@habx/nightwatchx

Nightwatchx 🦉

QA tool to run scenarii tests and compare screenshots for UI regression using Nightwatch integration with Browserstack integrated with Slack.

⚠️ This script requires a AWS S3, Browserstack and Slack API access ⚠️

• AWS S3: Storage of images for UI regression
• Browserstack: multiple devices tests
• Slack: tests notifications

Install

# in your project (better if you want to use some of our code for tests)
npm i @habx/nightwatchx

# globaly, you handle tests writing on your own
npm i -g @habx/nightwatchx

Usage

Build

# build tests
npx nightwatchx build

# build local tests
npx nightwatchx build local

Test

# run your tests
npx nightwatchx test

# run specific tests
npx nightwatchx test --only=login

# run on a local chromium
npm run  test:local --only=login

Config

If you want to run tests on browser stack using npx nightwatchx test, you'll need the following environement variables:

BROWSERSTACK_ACCESS_KEY=***
AWS_SECRET_ACCESS_KEY=***
SLACK_TOKEN=***

AWS_ACCESS_KEY_ID=***
SLACK_HOOK=***
SLACK_HOOK_ERROR_ONLY=***
AWS_ACCOUNT_ID=***
AWS_REGION=***
BROWSERSTACK_USERNAME=***

Options

param	description
only	run only specified test suites
silent	do not send slack message
local	use the local config that uses a local chromium

How tests are ran ?

There are 4 levels of configuration

├─ test suites/folders
│   ├─ runs <--- defined in `manifest.json` per test
│   │   ├─ environments <--- list of options defined in `environments.ts` (in our case it contains the url [prod/staging/dev]) 
│   │   │   ├─ devices <--- list of options defined in `devices.ts`

Test suite config

Each test suite should be associated with a manifest.json file. The later contains the configuration for the tests to be run with browserstack. The whole configuration is passed down to the globals of nightwatch

{
  "runs": {
    "default": {
      "environments": [ // All the environements that should be run for this test
        "prod",
        "dev"
      ]
      "devices": [ // All the devices that should be run for this test
        "win10_edge",
        "mac10.14_chrome70",
        "win10_firefox"
      ],
      "frequencyMinutes": 1440 // The frequency
    },
  },
}

Run options

option	type	default	description
environments	array	["default"]	environments to run the test on
devices	array	["default"]	devices to run the test on
frequencyMinutes	number (in minutes)	0	frequency of the test
threshold	float	0.2

Environment config

All the environments options should be described in the file config/environments.json. You could also configure specific parameters by environment for tests.

{
  "default": {
    "url": "https://www.habx.fr"
  },
  "prod": {
    "url": "https://www.habx.fr"
  },
  "prod_en": {
    "url": "https://www.habx.com/en"
  }
}

Devices config

All the devices settings used by browserstack should be described in the file config/devices.json. You could also configure specific parameters by device for tests. See how to configure a device for browserstack

{
  "default": {
    "env": "chrome",
    "desiredCapabilities": {
      "browser": "chrome",
      "resolution": "1024x768",
      "browser_version": "74.0",
      "os": "Windows",
      "os_version": "10"
    }
  }
}

Predefined devices

Some devices are already defined in the library:

• win10_chrome74
• win10_firefox66
• win10_edge18
• win10_ie11
• android_samsungS9
• ios_iphoneX

Slack integration

A slack message is sent at each run. It contains a summary of the run and its potential errors or warnings

If everything goes well

If a screenshot comparison exceeds the acceptable diff

• We show which device(s) failed with a ⚠️
• The first screenshot which exceeded the comparison threshold is attached to the message

If the test suite fails

• We show which device(s) failed with a ❗️
• A screenshot of the screen just after the test failed is attached to the message

Run test with github workflow

Use the folowing github workflow config to run your tests every hour 🎉

name: Run tests

on:
  schedule:
  - cron: '0 * * * *'

jobs:
  build:

    runs-on: ubuntu-latest

    strategy:
      matrix:
        node-version: [8.x]

    steps:
    - uses: actions/checkout@v1
    - name: Use Node.js ${{ matrix.node-version }}
      uses: actions/setup-node@v1
      with:
        node-version: ${{ matrix.node-version }}
    - name: npm install, build, and test
      env:
        SLACK_HOOK: https://hooks.slack.com/services/*****
        SLACK_HOOK_ERROR_ONLY: https://hooks.slack.com/services/*****
        AWS_ACCESS_KEY_ID: ****
        AWS_ACCOUNT_ID: ****
        AWS_REGION: eu-west-1
        AWS_SECRET_ACCESS_KEY: ${{ secrets.AWS_SECRET_ACCESS_KEY }}
        BROWSERSTACK_ACCESS_KEY: ${{ secrets.BROWSERSTACK_ACCESS_KEY }}
        BROWSERSTACK_USERNAME: ${{ secrets.BROWSERSTACK_USERNAME }}
        SLACK_TOKEN: ${{ secrets.SLACK_TOKEN }}
        BUCKET_NAME: ${{ secrets.BUCKET_NAME }}

      run: |
        npm install
        npm run build
        npm run test

API

UI regression

browser.compareScreenshot(fileName: string, { threshold: number, waitBetweenScreenshots: number }