The repository for Project Bluebottle, the crowdsourcing framework initiated by GoodUp.

For those who want to try out to the BlueBottle project, here's to get started:

1. Fork and/or clone the repository.
2. Navigate to your local repository directory.
3. Create a bluebottle/settings/local.py base on bluebottle/settings/secrets.py.example

Now you have the option to install the application with, or without Docker. If you wish to not use Docker, continue below. Otherwise, see :ref:`docker`.

1. Create a virtual environment within the repository directory, for example:

   $ virtualenv env
   $ source env/bin/activate
   

2. Make sure you have a recent Python distro (3.7+ recommended).

3. Make sure (a recent) virtualenv is installed.

4. Install the project:

   $ pip install -e .[test]
   

5. Migrate the database schemas:

   $ python manage.py migrate_schemas --shared --settings=bluebottle.settings.testing
   

6. (optional) Create a new tenant (if you haven't imported a database dump):

   $ python manage.py create_tenant
   

7. (optional) Re-index any imported data using Elasticsearch

   • For all tenants:

     $ python manage.py tenant_command search_index --rebuild -f
     

   • If for whatever reason this doesn't work, try running the migration step for a single tenant only, and indexing that tenant's data. E.g.:

     $ python manage.py migrate_schemas -s onepercent --settings=bluebottle.settings.testing
     $ python manage.py tenant_command --schema onepercent search_index --rebuild -f
     

8. Start the server:

   $ python manage.py runserver
   

9. You might still need to

   • Install libraries to get pip install working, like libxmlsec1, postgresql, postgis, elasticsearch 6.x

     • Installing these is not necessary when using Docker

   • Alter your hosts file (e.g. in /etc/hosts on Linux/OSX) to contain tenants you've created like:

     127.0.0.1 tenant35.localhost
     

It is possible to run thy Python server, PostgreSQL and Elasticsearch in a Docker environment using docker-compose. To get started, make sure to download a Docker client, like Docker Desktop.

Make sure to download Docker (e.g. Docker Desktop) first.

In your local.py file, set the DATABASES variable to the following:

DATABASES = {
    'default': {
        'ENGINE': 'bluebottle.clients.postgresql_backend',
        'HOST': 'postgres',
        'PORT': '5432',
        'NAME': 'reef',
        'USER': 'postgres',
        'PASSWORD': 'postgres',
        'DISABLE_SERVER_SIDE_CURSORS': True # this prevents issues with connection pooling
    },
}

ELASTICSEARCH_DSL = {
    'default': {
        'hosts': 'elasticsearch:9200'
    },
}

• To start the containers (if this is your first time, all dependencies will be installed automatically):

  $ docker-compose -u -d
  

• To import a database dump file (please note that the last two commands run without the -t flag):

  $ docker exec -it -u postgres postgres dropdb reef
  $ docker exec -it -u postgres postgres createdb reef
  $ bzcat reef-prod-current.sql.bz2 | docker exec -i -u postgres postgres psql reef
  $ echo "UPDATE clients_client SET domain_url=CONCAT(client_name, '.localhost');" | docker exec -i -u postgres postgres psql reef
  

• If you are running into errors related to max_map_count, then run the following command in your terminal: sysctl -w vm.max_map_count=262144

To run the containers:

$ docker-compose -u -d

To run one specific container:

$ docker-compose -u -d [CONTAINER_NAME]

Or on other systems (some OSX):

$ docker compose up -d elasticsearch

To shut them down:

$ docker-compose down

The environment also comes with pgAdmin included so you can inspect the local database. Navigate to http://localhost:5050 and login with these credentials:

• Email: [email protected]
• Password: admin

After that, you can add a new server using the details below to inspect the PostgreSQL database:

• Host: host.docker.internal
• Username: postgres
• Password: postgres

To run commands in the Python container:

$ docker exec -it bluebottle python manage.py [YOUR_COMMAND]

- For example::

    $ docker exec -it bluebottle python manage.py migrate_schemas -s onepercent --settings=bluebottle.settings.local

If you need to rebuild the container, for example when you want to apply changes after pulling the latest version of a branch, run:

$ docker compose up --build

The BlueBottle test suite can be run completely using:

1. Install the dependencies

   $ pip install -e .[test,dev]

2. Create test db and restore testdata

   $ createdb test_reef $ psql test_reef < testdata.sql

3. Run the tests

   $ python manage.py test -k