## This is forked from [twilio/twilio-voice](https://github.com/twilio/twilio-voice.js)

@twilio/voice-sdk
=================

[![Travis Build Status](https://travis-ci.com/twilio/twilio.js.svg?branch=master)](https://travis-ci.com/twilio/twilio-voice.js) [![NPM](https://img.shields.io/npm/v/%40twilio/voice-sdk.svg)](https://www.npmjs.com/package/%40twilio/voice-sdk)

Twilio's Voice SDK allows you to add real-time voice and PSTN calling to your web apps.

* [API Docs](https://twilio.github.io/twilio-voice.js/index.html)
* [More Docs](https://www.twilio.com/docs/voice/sdks/javascript)
* [Quickstart](https://www.twilio.com/docs/voice/client/javascript/quickstart)
* [Changelog](https://github.com/twilio/twilio-voice.js/blob/master/CHANGELOG.md)

### Technical Support
If you need technical support, contact
[help@twilio.com](mailto:help@twilio.com).

Installation
------------

### NPM

We recommend using `npm` to add the Voice SDK as a dependency.

```
npm install @twilio/voice-sdk --save
```

Using this method, you can `import` the Voice SDK using ES Module or TypeScript syntax:

```js
import { Device } from '@twilio/voice-sdk';

```

Or using CommonJS:

```js
const Device = require('@twilio/voice-sdk').Device;
```

### CDN
As of 2.0, the Twilio Voice SDK is no longer hosted via CDN.

### GitHub

Although we recommend using `npm` to add the Voice SDK as a dependency, you can also get the Twilio Voice SDK code
from GitHub and include it in your project directly. To do so, navigate to
["Releases"](https://github.com/twilio/twilio-voice.js/releases) and find the most recent release, or
the particular release version you'd like to use.

> **Note:** releases tagged with "-rc" are "Release Candidate" versions
> and are still being tested. Unless you specifically know that you want to use a release candidate
> version, you should not use a release with "-rc" in the name.

Download either the `zip` or the `tar.gz` and then extract the files. For example, if you downloaded the
tarball for the `2.0.0` release, you could then extract the files with the `tar` command:

```
tar -xvzf twilio-voice.js-2.0.0.tar.gz
cd twilio-voice.js-2.0.0
```

Once you've extracted the folder, the `twilio.js` and `twilio.min.js` files that
you can include in your project will be in the `/dist` directory. `twilio.min.js` is the
minified version of the code.

You can copy either the `twilio.js` or the `twilio.min.js` file into your project and
then provide a link to it in your html. For example:

```
<script type="text/javascript" src="twilio.min.js"></script>
```

Using this method, you can access the SDK through the browser global:

```
const Device = Twilio.Device;
```

Testing
-------

Running unit tests requires no setup aside from installation (above). You can run unit tests via:

```
npm run test:unit
```

Integration tests require some set up:

1. If the account you want to use doesn't already have a TwiML app set up, create one using the
   TwiML code below.
2. Copy config.example.yaml to config.yaml, replacing the placeholder information with valid credentials.

```xml
<?xml version="1.0" encoding="UTF-8"?>
<Response>
  <Dial>
    <Client>
      <Identity>{{To}}</Identity>
      <Parameter name="duplicate" value="12345" />
      <Parameter name="duplicate" value="123456" />
      <Parameter name="custom + param" value="我不吃蛋" />
      <Parameter name="foobar" value="some + value" />
      <Parameter name="custom1" value="{{Custom1}}" />
      <Parameter name="custom2" value="{{Custom2}}" />
      <Parameter name="custom3" value="{{Custom3}}" />
    </Client>
  </Dial>
</Response>
```

Integration tests can be run via:

```
npm run test:integration
```

These tests will run via karma, one at a time, in your system's default Chrome and then Firefox.

Network tests have been split out into their own docker processes, and can be run via

```
npm run test:docker
```

Content Security Policy (CSP)
----------------------------

Use the following policy directives to enable [CSP](https://developer.mozilla.org/en-US/docs/Web/HTTP/CSP) that is compatible with twilio-voice.js.

```
script-src https://media.twiliocdn.com https://sdk.twilio.com
media-src mediastream https://media.twiliocdn.com https://sdk.twilio.com
connect-src https://eventgw.twilio.com wss://chunderw-vpc-gll.twilio.com https://media.twiliocdn.com https://sdk.twilio.com
```

If you are providing a non-default value for `Device.ConnectOptions.edge` parameter, you need to add the Signaling URI `wss://chunderw-vpc-gll-{regionId}.twilio.com` in your `connect-src` directive where `regionId` is the `Region ID` as defined in this [page](https://www.twilio.com/docs/global-infrastructure/edge-locations/legacy-regions). See examples below.

**If `Device.ConnectOptions.edge` is `ashburn`**

```
connect-src https://eventgw.twilio.com https://media.twiliocdn.com https://sdk.twilio.com wss://chunderw-vpc-gll-us1.twilio.com
```

**If `Device.ConnectOptions.edge` is `['ashburn', 'sydney', 'roaming']`**

```
connect-src https://eventgw.twilio.com https://media.twiliocdn.com https://sdk.twilio.com wss://chunderw-vpc-gll-us1.twilio.com wss://chunderw-vpc-gll-au1.twilio.com wss://chunderw-vpc-gll.twilio.com
```

If you are providing a home region grant into your [Twilio access token](https://www.twilio.com/docs/iam/access-tokens), you need to add the insights endpoint in your `connect-src` directive using `eventgw.{homeRegion}.twilio.com` format. Below is an example if your home region grant is `sg1`.

```
connect-src https://eventgw.sg1.twilio.com wss://chunderw-vpc-gll.twilio.com https://media.twiliocdn.com https://sdk.twilio.com
```

License
-------

See [LICENSE.md](https://github.com/twilio/twilio-voice.js/blob/master/LICENSE.md)