Bullet Train PHP sample

An example of a project to take as a proof of concept that helps to integrate the Bullet Train API with PHP.

This project does not aim to cover best practices for PHP project development as a whole. For example, it does not provide an exhaustive documentation, or unit tests.

However, it provides a good foundation for integrating your developments under the continuous deployment approach, including the implementation of feature flags via Bullet Train.

Getting started

You'll need docker and docker-compose.

To work properly, Bullet Train requires a Google Analytics account. For testing purposes, I abandoned this feature by injecting a workaround into the API.

Do not use this in production !

Clone this project

git clone https://github.com/joogoo/bullet-train-php-sample.git;

Install vendor libraries

cd bullet-train-php-sample;

docker run --rm --interactive --tty \
  --volume $PWD:/app \
  composer install

Run the sample services

docker-compose up -d

Create a super admin account

docker exec -it bullet-train-php-sample_api_1 pipenv run python src/manage.py createsuperuser

Configure your feature flags

• Go to http://localhost:8082/ and sign in with the super admin account you created previously.

• Create an organization, a project and finally 2 feature flags:

ID	Description	Enabled
login	Login feature	your choice
nice_ui	Nice UI feature	your choice

• Get the environment key for this project (see the environmentID value in the Code example: 2: Initialising your project section) and save it as your API_KEY in a conf/client.local.php

• Go back to http://localhost:8080/ and enjoy switching features in Bullet Train [http://localhost:8082].

Go further

Three fake accounts are available for testing purposes, so you can go further activating features for some users only with Segments and User Traits.

Login	Password
callisto	fake1
ganymede	fake2
europa	fake3