README

A simple JavaScript wrapper for the PagerDuty APIs.

• Supports Node and Browser environments
• Supports REST and Events v2 APIs.
• Supports both offset and cursor based pagination

For full API Reference see this page.

Installation

npm install --save @pagerduty/pdjs

Usage

REST API

REST API calls can be done using the convenience methods or by passing in a url or endpoint.

Convenience Methods

There are some simple convience methods. get(), post(), put(), and delete().

import {api} from '@pagerduty/pdjs';

const pd = api({token: 'someToken1234567890'});

pd.get('/incidents')
  .then({data, resource, next} => console.log(data, resource, next))
  .catch(console.error);

// Similarly, for `post`, `put`, `patch` and `delete`.
pd.post('/incidents', { data: { ... } }).then(...);

// If you need to supply headers:
pd.post('/incidents/id/status_updates', {
  headers: {
    From: 'user@example.com',
  },
  data: { ... },
}).then();

tokenType

Allows you to set either token or bearer tokens. Defaults to token but provides ability to use bearer as well.

• Tokens are generated in your PagerDuty account.
• Bearer tokens are generated through an OAuth 2.0 authorization flow. For example, to use a Bearer token when initializing api you'll pass int the tokenType parameter, like so:

const pd = api({token: 'someBearerToken1234567890', tokenType: 'bearer'});

url or endpoint

// Calling the returned function with a `endpoint` or `url` will also send it.
pd({
  method: 'post',
  endpoint: '/incidents',
  data: {
    ...
  }
}).then(...)

The Response Object

The PD object always returns an APIResponse object which contains some raw data as well as a convenient shortcut to directly access the returns Resource.

pd.get('/incidents')
  .then({resource, data, response, next} => {
    console.log(resource); // Contains the data of resource request. In this example the 'incidents' data.
    console.log(data);     // The raw data returned from the API, also contains pagination information.
    console.log(response); // The response object returned from the cross-fetch
    console.log(next);     // A convenience function to help with pagination
  })
  .catch(console.error);

Pagination

There's an convience method all that attempts to fetch all pages for a given endpoint and set of parameters. For convenience this function supports both offset and cursor based pagination.

Note that the PagerDuty API has a limit for most endpoints and recommends using parameters to refine searches where more results are necessary. More information can be found in the Developer Documentation.

The response object of an API call also contains a nextFunc which can be used to define your own Pagination function if you feel included.

import {api} from '@pagerduty/pdjs';

const pd = api({token: 'someToken1234567890'});

pd.all('/incidents')
  .then({data, resource} => console.log(data, resource))
  .catch(console.error);

Options

The API interface allows for some extra parameters to be included.

server

To use this library with a different service region use this parameter to change the root url of requests. Default: api.pagerduty.com.

pd({
  method: 'post',
  endpoint: '/incidents',
  server: 'api.eu.pagerduty.com',
  data: {
    ...
  }
}).then(...)

headers

Some endpoints require the setting of extra headers such as a From header.

pd({
  method: 'post',
  endpoint: '/incidents',
  headers: {
    'From': "me@example.com"
  },
  data: {
    ...
  }
}).then(...)

Retries

There is some very simple retry logic baked into each request in the case that the PagerDuty API rate limits your requests (only when it responds HTTP Code 429). Requests will retry 3 times waiting 20 seconds between each request. If the request is still being rate limited after 3 attempts the client will simply return the 429 response.

Events API

Events V2 is supported along with Change Events.

import {event} from '@pagerduty/pdjs';
event({
  data: {
    routing_key: 'YOUR_ROUTING_KEY',
    event_action: 'trigger',
    dedup_key: 'test_incident_2_88f520',
    payload: {
      summary: 'Test Event V2',
      source: 'test-source',
      severity: 'error',
    },
  },
})
  .then(console.log)
  .catch(console.error);

Convenience Methods

import {change, trigger, acknowledge, resolve} from '@pagerduty/pdjs';

change({
    "routing_key": "YOUR_ROUTING_KEY",
    "payload": {
      "summary": "Build Success:!",
      "timestamp": "2015-07-17T08:42:58.315+0000",
      "source": "prod-build-agent",
      "custom_details": {
        "build_state": "passed",
        "build_number": "220",
        "run_time": "1337s"
      }
    },
    "links": [{
      "href": "https://buildpipeline.com",
      "text": "View in Build Pipeline"
    }]
})
  .then(console.log)
  .catch(console.error);

trigger({...})
  .then(console.log)
  .catch(console.error);
acknowledge({...})
  .then(console.log)
  .catch(console.error);
resolve({...})
  .then(console.log)
  .catch(console.error);

Browser

Two browser-ready scripts are provided:

• dist/pdjs.js: For browsers supporting fetch.
• dist/pdjs-legacy.js: For older browsers requiring a fetch polyfill -- mostly IE 11.

Either of these files can be used by copying them into your project and including them directly, with all functions namespaced PagerDuty:

<script src="https://raw.githubusercontent.com/PagerDuty/pdjs/main/dist/pdjs.js"></script>
<script>
  PagerDuty.api({token: 'someToken1234567890', endpoint: '/incidents'})
    .then(response => console.log(response.data))
    .catch(console.error);
</script>