas to the new schema. When opening a Realm the migration function will be applied to update the Realm to the given schema version only if a migration is needed.

If no migration function is supplied then any new properties are automatically added and old properties are removed from the database when updating to the new schemaVersion. If you need to update old or populate new properties when upgrading your version you can do this in the migration function. For example, suppose we want to migrate the Person model declared earlier. You can populate the name property of the new schema using the old firstName and lastName properties:

Realm.open({
  schema: [PersonSchema],
  schemaVersion: 1,
  migration: (oldRealm, newRealm) => {
    // only apply this change if upgrading to schemaVersion 1
    if (oldRealm.schemaVersion < 1) {
      const oldObjects = oldRealm.objects('Person');
      const newObjects = newRealm.objects('Person');

      // loop through all objects and set the name property in the new schema
      for (let i = 0; i < oldObjects.length; i++) {
        newObjects[i].name = oldObjects[i].firstName + ' ' + oldObjects[i].lastName;
      }
    }
  }
}).then(realm => {
  const fullName = realm.objects('Person')[0].name;
});

Once the migration is successfully completed the Realm and all of its objects can be accessed as usual by your app.

Linear Migrations

With the migration pattern described above you can potentially run into issues when migrating over multiple versions. This could happen if a user skips an app update and a property has been changed multiple times in the versions being skipped. In this case you may need to edit old migration code to correctly update data from old schema to the latest schema.

It’s possible to avoid this issue by running multiple migrations sequentially, making sure that the database is upgraded to each previous version and that the associated migration code is run. When following this pattern old migration code should never have to be modified, although you will need to keep all old schema and migration blocks for future use. An example of what this would look like:

const schemas = [
  { schema: schema1, schemaVersion: 1, migration: migrationFunction1 },
  { schema: schema2, schemaVersion: 2, migration: migrationFunction2 },
  ...
]

// The first schema to update to is the current schema version
// since the first schema in our array is at nextSchemaIndex:
let nextSchemaIndex = Realm.schemaVersion(Realm.defaultPath);

// If Realm.schemaVersion() returned -1, it means this is a new Realm file
// so no migration is needed.
if (nextSchemaIndex !== -1) {
  while (nextSchemaIndex < schemas.length) {
    const migratedRealm = new Realm(schemas[nextSchemaIndex++]);
    migratedRealm.close();
  }
}

// Open the Realm with the latest schema
Realm.open(schemas[schemas.length-1]);

Notifications

The Realm, Results and List objects provide addListener methods to register notification callbacks. Whenever the object is updated, the change notification callback will be called.

There are two kinds of notifications, “Realm Notifications” (simple callbacks notified when write transactions are committed) and “Collection Notifications” (more sophisticated callbacks which receive change metadata on insertions, deletions and updates).

In some cases the listener may be called when the transaction starts—if the Realm is advanced to the latest version, or Realm entities being observed were modified or deleted in a way that triggers notifications. In those cases, the listener runs within the context of the current write transaction, so an attempt to begin a new write transaction within the notification handler will throw an exception. You can use the Realm.isInTransaction property to determine if your code is executing within a write transaction.

Realm Notifications

Realm instances send out notifications to other instances every time a write transaction is committed. To register for notifications:

function updateUI() {
  // ...
}

// Observe Realm Notifications
realm.addListener('change', updateUI);

// ..later remove the listener
realm.removeListener('change', updateUI);

// ..or unregister all listeners
realm.removeAllListeners();

Collection Notifications

Collection notifications contain information that describe what changes have occurred at a fine-grained level. This consists of the indices of objects that have been inserted, deleted, or modified since the last notification. Collection notifications are delivered asynchronously: first with the initial results, and then after any write transaction which modifies any of the objects in the collection, deletes objects from the collection, or adds new objects to the collection.

The notification callback function given to addListener receives two parameters when these changes occur. The first one is the collection that changed, and the second one is a changes object with information about the collection indices affected by deletions, insertions and modifications.

The former two, deletions and insertions, record the indices whenever objects start and stop being part of the collection. This takes into account when you add objects to the Realm or delete them from the Realm. For Results this also applies when you filter for specific values and the object was changed so that it is now matching the query or not matching anymore. For collections based on List, this applies when objects are added or removed from the relationship.

Your application is notified about modifications whenever a property of an object has changed, which was previously part of the collection and is still part of it. This happens as well when to-one and to-many relationships change, but doesn’t take changes on inverse relationships into account.

class Dog {}
Dog.schema = {
  name: 'Dog',
  properties: {
    name:  'string',
    age: 'int',
  }
};
class Person {}
Person.schema = {
  name: 'Person',
  properties: {
    name:    {type: 'string'},
    dogs:    {type: 'list', objectType: 'Dog'},
  }
};

Let’s assume you’re observing a list of dog owners as given by the model code above. You will be notified about modifications for a matched Person object when:

• You modify the Person’s name property.
• You add or remove a Dog to the Person’s dogs property.
• You modify the age property of a Dog belonging to that Person.

This makes it possible to discretely control the animations and visual updates made to the content inside your UI, instead of arbitrarily reloading everything each time a notification occurs.

collectionListenerRetainer = realm.objects('Dog').filtered('age < 2');

// Observe Collection Notifications
function listener(puppies, changes) {

  // Update UI in response to inserted objects
  changes.insertions.forEach((index) => {
    let insertedDog = puppies[index];
    ...
  });

  // Update UI in response to modified objects
  changes.modifications.forEach((index) => {
    let modifiedDog = puppies[index];
    ...
  });

  // Update UI in response to deleted objects
  changes.deletions.forEach((index) => {
    // Deleted objects cannot be accessed directly
    // Support for accessing deleted objects coming soon...
    ...
  });

});

collectionListenerRetainer.addListener(listener);

// Unregister all listeners
realm.removeAllListeners();

// OR Unregister this listener
collectionListenerRetainer.removeListener(listener);

Working with a Synchronized Realm

Encryption

Realm supports encrypting the database file on disk with AES-256+SHA2 by supplying a 64-byte encryption key when creating a Realm.

var key = new Int8Array(64);  // pupulate with a secure key
Realm.open({schema: [CarObject], encryptionKey: key})
  .then(realm => {
    // Use the Realm as normal
    var dogs = realm.objects('Car');
  });

This makes it so that all of the data stored on disk is transparently encrypted and decrypted with AES-256 as needed, and verified with a SHA-2 HMAC. The same encryption key must be supplied every time you obtain a Realm instance.

There is a small performance hit (typically less than 10% slower) when using encrypted Realms.

Examples

Examples can be found on GitHub in the realm-js repository.

Note that on Android, you need the NDK installed and must have set the ANDROID_NDK environment variable.

export ANDROID_NDK=/usr/local/Cellar/android-ndk/r10e

Troubleshooting

Missing Realm Constructor

If your app crashes, telling you that the Realm constructor was not found, there are a few things you can try:

First of all, run react-native link realm

If that doesn’t help, and your problem is on Android, try:

Add the following in your MainApplication.java file:

import io.realm.react.RealmReactPackage;

And add the RealmReactPackage to the list of packages:

protected List getPackages() {
    return Arrays.asList(
        new MainReactPackage(),
        new RealmReactPackage() // add this line
    );
}

add following two lines in settings.gradle:

include ':realm'
project(':realm').projectDir = new File(settingsDir, '../node_modules/realm/android')

If your problem is on iOS, try:

1. Close all simulators/device builds
2. Stop the package manager running in terminal (or better yet, just restart terminal)
3. Open the ios folder in your app root in finder
4. Go into the build folder (note: you won’t see this build folder in atom, so just right click ios and click open in finder)
5. Delete everything inside of the build folder (just move to trash and keep trash around in case you’re worried)
6. Run react-native run-ios to rebuild the whole thing

Chrome Debugging is slow

We are aware of this. The reason for this is that since Realm is written in C++ and runs native code, it has to run on the device/simulator. But given the zero-copy architecture, we need to send values in realm objects over the RPC wire every time you inspect an object that is stored in a Realm.

We are investigating various potential solutions for this problem. If you want to keep track of it, you can follow the GitHub issue.

Cannot download realm-sync-cocoa

We have seen a few reports where users are not able to build their app due to download problems. The symptome is that you see an error message like Error: unexpected end of file at Zlib.zlibOnError [as onerror] (zlib.js:142:17) errno: -5, code: 'Z_BUF_ERROR' }.

It is possible to manually download the required files, and then build the app. The steps are:

1. Locate to your project directory, and find node_modules/realm/vendor/realm-ios. It will be empty.
2. Create the file download-realm.lock using the following command: echo SYNC_SERVER_FOLDER=sync SYNC_ARCHIVE=realm-sync-cocoa-3.7.0.tar.gz SYNC_ARCHIVE_ROOT=core > download-realm.lock. The version number (here 3.7.0) must match the Realm Sync version in the change log.
3. Locate to your downloaded file realm-sync-cocoa-3.7.0.tar.gz excute this command tar -xzvf realm-sync-cocoa-3.7.0.tar.gz -C yourProjectDirectory/node_modules/realm/vendor/realm-ios.
4. You will find the directory is not empty any more. Copy all the files under directory core and paste them under directory yourProjectDirectory/node_modules/realm/vendor/realm-ios.

Crash Reporting

We encourage you to use a crash reporter in your application. Many Realm operations could potentially fail at runtime (like any other disk IO), so collecting crash reports from your application will help identify areas where either you (or us) can improve error handling and fix crashing bugs.

Most commercial crash reporters have the option of collecting logs. We strongly encourage you to enable this feature. Realm logs metadata information (but no user data) when throwing exceptions and in irrecoverable situations, and these messages can help debug when things go wrong.

Getting help

• Professional Support We offer professional support for Realm Database and Realm Platform. If you have a support agreement, you can reach us via our support portal.
• Need help with your code? Ask on StackOverflow. We actively monitor & answer questions on SO!
• Have a bug to report? Open an issue on our repo. If possible, include the version of Realm, a full log, the Realm file, and a project that shows the issue.
• Have a feature request? Open an issue on our repo. Tell us what the feature should do, and why you want the feature.
• Love to follow what comes up next? Look at our changelog. The log shows the latest additions and changes we plan to release soon, and the history of how Realm has evolved.