@ianwalter/bff v11.2.2

@ianwalter/bff

Your friendly test runner/framework

Installation

yarn add @ianwalter/bff --dev

CLI Usage

You can pass all configuration described in the next section through the CLI using a dot syntax for nested configuration properties, e.g. bff --puppeteer.dumpio true. Test files can be specified as input without a flag, e.g. bff tests/example.tests.js. Array values can be specified with multiple flag instances, e.g. bff -t dev -t qa.

The following top-level options are also aliased to shorter flags for convenience:

• --log-level, -l [level] Specifies bff's logging level.
• --fail-fast, -f Specifies whether to exit when a test fails instead of continuing to run tests.
• --update-snapshot, -u Specifies whether snapshots should be created or updated.
• --concurrency, -c [#] Specifies how many tests/workers to run in parallel.
• --tag, -t [name] Specifies which test tags should be used to match tests. How it matches the tags depends on the match option below.
• --match, -m [type] Specifies whether a test needs some or every specified tag in order for it to be run.
• --timeout, -T [milliseconds] Specifies how long a test should take before it's marked as failed for timing out.
• --junit, -j [file] Specifies whether or not to write the results to a junit report file.
• --verbose, -V Prints more information for each test: test tags, relative file path, and timing information.

Configuration

Configuration can be specified under the bff property in the current working directory's package.json. The following is a snapshot of the available options with their default values:

{
  bff: {
    // Specifies what files to consider test files.
    // Value should be an array of file paths/globs.
    tests: [
      'tests.js',
      'tests.pptr.js',
      'tests/**/*tests.js',
      'tests/**/*pptr.js'
    ],
    // Specifies bff's logging level.
    // Possible values include: 'debug', 'info', 'warn', 'error', 'fatal'
    logLevel: 'info',
    // Specifies whether to exit when a test fails instead of continuing to run
    // tests.
    // Value should be a boolean.
    failFast: undefined,
    // Specifies whether snapshots should be created or updated.
    // Value should be a boolean (although in practice it only makes sense to
    // use this option with the CLI).
    updateSnapshot: undefined,
    // Specifies how many tests/workers to run in parallel.
    // Value should be a integer.
    concurrency: 4, // This defaults to the number of the host system's CPU
                    // cores minus one.
    // Specifies which test tags should be used to match tests. How it matches
    // the tags depends on the `match` option below.
    // Value should be an array of strings, e.g. ['dev', 'qa'].
    tag: undefined
    // Specifies whether tests have to contain every or at least one specified
    // tag in order to be matched and run.
    // Value should be an array test method ('some' or 'every').
    match: 'some',
    // Specifies how long a test should take before it is marked as failed for
    // timing out.
    // Value should be a length of time in milliseconds (integer).
    timeout: 60000,
    // Specifies whether or not to write the results to a junit report file.
    // Value should either be a boolean or a string specifying the relative path
    // of the report file.
    junit: undefined,

    // Configuration specified under this property will be passed directly to
    // Puppeteer's launch method. See Puppeteer's documentation for information
    // on those options:
    // https://pptr.dev/#?product=Puppeteer&show=api-puppeteerlaunchoptions
    puppeteer: {
      // Treat all test files as puppeteer test files regardless of whether the
      // test file contains the .pptr.js file extension or not.
      // Value should be a boolean.
      all: undefined
    }
  }
}

Writing tests

Declare a test by calling the test function with a name and a function:

const { test } = require('@ianwalter/bff')
const someFunctionality = require('./someFunctionality')

test('some functionality', ({ expect }) => {
  expect(someFunctionality()).toBeTruthy()
})

If the test name/description is really long, you can also pass the test function in a second call:

test(`
  some functionality when some environment variable is set, the user has some
  localStorage value, and some query parameter has a certain value
`)(({ expect }) => {
  expect(scenario).toMatchSnapshot()
})

You can skip individual tests by adding the .skip modifier:

test.skip('something', ({ expect }) => {
  expect(something).toBe(somethingElse)
})

You can also have only designated tests in a test file run with the .only modifier:

test.only('focus', ({ expect }) => {
  expect({}).toEqual({})
})

If you have flaky tests and you don't necessarily want them to fail a test run but would like to be warned when they fail, you can use the .warn modifier:

test.warn('Payfail checkout', asnyc ({ expect }) => {
  expect(await payfailCheckout()).toEqual({ success: true })
})

Related

• @ianwalter/bff-webdriver - A bff plugin to enable WebDriver-based testing
• @ianwalter/bff-puppeteer - A bff plugin to enable Puppeteer-based testing

License

Apache 2.0 with Commons Clause - See LICENSE

Created by Ian Walter