DEPRECATED

See the following repos as a replacement

ember-firebase-adapter for Flexible Adapter and Serializer part
ember-firebaseui for FirebaseUI Component
ember-computed-query for hasFiltered relationship
No replacement for firebase-util service. Before, it was useful because of some power features but they've since been split over to the adapter side. What's left now of the service are merely just sugar syntax over firebase.

EmberFire Utilities

This addon provides some useful utilities on top of EmberFire.

Installation

ember install emberfire-utils

Your app needs to have EmberFire installed for this addon to work.

Configuration

You can optionally specify what libraries you'd want to exclude in your build within your ember-cli-build.js. Here's how:

var app = new EmberApp(defaults, {
  'emberfire-utils': {
    exclude: [ 'firebase-flex', 'firebase-util', 'firebase-ui' ],
  },
});

Possible exclusions are firebase-flex, firebase-util, and firebase-ui.

Flexible Adapter and Serializer

This is a standard Ember Data adapter that supports: createRecord(), destroyRecord(), findRecord(), findAll(), queryRecord(), and query(). However, its extended to allow some power features that's available to Firebase users.

Usage

Setup your application adapter like this:

// app/adapters/application.js
import FirebaseFlexAdapter from 'emberfire-utils/adapters/firebase-flex';

export default FirebaseFlexAdapter.extend();

Save and delete records with fan-out

// Saving a new record with fan-out
this.get('store').createRecord('post', {
  title: 'Foo',
  message: 'Bar'
}).save({
  adapterOptions: {
    include: {
      '/userFeeds/user_a/$id': true,
      '/userFeeds/user_b/$id': true,
    }
  }
});

// Deleting a record with fan-out
this.get('store').findRecord('post', 'post_a').then((post) => {
  post.deleteRecord();
  post.save({
    adapterOptions: {
      include: {
        '/userFeeds/user_a/post_a': null,
        '/userFeeds/user_b/post_a': null,
      }
    }
  });
});

// Alternatively, you can use `destroyRecord` with fan-out too
this.get('store').findRecord('post', 'post_a').then((post) => {
  post.destroyRecord({
    adapterOptions: {
      include: {
        '/userFeeds/user_a/post_a': null,
        '/userFeeds/user_b/post_a': null,
      }
    }
  });
});

Notice the $id. It's a keyword that will be replaced by the model's ID.

Save records with path

this.get('store').createRecord('comment', {
  title: 'Foo',
  message: 'Bar'
}).save({
  adapterOptions: { path: 'comments/post_a' }
});

Update only the changed attributes of a record

By default, only the changed attributes will be updated in Firebase whenever we call save(). This way, we can now have rules that doesn't allow some attributes to be edited.

Query records with path and infinite scrolling

The query params here uses the same format as the one in EmberFire with the addition of supporting the following:

orderBy: '.value'.
path to query the data from
isReference to know if the path is just a reference to a model in a different node (see example below)
cacheId to prevent duplicate listeners and make the query result array update in realtime
- Without cacheId, the query result array won't listen for child_added or child_removed changes. However, the models that are already inside of it will still update in realtime.
- cacheId isn't available in queryRecord.

With path

Let's assume the following data structure.

{
  "chats": {
    "one": {
      "title": "Historical Tech Pioneers",
      "lastMessage": "ghopper: Relay malfunction found. Cause: moth.",
      "timestamp": 1459361875666
    },
    "two": { ... },
    "three": { ... }
  },

  "members": {
    "one": {
      "ghopper": true,
      "alovelace": true,
      "eclarke": true
    },
    "two": { ... },
    "three": { ... }
  },

  "messages": {
    "one": {
      "m1": {
        "name": "eclarke",
        "message": "The relay seems to be malfunctioning.",
        "timestamp": 1459361875337
      },
      "m2": { ... },
      "m3": { ... }
    },
    "two": { ... },
    "three": { ... }
  },

  "users": {
    "ghopper": { ... },
    "alovelace": { ... },
    "eclarke": { ... }
  }
}

To fetch the chat members, you need to set the path and isReference. The isReference boolean indicates that the nodes under members/one are simply references to the user model which is represented by the users node.

this.get('store').query('user', {
  path: 'members/one',
  isReference: true,
  limitToFirst: 10
});

To fetch the chat messages, you just need to set the path and leave out the isReference. Without the isReference boolean, it indicates that the messages/one/m1, messages/one/m2, etc. are a direct representation of the message model.

this.get('store').query('message', {
  path: 'messages/one',
  limitToFirst: 10
});

With cacheId

this.get('store').query('post', {
  cacheId: 'my_cache_id',
  limitToFirst: 10
});

Infinite scrolling

this.get('store').query('post', {
  limitToFirst: 10
}).then((posts) => {
  posts.get('firebase').next(10);
});

Caveats

Relationship won't get updated when firing `save()`

As explained above, only the changed attributes will be saved when we call it. Ember Data currently doesn't provide a way to check if a relationship has changed. As a workaround, we need to fan-out the relationship to save it.

e.g.

const store = this.get('store');

store.findRecord('comment', 'another_comment').then((comment) => {
  store.findRecord('post', 'post_a').then((post) => {
    post.get('comments').addObject(comment);
    post.save({
      adapterOptions: {
        include: {
          'posts/post_a/comments/another_comment': '<some_value_here>'
        }
      }
    });
  });
});

However, there's a good side to this. Now we can provide different values to those relationships rather than the default true value in EmberFire.

`hasFiltered` relationship (not really a relationship)

Most of the time, we don't want to use the hasMany() relationship in our models because:

It's not flexible enough to fetch from paths we want.
It loads all the data when we access it.
Even if we don't access it, those array of IDs are still taking up internet data usage.

To solve those 2 problems above, use hasFiltered() relationship. It has the same parameters as store.query() and it also works with infinite scrolling as explained above.

// app/models/post
import Model from 'ember-data/model';
import attr from 'ember-data/attr';

import hasFiltered from 'emberfire-utils/utils/has-filtered';

export default Model.extend({
  title: attr('string'),

  _innerReferencePath: attr('string'),

  comments: hasFiltered('comment', {
    cacheId: '$id_comments',
    path: '/comments/$innerReferencePath/$id',
    limitToFirst: 10,
  })
});

Notice the following:

$id
- This is a keyword that will be replaced by the model's ID.
- This works for both cacheId and path.
_innerReferencePath
- This will be replaced by the inner Firebase reference path of the model.
- If post model lives in /posts/forum_a/post_a, the value would be forum_a.
  - Another example, /posts/foo/bar/post_a -> foo/bar.
- This is a client-side only property. It won't be persisted in the DB when you save the record.
$innerReferencePath
- This is a keyword that will be replaced by _innerReferencePath.
- This only works for path.
- Won't work when _innerReferencePath isn't defined.
- This is useful for when let's say your comments lives in /comments/<forum_id>/<post_id> and you know the value of the <post_id> through $id but don't know the value of <forum_id>.

hasFiltered() are read only.

Utility Service

Usage

Simply inject the firebase-util service.

Multi-path updates

To write on multiple paths atomically in Firebase, call update().

const fanoutObject = {};

fanoutObject['users/foo/firstName'] = 'Foo';
fanoutObject['users/bar/firstName'] = 'Bar';

this.get('firebaseUtil').update(fanoutObject).then(() => {
  // Do something after a succesful update
}).catch(error => {
  // Do something with `error`
});

Generate Firebase push ID

Should you need to generate a Firebase push ID for your multi-path updates, you can use generateIdForRecord(). This returns a unique ID generated by Firebase's push() method.

const pushId = this.get('firebaseUtil').generateIdForRecord();

Storage manipulations

Uploading a file to Firebase Storage

To upload files in Firebase storage, call uploadFile().

function onStateChange(snapshot) {
  const progress = (snapshot.bytesTransferred / snapshot.totalBytes) * 100;

  console.log('Upload is ' + progress + '% done');
}

this.get('firebaseUtil').uploadFile('images/foo.jpg', file, metadata, onStateChange).then(downloadURL => {
  // Do something with `downloadURL`
}).catch(error => {
  // Do something with `error`
});

file should be a Blob or a Uint8Array.
metadata and onStateChange are optional params.

Deleting a file in Firebase Storage

To delete files in Firebase storage, call deleteFile().

this.get('firebaseUtil').deleteFile(url).then(() => {
  // Do something on success
}).catch(error => {
  // Do something with `error`
});

url should be the HTTPS URL representation of the file. e.g. https://firebasestorage.googleapis.com/b/bucket/o/images%20stars.jpg

Queries for non-model data

For the examples below, assume we have the following Firebase data:

{
  "users" : {
    "foo" : {
      "photoURL" : "foo.jpg",
      "username" : "bar"
    },
    
    "hello" : {
      "photoURL" : "hello.jpg",
      "username" : "world"
    }
  }
}

Query a single record

To query a single record, call queryRecord(). This will return a promise that fulfills with the requested record in a plain object format.

this.get('firebaseUtil').queryRecord('users', { equalTo: 'foo' }).then((record) => {
  // Do something with `record`
}).catch(error => {
  // Do something with `error`
});

Params:

path - Firebase path
options - An object that can contain the following:
- cacheId - Prevents duplicate listeners and returns cached record if it already exists. When not provided, Firebase won't listen for changes returned by this function.
- EmberFire queries with the addition of .value for orderBy and forcing of limitToFirst or limitToLast to 1.

limitToFirst and limitToLast is forced to 1 because this method will only return a single record. If you provided an option of limitToFirst, it will set it to 1 regardless of the value that you've set. Same goes for limitToLast respectively.

Query multiple records

To query for multiple records, call query(). This will return a promise that fulfills with the requested records; each one in a plain object format.

this.get('firebaseUtil').query('users', { limitToFirst: 10 }).then((records) => {
  // Do something with `records`
}).catch(error => {
  // Do something with `error`
});

Params:

path - Firebase path
options - An object that can contain the following:
- cacheId - Prevents duplicate listeners and returns cached record if it already exists. When not provided, Firebase won't listen for changes returned by this function.
- EmberFire queries with the addition of .value for orderBy.

Serialized to plain objects

For queryRecord() and query(), the records are serialized in plain object. For the queryRecord() example above, the record will be serialized to:

record = {
  id: 'foo',
  photoURL: 'foo.jpg',
  username: 'bar'
};

For query():

records = [{
  id: 'foo',
  photoURL: 'foo.jpg',
  username: 'bar'
}, {
  id: 'hello',
  photoURL: 'hello.jpg',
  username: 'world'
}];

Should we retrieve a record who's value isn't an object (e.g. users/foo/username), the record will be serialized to:

record = {
  id: 'username',
  value: 'bar'
};

Loading more records for `query()`

To load more records in the query() result, call next().

const firebaseUtil = this.get('firebaseUtil');

firebaseUtil.query('users', {
  cacheId: 'cache_id',
  limitToFirst: 10,
}).then(() => {
  firebaseUtil.next('cache_id', 10);
});

Checking if record exists

To check if a record exists, call isRecordExisting(). This returns a promise that fulfills to true if the record exists. Otherwise, false.

this.get('firebaseUtil').isRecordExisting('users/foo').then((result) => {
  // Do something with `result`
}).catch(error => {
  // Do something with `error`
});

FirebaseUI

Auth

A component is provided for rendering FirebaseUI Auth. Here's how:

First setup your uiConfig which is exactly the same with Firebase UI Auth.

import firebase from 'firebase';
import firebaseui from 'firebaseui';

let uiConfig = {
  credentialHelper: firebaseui.auth.CredentialHelper.NONE,
  signInSuccessUrl: '<url-to-redirect-to-on-success>',
  signInOptions: [
    firebase.auth.GoogleAuthProvider.PROVIDER_ID,
    firebase.auth.FacebookAuthProvider.PROVIDER_ID,
    firebase.auth.TwitterAuthProvider.PROVIDER_ID,
    firebase.auth.GithubAuthProvider.PROVIDER_ID,
    firebase.auth.EmailAuthProvider.PROVIDER_ID
  ],
};

Then pass that uiConfig into the firebase-ui-auth component.

{{firebase-ui-auth uiConfig=uiConfig}}

Compatibility

This addon is compatible with EmberFire 2.0.x.

Contributing

Installation

git clone <repository-url> this repository
cd emberfire-utils
npm install

Running

ember serve
Visit your app at http://localhost:4200.

Running Tests

npm test (Runs ember try:each to test your addon against multiple Ember versions)
ember test
ember test --server

Building

ember build

For more information on using ember-cli, visit https://ember-cli.com/.

Name		Name	Last commit message	Last commit date
Latest commit History 170 Commits
addon		addon
app		app
config		config
tests		tests
vendor		vendor
.editorconfig		.editorconfig
.ember-cli		.ember-cli
.eslintignore		.eslintignore
.eslintrc.js		.eslintrc.js
.gitignore		.gitignore
.npmignore		.npmignore
.travis.yml		.travis.yml
.watchmanconfig		.watchmanconfig
LICENSE.md		LICENSE.md
README.md		README.md
ember-cli-build.js		ember-cli-build.js
index.js		index.js
package.json		package.json
testem.js		testem.js
yarn.lock		yarn.lock

Folders and files

Latest commit

History

Repository files navigation

DEPRECATED

EmberFire Utilities

Installation

Features

Configuration

Flexible Adapter and Serializer

Usage

Save and delete records with fan-out

Save records with path

Update only the changed attributes of a record

Query records with path and infinite scrolling

With path

With cacheId

Infinite scrolling

Caveats

Relationship won't get updated when firing save()

hasFiltered relationship (not really a relationship)

Utility Service

Usage

Multi-path updates

Generate Firebase push ID

Storage manipulations

Uploading a file to Firebase Storage

Deleting a file in Firebase Storage

Queries for non-model data

Query a single record

Query multiple records

Serialized to plain objects

Loading more records for query()

Checking if record exists

FirebaseUI

Auth

Compatibility

Contributing

Installation

Running

Running Tests

Building

About

Resources

License

Uh oh!

Stars

Watchers

Forks

Releases 11

Packages 0

Uh oh!

Uh oh!

Contributors

Uh oh!

Languages

Relationship won't get updated when firing `save()`

`hasFiltered` relationship (not really a relationship)

Loading more records for `query()`

Packages