Gatsby blog as PWA (Progressive Web App)

November 26, 2022

Starting with some of the benefits, installed PWAs can bring more user engagement and conversions. On the user side, it brings the possibility to read posts offline. Progressive Web App 101 post covers more details about PWAs.


  • bootstrapped Gatsby blog
  • installed manifest (gatsby-plugin-manifest) and offline (gatsby-plugin-offline) plugins


Add plugin configurations to the Gatsby configuration file. The manifest plugin should be loaded before the offline plugin.

Prepare the app icon in 512x512 pixels, and the manifest plugin will generate the icons in all the necessary dimensions. PWA usage can be logged with the UTM link in start_url property.

Runtime caching for static resources (JavaScript, CSS, and page data JSON files) is set to network-first caching, so it retrieves the latest changes before showing them to the user. In case of issues with caching in a local environment, an offline plugin can be disabled.

// gatsby-config.js
const plugins = [
// ...
resolve: `gatsby-plugin-manifest`,
options: {
name: `app name`,
short_name: `app name`,
start_url: `/?utm_source=pwa&utm_medium=pwa&utm_campaign=pwa`,
background_color: `#FFF`,
theme_color: `#2F3C7E`,
display: `standalone`,
icon: `src/assets/icon.png`
if (process.env.NODE_ENV !== 'development') {
resolve: `gatsby-plugin-offline`,
options: {
workboxConfig: {
runtimeCaching: [
urlPattern: /(\.js$|\.css$|static\/)/,
handler: `NetworkFirst`
urlPattern: /^https?:.*\/page-data\/.*\.json/,
handler: `NetworkFirst`
urlPattern: /^https?:.*\.(png|jpg|jpeg|webp|svg|gif|tiff|js|woff|woff2|json|css)$/,
handler: `StaleWhileRevalidate`
urlPattern: /^https?:\/\/fonts\.googleapis\.com\/css/,
handler: `StaleWhileRevalidate`
module.exports = {
// ...

Service worker updates can also be detected. For a better user experience, a user should approve refreshing the page before updating it to the latest version.

// gatsby-browser.js
exports.onServiceWorkerUpdateReady = () => {
const shouldReload = window.confirm(
'This website has been updated. Reload to display the latest version?'
if (shouldReload) {
window.location.href = window.location.href.replace(/#.*$/, '');
exports.onRouteUpdate = () => {
.then((registration) => registration.update())

Documenting JavaScript code with JSDoc

November 24, 2022

JSDoc provides adding types to the JavaScript codebase with appropriate conventions inside comments so different IDEs like Visual Studio Code can recognize defined types, show them and make coding easier with auto-completion. Definitions are put inside /** */ comments.


Custom types can be defined with @typedef and @property tags. Every property has a type and if the property is optional, its name is put between square brackets, and a description can be included after the property name. Global types should be defined in *.jsdoc.js files so they can be used in multiple files without importing. * represents any type.

* @typedef {object} CollectionItem
* @property {string} [collectionName] - collection name is optional string field
* @property {boolean} isRevealed - reveal status
* @property {number} floorPrice - floor price
* @property {?string} username - username is a nullable field
* @property {Array.<number>} prices - prices array
* @property {Array.<string>} [buyers] - optional buyers array
* @property {Array.<Object<string, *>>} data - some data

Classes are auto recognized so @class and @constructor tags can be omitted.

* Scraper for websites
class Scraper {
* Create scraper
* @param {string} url - website's URL
constructor(url) {
this.url = url;
// ...

Comments starting with the description can omit @description tag. Function parameters and function return types can be defined with @param and @returns tags. Multiple return types can be handled with | operator. Deprecated parts of the codebase can be annotated with @deprecated tag.

* Gets prices list
* @private
* @param {Array.<number>} prices - prices array
* @returns {string|undefined}
const getPricesList = (prices) => {
if (prices.length > 0) return prices.join(',');
* Get data from the API
* @deprecated
* @returns {Promise<CollectionItem>}
const getData = async () => {
// ...

Variable types can be documented with @type tags and constants can utilize @const tags.

* Counter for the requests
* @type {number}
let counter;
* HTTP timeout in milliseconds
* @const {number}
const HTTP_TIMEOUT_MS = 3000;

Enums can be documented with @enum and @readonly tags.

* Some states
* @readonly
* @enum {string}
const state = {

Docs validation

Linter can validate the docs. Add the following package and update the linter configuration file.

npm i -D eslint-plugin-jsdoc
// .eslintrc.js
module.exports = {
extends: ['plugin:jsdoc/recommended'],

Run the linter and it will show warnings if something has to be improved.

Generating the docs overview

Run the following command to recursively generate the HTML files with the docs overview, including the README.md and package.json content. Symbols marked with @private tags will be skipped.

npx jsdoc src -r --destination docs --readme ./README.md --package ./package.json

This command can be included in the CI/CD pipeline, it depends on the needs of the project.

Timeout with Fetch API

November 2, 2022

Setting up a timeout for HTTP requests can prevent the connection from hanging forever, waiting for the response. It can be set on the client side to improve user experience, and on the server side to improve inter-service communication. Fetch API is fully available in Node as well from version 18.

AbortController can be utilized to set up timeouts. Instantiated abort controller has a signal property which represents reference to its associated AbortSignal object. Abort signal object is used as a signal parameter in the request with Fetch API, so HTTP request is aborted when abort method is called.

const HTTP_TIMEOUT = 3000;
const URL = 'https://www.google.com:81';
(async () => {
const controller = new AbortController();
const timeoutId = setTimeout(() => controller.abort(), HTTP_TIMEOUT);
try {
const response = await fetch(URL, {
signal: controller.signal
}).then((res) => res.json());
} catch (error) {
} finally {

Further reading

Progressive Web Apps 101

March 5, 2022

Progressive Web Apps bring some advantages over native mobile apps

  • automatic updates can be implemented
  • the installed app takes less memory
  • installable on phones, tablets, desktops

Prerequisites for installation

  • web app is running over an HTTPS connection
  • service worker is registered
  • web app manifest (manifest.json) is included

Service worker

Read more about it on Caching with service worker and Workbox


Following fields can be included

  • name is a full name used when the app is installed
  • short_name is a shorter version of the name that is shown when there is insufficient space to display the full name
  • background_color is used on a splash screen
  • description is shown on an installation pop-up
  • display customizes which browser UI is shown when the app is launched (standalone, fullscreen, minimal-ui, browser)
  • icons is a list of icons for the browser used in different places (home screen, app launcher, etc.)
  • scope specifies the navigation scope of the PWA. It should start with the URL from start_url value. If the user navigates outside the scope, PWA won't be open from external URLs
  • screenshots is a list of screenshots shown on the installation pop-up
  • start_url is a relative URL of the app which is loaded when the installed app is launched. PWA usage can be tracked by adding UTM parameters within the URL.
  • theme_color sets the color of the toolbar, it should match the meta theme color specified in the document head

Description and screenshots are shown only on mobile phones.

"name": "App name",
"short_name": "App short name",
"background_color": "#ffffff",
"description": "App description",
"display": "standalone",
"icons": [
"src": "icons/icon-128x128.png",
"sizes": "128x128",
"type": "image/png"
"src": "icons/icon-144x144.png",
"sizes": "144x144",
"type": "image/png"
"src": "icons/icon-152x152.png",
"sizes": "152x152",
"type": "image/png"
"src": "icons/icon-192x192.png",
"sizes": "192x192",
"type": "image/png"
"src": "icons/icon-512x512.png",
"sizes": "512x512",
"type": "image/png"
"scope": "/app",
"screenshots": [{
"src": "screenshots/main.jpg",
"sizes": "1080x2400",
"type": "image/jpg"
"start_url": "/app?utm_source=pwa&utm_medium=pwa&utm_campaign=pwa",
"theme_color": "#3366cc"

Manifest file should be included via link tag

<link rel="manifest" href="/manifest.json">

In-app installation experience

It can be implemented on Google Chrome and Edge.

  • listen for the beforeinstallprompt event
  • save beforeinstallprompt event so it can be used to trigger the installation
  • provide a button to start the in-app installation flow
let deferredPrompt;
let installable = false;
window.addEventListener("beforeinstallprompt", (event) => {
deferredPrompt = event;
installable = true;
document.getElementById("installable-btn").innerHTML = "Install";
window.addEventListener("appinstalled", () => {
installable = false;
document.getElementById("installable-btn").addEventListener("click", () => {
if (installable) {
deferredPrompt.userChoice.then((choiceResult) => {
if (choiceResult.outcome === "accepted") {
document.getElementById("installable-btn").innerHTML = "click!";
} else {


chrome://webapks page on mobile phones shows the list of installed PWAs with their details. Last Update Check Time is useful for checking when the manifest file was updated. The app is updated once a day if there are some manifest changes.


A working example is available at https://github.com/zsevic/pwa-starter


Spies and mocking with Jest

August 19, 2021

Unit testing, in addition to output testing, includes the usage of spies and mocking. Spies are functions that let you spy on the behavior of functions called indirectly by some other code. Spy can be created by using jest.fn(). Mocking injects test values into the code during the tests. Some of the use cases will be presented below.

  • Async function and its resolved value can be mocked using mockResolvedValue. Another way to mock it is by using mockImplementation and providing a function as an argument.
const calculationService = {
calculate: jest.fn()
jest.spyOn(calculationService, 'calculate').mockResolvedValue(value);
.spyOn(calculationService, 'calculate')
.mockImplementation(async (a) => Promise.resolve(a));
  • Rejected async function can be mocked using mockRejectedValue and mockImplementation.
.spyOn(calculationService, 'calculate')
.mockRejectedValue(new Error(errorMessage));
.spyOn(calculationService, 'calculate')
.mockImplementation(async () => Promise.reject(new Error(errorMessage)));
await expect(calculateSomething(calculationService)).rejects.toThrowError(
  • Sync function and its return value can be mocked using mockReturnValue and mockImplementation.
jest.spyOn(calculationService, 'calculate').mockReturnValue(value);
jest.spyOn(calculationService, 'calculate').mockImplementation((a) => a);
  • Chained methods can be mocked using mockReturnThis.
// calculationService.get().calculate();
jest.spyOn(calculationService, 'get').mockReturnThis();
  • Async and sync functions called multiple times can be mocked with different values using mockResolvedValueOnce and mockReturnValueOnce, respectively, and mockImplementationOnce.
.spyOn(calculationService, 'calculate')
.spyOn(calculationService, 'calculate')
.spyOn(calculationService, 'calculate')
.mockImplementationOnce((a) => a + 3)
.mockImplementationOnce((a) => a + 5);
  • External modules can be mocked similarly to spies. For the following example, let's suppose axios package is already used in one function. The following example represents a test file where axios is mocked using jest.mock().
import axios from 'axios';
// within test case
  • Manual mocks are resolved by writing corresponding modules in __mocks__ directory, e.g., fs/promises mock will be stored in __mocks__/fs/promises.js file. fs/promises mock will be resolved using jest.mock() in the test file.
  • To assert called arguments for a mocked function, an assertion can be done using toHaveBeenCalledWith matcher.
const spy = jest.spyOn(calculationService, 'calculate');
expect(spy).toHaveBeenCalledWith(firstArgument, secondArgument);
  • To assert skipped call for a mocked function, an assertion can be done using not.toHaveBeenCalled matcher.
const spy = jest.spyOn(calculationService, 'calculate');
  • To assert called arguments for the exact call when a mocked function is called multiple times, an assertion can be done using toHaveBeenNthCalledWith matcher.
const argumentsList = [0, 1];
argumentsList.forEach((argumentList, index) => {
index + 1,
  • Methods should be restored to their initial implementation before each test case.
// package.json
"jest": {
// ...
"restoreMocks": true
// ...

Further reading

Server-Sent Events 101

August 18, 2021

Server-Sent Events (SSE) is a unidirectional communication between the client and server. The client initiates the connection with the server using EventSource API. The previously mentioned API can also listen to the events from the server, listen for errors, and close the connection.

const eventSource = new EventSource(url);
eventSource.onmessage = ({ data }) => {
const eventData = JSON.parse(data);
// handling the data from the server
eventSource.onerror = () => {
// error handling

A server can send the events in text/event-stream format to the client once the client establishes the client-server connection. A server can filter clients by query parameter and send them only the appropriate events. In the following example, the NestJS server sends the events only to a specific client distinguished by its e-mail address.

import { Controller, Query, Sse } from '@nestjs/common';
import { EventEmitter2 } from '@nestjs/event-emitter';
import { Observable, Subject } from 'rxjs';
import { map } from 'rxjs/operators';
import { MessageEvent, MessageEventData } from './message-event.interface';
import { SseQueryDto } from './sse-query.dto';
export class AppController {
constructor(private readonly eventService: EventEmitter2) {}
sse(@Query() sseQuery: SseQueryDto): Observable<MessageEvent> {
const subject$ = new Subject();
this.eventService.on(FILTER_VERIFIED, data => {
if (sseQuery.email !== data.email) return;
subject$.next({ isVerifiedFilter: data.isVerified });
return subject$.pipe(
map((data: MessageEventData): MessageEvent => ({ data })),
// ...

Emitting the event mentioned above is done in the following way.

const filterVerifiedEvent = new FilterVerifiedEvent();
filterVerifiedEvent.email = user.email;
filterVerifiedEvent.isVerified = true;
this.eventService.emit(FILTER_VERIFIED, filterVerifiedEvent);


© 2023