BDD/Gherkin QA robot for effortless software testing

As a QA Specialist, I want to generate executable BDD tests for APIs and web apps

QAshield:bot is a testing, automation and orchestration tool for agile software teams.

QAshield uses English as a scripting language to help simplify, standardise and socialise software deliverables.

We take a Business Driven Development (BDD) approach to designing, deploying and governing complex software systems.

Tests are written in English so that every stakeholder (including the CIO :-) can make sense of them.

View the PDF introduction.

I want to automate myself

We're all busy. We write scripts, or at we should. But our scripts are unreadable by our stakeholders - often even ourselves.

With QAshield, you write your requirements in a simplified dialect of English - called Gherkin.

These "Features" contain executable English instructions - actions and assertions - that are easy to read, write and socialise.

Features are lists of related scenarios. It's the scenario's that do all the heavy lifting.

A scenario describes your pre-conditions, actions and outcomes in a way that is both human and machine friendly.

QAshield can be invoked elegantly from an API, the command line, Mocha, your IDE or your DevOps workflow.

I want to learn Executable English

The BDD notation for a feature / scenario is:

Feature: Basic Examples

Scenario: An example
	GIVEN   some context
	WHEN    an action is performed
	THEN    an outcome is expected

The text following the keyword (GIVEN | WHEN | THEN) needs to match a phrase/pattern from a vocabulary.

This BDD notation is called "Gherkin". Gherkin is human and machine readable - business analysts, featureers, developers and robots can collaborate.

New features can be created using any simple text editor - where you'd use QAShield:bot's extensible vocabulary to write them.

You can download pre-packaged vocabularies (called Dialects) and/or roll your own with simple Javascript.

I want to see a working example

To improve readability, the keyword AND can be used instead in place of the verbs above.

You can influence what QAShield:bot understands using @dialect annotations.

Feature: Verify that Google is accessible

Background: Google Scenarios

    Given I am googling

  Scenario: Request Google homepage - with redirects

    Given I enable redirects
    When I GET http://google.com
    Then response code should be 200

  Scenario: Request Google homepage - no redirect

    Given I disable redirects
    When I GET http://google.com
    Then response code should be 3xx

View more .feature examples.

I want to test-drive QAShield:bot

QAShield:bot is built using NodeJS.

If you're new to NodeJS, pre-read Installing Node.

You install QAShield:bot as a system-wide CLI command:

$ npm install qashield-qabot -g

To run it simply type:

$ qa

However, it won't do much else until we provide some feature scenarios.

By default, QAShield:bot looks for ".feature" files recursively, by default they're stored in the "./features/" directory.

I want to create my first Feature

1. To quickly create a few examples in the ./features folder, type:

    $ qa --example

This will create the ./features folder and copy some simple examples.

It will also write your default configuration to ./qa.json

It won't damage if you run it again, except re-save your ./qa.json config.

Or, you can just create the ./features folder and a default "qa.json" without the examples:

$ qa --initialize

2. To execute your example ".feature" files, type:

    $ qa

3. If something goes wrong, enable the built-in debugger.

    $ export DEBUG=qa*
    $ qa

Now, the output is verbose and colour-coded to make your life easier.

To turn off debugging, type:

	$ export DEBUG=
	$ qa

I want to learn some vocabulary

QAShield:bot ships with few default vocabularies - variables, files, web apis, web apps, etc.

Try out the included .feature examples - they cover a range of powerful use cases.

To discover what phrases exist in the vocabularies, type:

$ qa --knows

Let's create a trivial example of a hypothetical test case.

Scenario: Trivial Test

    Given I am testing debug
    When debug works
	And I succeed

The steps are executed in sequence.

The GIVEN steps setup pre-conditions. The "Given I am $acting" phrase doesn't do much - except communicate intentions.

The "WHEN ... " steps do useful work that result in desirable outcomes. For example: writing a file, requesting a web page, etc.

In this example, we simply write a debug message to the console, so let's turn on debug output.

$ export DEBUG=qa*

You can adjust the logging scope - to see only Web API messages, use:

$ export DEBUG=qa:webapi

The "THEN ..." steps make assertions, that is they test that conditions are met. For example, you can use arbitrary Javascript if necessary:

    Then I assert $some_javascript

The "I succeed", "I pass" always meet their conditions. The inverse "I fail" forces the scenario to abort and report it's failure.

I want to learn more about QAShield:bot

For runtime options, type:

$ qa  -h

We actively add capabilities and fix bugs. To upgrade to the latest version, type:

$ npm update qashield-qabot -g

For more information, please read:

Example Features.

Basic Dialect.

Advanced Dialect.

Something Went Wrong.

Web API Dialect.

Web API Advanced.

Web Apps Dialect.

Web Apps Advanced.

Software Blueprint Dialect.

Software Blueprint Advanced.

I want to organise my work into folders

If your features are in a different location then use the "--features" option to locate them.

$ qa --archive ./my-archive --features ./my-features

These folders are not automatically created, they will throw an error if they're not found.

I want to re-use my features in other projects

QAShield:bot was designed to support a declarative style so that features are portable.

Supplying a different "config" file for each environment allows features to be re-used across multiple environments.

For example, features can adapt to dev, QA and live environments - injecting hostnames, credentials, etc as required.

Most Dialects configure themselves automatically.

If yours doesn't then there is alternative - use {{mustache}} templates to modify statements prior to execution.

Given I login as {{scope.actor}}

In this way, your BDD features are neatly abstracted from your runtime configuration.

To specify a runtime configuration for your features, type:

$ qa --config ./my-context.json

By default, QAShield:bot will try to load a configuration file called "qa.json" from your current directory.

If no file is found, then sensible default values are defined.

I want to do something before every scenario

Backgrounds are similar to scenarios, except they do not support annotations.

Any feature can contain a background, in which case the steps that carried out before each scenario.

Background: Authenticate

	GIVEN I login
	AND I use a valid client certificate

I want to know how it works

First, QAShield:bot parses the command line and initializes the Dialect, Features and Engine components.

Next it loads the default dialects. These can be specified on using the APIGEEK_DIALECT environment variable. Dialects can also be specified using --dialect option and within Feature: definitions using the @dialect annotation.

Each Dialect instructs the parser (Yadda) to match a set of Gherkin phrases to their related function.

The Feature manager converts features and scenarios into executable units.

User defined variables are scoped at the feature-level meaning they are shared between scenarios within the same feature.

The --config file is used as the basis for internal context variables. They are scoped to each scenario.

These variables - such as web requests/responses - can be accessed using the "this." qualifier - with due caution.

Next, the Engine runs each feature using the built-in Mocha runner. Results are correlated and output according to your CLI options.

I want to add comments

It's useful to document your intentions or to prevent a statement from running, for example during development.

Simple, place a # before any line and it will be ignored by QAshield.

# This is ignored by the parser
Scenario: Comments Example

	Given I am using comments
	# Then I fail
	Then I succeed

Instead, you should use @skip or @todo before a Feature: or Scenario: definition.

An @bug scenario will pass normally (skipped) but fail when --debug is used.

@bug=something is broken

Scenario: A Bug
	Given I am a bug
	Then I fail

I want to automate everything

That is our goal too. We'll continue to address the needs of Enterprise QAshields.

Competent software engineers can easily create "blueprints" that capture the patterns, templates and files used to build SDLC artefacts.

Then QAShield:bot can re-use those blueprints to build customised websites, portals, Apps, APIs, mock servers, micro services and more.

Please share any custom Blueprints and Dialects so that QAShield:bot becomes more useful for all of us.

If you need support to build or debug your community Blueprints or Dialects, please [email protected]

I want to license QAShield:bot

This software is licensed under the Apache 2 license, quoted below.

Copyright 2015-2016 AS Network

Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with the License. You may obtain a copy of the License at

[http://www.apache.org/licenses/LICENSE-2.0]

Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License.