A JavaScript library for sipgate.io

• Installation
• Available Functionality
  • SMS
  • Fax
  • Call
  • Webhook (node.js only)
  • Webhook Settings
  • Fluent interface
  • Contacts
  • History
  • Numbers
  • Real Time Call Manipulation (RTCM)
• Usage
  • Creating a Client
  • SMS
  • Fax
  • Call
  • Webhooks
  • Webhook Settings
  • Fluent interface
  • Contacts
  • History
  • Numbers
  • Real Time Call Manipulation (RTCM)
  • Voicemail
  • Device
• Examples
• Changelog
• Privacy Note

For use in node applications you can install with

npm install sipgateio --save

Alternatively, a bundled version can be obtained from the github releases page.
Or use a CDN like jsDelivr instead:
https://cdn.jsdelivr.net/gh/sipgate-io/sipgateio-node@latest/bundle/sipgate-io.min.js

The following features are already implemented in the current version of this library:

Send text messages, either instantly or scheduled. The caller ID can be set from the sipgate web interface, the default is the string "sipgate".

Send any PDF file as a fax to a single number.

Initiate a call between two phones of your choice – no matter if inside your sipgate account or outside.

Set up a webserver to process real-time call data from sipgate.io.

Configure the webhook functionality of sipgate.io. Currently, you can set URLs and whitelist extensions for triggering webhooks as well as toggle the debug log.

The fluent interface module is a wrapper for the webhook module to simplify its usage.

Import contacts in CSV format into your sipgate account.

Fetch multiple or a specific event from your history.

Receive all numbers assigned to your account.

Manipulate present calls that containts playing announcements with own audio-files. Also you can hangup, hold, mute an exist call and send DTMF singals.

You can connect the client by passing a valid OAuth token (You have to implement the OAuth flow yourself). You can find more information on our API documentation.:

const client = sipgateIO({
	token: '<valid Token>',
});

As alternative you can use personal access tokens to authenticate. You can find more information on our API documentation.

const client = sipgateIO({
	tokenId: '<your tokenId>',
	token: '<your personal access token>',
});

Possible authentication objects:

interface OAuthCredentials {
	token: string;
}

interface PersonalAccessTokenCredentials {
	tokenId: string;
	token: string;
}

The sipgateIO method accepts your valid sipgate credentials as defined in the AuthCredentials type and returns a sipgate.io Client. The client contains as members the supported modules (sms, fax, call, etc.).

The client provides the following functions:

async function getAuthenticatedWebuserId(): Promise<string>;
async function getWebUsers(): Promise<Webuser[]>;

The getAuthenticatedWebuserId() method returns the ID of the admin account.

The getWebUsers() method returns an array of the following type:

export interface Webuser {
	id: string;
	firstname: string;
	lastname: string;
	email: string;
	defaultDevice: string;
	busyOnBusy: string;
	addressId: string;
	directDialIds: string[];
	timezone: string;
	admin: string;
}

Note that creating the sipgate.io Client will not validate your credentials, you might want to call the method client.getAuthenticatedWebuserId() to check that your credentials are correct.

The SMS module provides the following functions:

async function send(sms: ShortMessage): Promise<void>;
async function schedule(sms: ShortMessage, sendAt: Date): Promise<void>;
async function getSmsExtensions(webuserId: string): Promise<SmsExtension[]>;

Note: sendAt can be 30 days in advance at max.

Note: you should be aware that the request will take a short time to be processed. Values for sendAt should not just be a few seconds in the future. If sendAt is in the past an error will be thrown.

The ShortMessage type requires the following fields:

interface ShortMessage {
	smsId: string;
	to: string;
	message: string;
}

or

interface ShortMessage {
	from: string;
	to: string;
	message: string;
}

The getSmsExtensions method returns an array of the following type:

export interface SmsExtension {
	id: string;
	alias: string;
	callerId: string;
}

The fax module provides the following functions:

async function send(fax: Fax): Promise<SendFaxSessionResponse>;
async function getFaxStatus(sessionId: string): Promise<FaxStatus>;
async function getFaxlines(): Promise<Faxline[]>;
async function getFaxlinesByWebUser(webuserId: string): Promise<Faxline[]>;

The send function allows you to send a fax by passing an object with the following fields:

interface Fax {
	to: string;
	fileContent: Buffer;
	filename?: string;
	faxlineId?: string;
}

and returns an SendFaxSessionResponse:

interface SendFaxSessionResponse {
	sessionId: string;
}

By using getFaxStatus and passing the sessionId you received from the send function, you will receive one of the following values:

enum FaxStatus {
	SENT = 'SENT',
	PENDING = 'PENDING',
	FAILED = 'FAILED',
	SENDING = 'SENDING',
	SCHEDULED = 'SCHEDULED',
}

With the getFaxlines function you can get a list of faxlines that belong to the currently authenticated user. It returns an array of the following type:

export interface Faxline {
	id: string;
	alias: string;
	tagline: string;
	canSend: boolean;
	canReceive: boolean;
}

With the getFaxlinesByWebUser function you can get a list of faxlines that belong to the specified user. It also returns an array of Faxline.

The call module provides the following function:

async function initiate(
	newCallRequest: CallData
): Promise<InitiateNewCallSessionResponse>;

The CallData contains the following fields:

interface CallData {
	deviceId?: string;
	from: string;
	to: string;
	callerId?: string;
}

The InitiateNewCallSessionResponse contains only a session ID:

interface InitiateNewCallSessionResponse {
	sessionId: string;
}

The following table shows valid parameter combinations

The displayed number at the callee device is determined by a hierarchy:

If not set it falls back to the next stage:

1. callerId
2. deviceId (related phone number)
3. from (related phone number if from is an extension)

The param deviceId is only mandatory if from is not an extension.

Valid extension types are e, p, x and y.

Scenario 1: basic call

const call = createCallModule(client);
const callData = {
	from: 'e14',
	to: '+4921165432',
};

call.initiate(callData);

Behavior: The phone with extension e14 rings first, after pick-up the callee is called. The default number associated with e14 is displayed at the callee device.

Scenario 2: custom caller id

const callData = {
	from: 'p0',
	to: '+4921165432',
	callerId: '+4917012345678',
};

call.initiate(callData);

Behavior: Same situation as previous example, but displayed number is now callerId (see hierarchy).

Scenario 3: group call

const callData = {
	from: '+4921123456',
	deviceId: 'e14',
	to: '+4921165432',
};

call.initiate(callData);

If the from number refers to a group of phones rather than a single one all phones in the group will ring and the first to be picked up will establish the call with the to.

The deviceId is needed for billing and determines the number which will be displayed at the to device. For instance, e14 has the default number '+4921156789'.

Please note: The webhook feature is only available in node.js environments.

The Webhook API provides processing of real-time call data.
A webhook is a POST request that sipgate.io makes to a predefined URL when a certain event occurs. These requests contain information about the event that occurred in application/x-www-form-urlencoded format.

The webhook module provides a simple means to set up a server for handling these webhooks.

The following types of events can trigger webhooks:

For any of those events, a callback function can be registered to be called upon receiving the respective webhook.

Additionally, for the types NEW_CALL and DATA a response may be returned containing commands to trigger actions like hanging up or redirecting calls.

For generating that response our library provides a convenient response interface.

To begin, instantiate the webhook module by calling createWebhookModule. The resulting object provides only one method, createServer which takes a configuration object of type ServerOptions.

export interface ServerOptions {
	// the local port to listen on
	port: number;
	// the publicly accessible server address
	// (including the port, if not standard)
	serverAddress: string;
	// an optional hostname (default: localhost)
	hostname?: string;
	// you can optionally skip the signature verification
	skipSignatureVerification?: boolean;
}

It returns a Promise<WebhookServer> which, when resolved, provides the following methods:

interface WebhookServer {
	onNewCall: (fn: HandlerCallback<NewCallEvent, ResponseObject | void>) => void;
	onAnswer: (fn: HandlerCallback<AnswerEvent, void>) => void;
	onHangup: (fn: HandlerCallback<HangUpEvent, void>) => void;
	onData: (fn: HandlerCallback<DataEvent, ResponseObject | void>) => void;
	stop: () => void;
}

The stop method simply kills the server, the other methods each take a callback function for handling the respective types of events suggested by their name.

Webhooks are verified to ensure that they are sent by sipgate.
You can optionally disable this via the skipSignatureVerification flag.

Additionaly you can check that webhooks are originated from push-api.sipgate.net (217.116.118.254 or 212.9.46.32).

Each of the four callback registration methods takes a single callback function that accepts a webhook object of the respective type (i.e. NewCallEvent, AnswerEvent, HangUpEvent, or DataEvent). In the case of onNewCall and onData the provided function may return a ResponseObject (details below)

Within the callback function the following fields are accessible:

The NewCallEvent type offers the following fields:

the AnswerEvent type offers the following fields:

the HangUpEvent type offers the following fields:

Hangups can occur due to these causes:

the DataEvent type offers the following fields:

For composing an XML response from withing a callback function our library offers a convenient response interface:

interface WebhookResponseInterface {
	redirectCall: (redirectOptions: RedirectOptions) => RedirectObject;
	sendToVoicemail: () => VoicemailObject;
	rejectCall: (rejectOptions: RejectOptions) => RejectObject;
	playAudio: (playOptions: PlayOptions) => PlayObject;
	playAudioAndTransfer: (
		playOptions: PlayOptions,
		transferOptions: TransferOptions,
		client: SipgateIOClient,
		callId: string,
		timeout?: number
	) => Promise<PlayObject>;
	playAudioAndHangUp: (
		playOptions: PlayOptions,
		client: SipgateIOClient,
		callId: string,
		timeout?: number
	) => Promise<PlayObject>;
	gatherDTMF: (gatherOptions: GatherOptions) => GatherObject;
	hangUpCall: () => HangUpObject;
}

The redirectCall method accepts an options object of type RedirectOptions with the following fields:

type RedirectOptions = {
	numbers: string[];
	anonymous?: boolean;
	callerId?: string;
};

The sendToVoicemail method accepts no further options.

The rejectCall method accepts an options object of type RejectOptions with a single field, the reason for rejecting the call. This reason may be one of the following:

enum RejectReason {
	BUSY = 'busy',
	REJECTED = 'rejected',
}

The playAudio method accepts an options object of type PlayOptions with a single field, the URL to a sound file to be played.

Note: Currently the sound file needs to be a mono 16bit PCM WAV file with a sampling rate of 8kHz. You can use conversion tools like the open source audio editor Audacity to convert any sound file to the correct format.

Linux users might want to use mpg123 to convert the file:

mpg123 --rate 8000 --mono -w output.wav input.mp3

The playAudioAndHangUp method accepts an options object of type PlayOptions with a single field, the URL to a sound file to be played. In addition, this also requires a sipgateIOClient, a unique callId from an current active call and a timeout which is optional.

After the audio file has been played and the additional timeout has expired, the call is terminated based on the callId.

Note: For any information about the audio file please look at play audio.

The playAudioAndTransfer method accepts an options object of type PlayOptions with a single field, the URL to a sound file to be played, and an object of type TransferOptions, which contains the phone number to which the call should be transferred. In addition, this also requires a sipgateIOClient, a unique callId from an current active call and a timeout which is optional.

After the audio file has been played and the additional timeout has expired, the call is terminated based on the callId.

Note: For any information about the audio file please look at play audio.

The gatherDTMF method accepts an options object of type GatherOptions with the following fields:

type GatherOptions = {
	announcement?: string;
	maxDigits: number;
	timeout: number;
};

maxDigits (> = 1) specifies to maximum number of DTMF tones to be gathered , the timeout (> = 0) is the period in milliseconds to wait for DTMF input from a caller before processing. Please note that the establishment of the call is delayed until this period has elapsed. By specifying a URL to a sound file as announcement an audio message can be played to inform callers what DTMF tones they should send.

Note: Please consider the above restrictions concerning the format of the announcement file.

The hangUpCall method accepts no further options.

The webhook settings module provides the following functions to update settings:

async function setIncomingUrl(url: string): Promise<void>;
async function clearIncomingUrl(): Promise<void>;

These two functions allow for the setting and clearing of the URL to be called when a webhook is triggered by an incoming call.

async function setOutgoingUrl(url: string): Promise<void>;
async function clearOutgoingUrl(): Promise<void>;

Analogous functions exist for the URL that handles outgoing calls.

async function setWhitelist(extensions: string[]): Promise<void>;
async function clearWhitelist(): Promise<void>;
async function disableWhitelist(): Promise<void>;

The whitelist specifies extensions that should trigger webhooks. By default, webhooks are enabled for all phoneline and group extensions. This behavior is restored by calling disableWhitelist. The disableWhitelist completely removes the whitelisting and enables all phoneline and group extensions.

async function setLog(value: boolean): Promise<void>;

The setLog function toggles, the function to display all incoming and outgoing events, which have been sent to your Incoming and Outgoing Url. These parameters can be set using these functions: setIncomingUrl and setOutgoingUrl.

interface WebhookSettings {
	incomingUrl: string;
	outgoingUrl: string;
	log: boolean;
	whitelist: string[] | null;
}

async function getWebhookSettings(): Promise<WebhookSettings>;

The getWebhookSettings function returns you the current settings of sipgate.io including the incoming/outgoing URL, whether logging is enabled or if a specific whitelist of devices is set for your incoming & outgoing URL.

The fluent interface module currently provides a wrapper for the following functions of the webhook module:

• port
• serverAddress
• AnswerCallback
• DataCallback
• HangUpCallback
• NewCallCallback
• startServer

Example:

new FluentWebhookServer()
	.setServerPort(port)
	.setServerAddress(serverAddress)
	.setOnNewCallListener((newCallEvent) => {
		console.log(`New call from ${newCallEvent.from} 	to ${newCallEvent.to}`);
	})
	.setOnAnswerListener((answerEvent) => {
		console.log(`Answer from: ${answerEvent.from}`);
	})
	.setOnHangupListener((hangupEvent) => {
		console.log(`Hangup with cause: ${hangupEvent.cause}`);
	})
	.setOnDataListener((dataEvent) => {
		console.log(`Data from Call: ${dataEvent.originalCallId}`);
	})
	.startServer();

The contacts module provides the following functions:

interface ContactImport {
	firstname: string;
	lastname: string;
	address?: Address;
	phone?: PhoneNumber;
	email?: Email;
	organization?: string[];
}

type ContactUpdate = ContactResponse;

interface ContactResponse {
	id: string;
	name: string;
	picture: string;
	emails: Email[];
	numbers: PhoneNumber[];
	addresses: Address[];
	organization: string[][];
	scope: Scope;
}

interface ContactsModule {
	create: (contact: ContactImport, scope: Scope) => Promise<void>;
	update: (contact: ContactUpdate) => Promise<void>;
	delete: (id: string) => Promise<void>;
	deleteAllPrivate: () => Promise<void>;
	deleteAllShared: () => Promise<void>;
	importFromCsvString: (csvContent: string) => Promise<void>;
	importVCardString: (vcardContent: string, scope: Scope) => Promise<void>;
	exportAsCsv: (
		scope: ExportScope,
		delimiter?: string,
		pagination?: Pagination,
		filter?: ContactsExportFilter
	) => Promise<string>;
	exportAsVCards: (
		scope: ExportScope,
		pagination?: Pagination,
		filter?: ContactsExportFilter
	) => Promise<string[]>;
	exportAsSingleVCard: (
		scope: ExportScope,
		pagination?: Pagination,
		filter?: ContactsExportFilter
	) => Promise<string>;
	get: (
		scope: ExportScope,
		pagination?: Pagination,
		filter?: ContactsExportFilter
	) => Promise<ContactResponse[]>;
}

Given a ContactImport and Scope, creates a contact.

Takes a ContactResponse, the output of ContactsModule.get, and updates the corresponding contact.

Deletes a contact by a given id.

Deletes all private contacts.

Deletes all shared contacts.

It takes a valid ContactImport Object and creates a Contact in the requested Scope.

It takes a valid CSV-formatted string (columns separated by ",") containing at least the following fields:

• firstname
• lastname
• number

These fields may be provided in an arbitrary order. Additional fields as well as empty lines will be ignored. Empty records (i.e. just separators) produce a warning but no error. The same is true for strings containing only the header row.

Example:

lastname,firstname,number
Turing,Alan,+4921163553355
Lovelace,Ada,+4921163553355

It takes a valid VCard 4.0 string, containing at least the following fields:

• name contains firstname and lastname
• number

It returns a csv strings containing all contacts for the given scope.
You can also add a specific delimiter for the csv format.

Note: using a filter will ignore pagination

It returns JSON objects containing all contacts for the given scope.

Note: using a filter will ignore pagination

It returns mulitple vCard-strings containing all contacts for the given scope

Note: using a filter will ignore pagination

It returns a vCard-address-book containing all contacts for the given scope

Note: using a filter will ignore pagination

It returns a list of contacts for the given scope as described in the following interface.

Note: using a filter will ignore pagination

interface ContactRequest {
	id: string;
	name: string;
	picture: string;
	emails: { email: string; type: string[] }[];
	numbers: { number: string; type: string[] }[];
	addresses: Address[];
	organization: string[][];
	scope: Scope;
}

The PRIVATE Scope contains all contacts created by yourself and not shared with other people.

The SHARED Scope includes all contacts created by anyone in your organization and are therefore shared with other people.

The INTERNAL Scope contains the contacts which are created by sipgate such as a contact for any webuser in your organization.

Adress and Numbers:

You can only save one address and one number using the Format.

The history module provides functionality to fetch all or specific history events.

interface HistoryModule {
	fetchAll: (
		filter?: HistoryFilter,
		pagination?: Pagination
	) => Promise<HistoryEntry[]>;
	fetchById: (entryId: string) => Promise<HistoryEntry>;
	deleteByListOfIds: (entryIds: string[]) => Promise<void>;
	deleteById: (entryId: string) => Promise<void>;
	batchUpdateEvents: (
		events: HistoryEntry[],
		callback: (entry: HistoryEntry) => HistoryEntryUpdateOptions
	) => Promise<void>;
	exportAsCsvString: (
		filter?: BaseHistoryFilter,
		pagination?: Pagination
	) => Promise<string>;
}

The fetchAll method can filter the result by using the 'HistoryFilter' interface. You can decide how many history events you receive by adjusting the values in the pagination object.

interface HistoryFilter {
	connectionIds?: string[];
	types?: HistoryEntryType[];
	directions?: Direction[];
	archived?: boolean;
	starred?: boolean;
	from?: Date;
	to?: Date;
	phonenumber?: string;
}

interface Pagination {
	offset?: number;
	limit?: number;
}

fetchById and fetchAll methods returns one or multiple history events described by the following base-structure:

interface BaseHistoryEntry {
	id: string;
	source: string;
	target: string;
	sourceAlias: string;
	targetAlias: string;
	type: HistoryEntryType;
	created: Date;
	lastModified: Date;
	direction: Direction;
	incoming: boolean;
	status: string;
	connectionIds: string[];
	read: boolean;
	archived: boolean;
	note: string;
	endpoints: RoutedEndpoint[];
	starred: boolean;
	labels: string[];
}

There are multiple event-types, such as:
CallHistoryEntry, FaxHistoryEntry, SmsHistoryEntry, VoicemailHistoryEntry.
A more detailed description of these types can be found here.

You have to pass events, which should be updated, to the batchUpdateEvents function. The second parameter is a mapping function which gives you the ability to return HistoryEntryUpdateOptions for every event which is used to set the specified parameters in the history entry.

export interface HistoryEntryUpdateOptions {
	archived?: boolean;
	starred?: boolean;
	note?: string;
	read?: boolean;
}

The deleteById method allows you to delete an history entry with the given id.

The deleteByListOfIds method allows you to delete multiple history entries by a given list of ids.

The exportAsCsvString method allows you to export your history entries as a csv string. Optionally you can filter and paginate the response by using the following parameters:

export interface BaseHistoryFilter {
	connectionIds?: string[];
	types?: HistoryEntryType[];
	directions?: HistoryDirection[];
	archived?: boolean;
	starred?: boolean;
	startDate?: Date;
	endDate?: Date;
}

interface Pagination {
	offset?: number;
	limit?: number;
}

The numbers module provides the following function:

interface NumbersModule {
	getAllNumbers: (pagination?: Pagination) => Promise<NumberResponseItem[]>;
}

The method returns an array of objects with the type NumberResponseItem, which has the following structure:

interface NumberResponseItem {
	id: string;
	number: string;
	localized: string;
	type: NumberResponseItemType;
	endpointId: string;
	endpointAlias: string;
	endpointUrl: string;
	mnpState?: NumberMnpState;
	portId?: number;
}

Where NumberResponseItemType is one value out of the following enum:

enum NumberResponseItemType {
	MOBILE = 'MOBILE',
	LANDLINE = 'LANDLINE',
	QUICKDIAL = 'QUICKDIAL',
	INTERNATIONAL = 'INTERNATIONAL',
}

The mnpState of type NumberMnpState only appears in a mobile number entry and looks like this:

interface NumberMnpState {
	isReleased: boolean;
	releasedUntil: Date;
}

The real time call manipulation module provides the following functions:

interface RTCMModule {
	getEstablishedCalls: () => Promise<RTCMCall[]>;
	mute: (call: RTCMCall, status: boolean) => Promise<void>;
	record: (call: RTCMCall, recordOptions: RecordOptions) => Promise<void>;
	announce: (call: RTCMCall, announcement: string) => Promise<void>;
	transfer: (call: RTCMCall, transferOptions: TransferOptions) => Promise<void>;
	sendDTMF: (call: RTCMCall, sequence: string) => Promise<void>;
	hold: (call: RTCMCall, status: boolean) => Promise<void>;
	hangUp: (call: RTCMCall) => Promise<void>;
}

The structure of a present call is provided by a RTCMCall and containts the following attributes:

interface RTCMCall {
	callId: string;
	muted: boolean;
	recording: boolean;
	hold: boolean;
	participants: Participant[];
}

interface Participant {
	participantId: string;
	phoneNumber: string;
	muted: boolean;
	hold: boolean;
	owner: boolean;
}

It returns all present calls of this account and return an array of RTCMCall.

You can pass a RTCMCall and set your microphone muted or unmuted.

You can start or stop a recording and find this later in the history entry.

You can play an audiofile that needs to be a mono 16bit PCM WAV file with a sampling rate of 8kHz. Insert the URL of this audiofile as parameter in the announcment.

You can attend a call and transfer it to a phonenumber with the following structure:

interface TransferOptions {
	attended: boolean;
	phoneNumber: string;
}

You can send a sequence of valid DTMFs digits to the passed call.

You can hold and continue a present call.

You can abort or terminate a current call.

The voicemail module provides the following functions:

async function getVoicemails(): Promise<Voicemail[]>;

The getVoicemails method returns an array of the following type:

export interface Voicemail {
	id: string;
	alias: string;
	belongsToEndpoint: {
		extension: string;
		type: string;
	};
}

The device module provides the following functions:

async function getDevices(webuserId: string): Promise<Device[]>;

The getDevices method returns an array of the following type:

export type DeviceType = 'REGISTER' | 'MOBILE' | 'EXTERNAL';

export interface ActiveRouting {
	id: string;
	alias: string;
}

export interface Device {
	id: string;
	alias: string;
	type: DeviceType;
	dnd: boolean;
	online: boolean;
	callerId?: string;
	owner?: string;
	activePhonelines: ActiveRouting[];
	activeGroups: ActiveRouting[];
}

For some examples on how to use the library, please refer to this repository: sipgateio-node-examples

npx ts-node some_example.ts

You can find the changelog here.

This library sets the following headers for every request made to the sipgate REST-API to obtain statistics about versions currently in use and to differentiate sipgate.io-users from sipgate software:

• X-Sipgate-Client
• X-Sipgate-Version