NYPR Crowdsourcing

Development

Prerequisites

You will need the following software properly installed on your computer. Please follow all the directions as written on the websites of these projects and do not neglect to install their dependencies.

Git
Docker

Getting started

Clone the repo.

$ git clone git@github.com:nypublicradio/crowdsourcing
$ cd crowdsourcing

Create an .env file for environment variables

$ cp .env.sample .env

Build the image.

$ docker-compose build

Run the development server.

$ docker-compose up

Run the migrations.

$ docker-compose exec django manage.py migrate

Create a superuser.

$ docker-compose exec django manage.py createsuperuser

Running tests

Tests are executed via pytest using NYPR's setuptools extensions. Use exec to run tests on running development containers and run to run tests within a new container.

$ docker-compose exec django python setup.py test

For faster testing in development, test dependencies can be permanently installed.

$ docker-compose exec django python setup.py test_requirements

Interactive debugging

To enable ipdb breakpoints developers need to attach to the Docker container running the Django development server.

Start the containers and detach from the log output.

$ docker-compose up -d

If using ipdb for debugging it will need to be installed in the development container.

$ docker-compose exec django pip install ipdb

Attach to the container. The example below provides the likely name for the Django container, but if incorrect it can be obtained via docker-compose ps.

$ docker attach crowdsourcing_django_1

Alternate workflow

For developers who prefer to run the Django server directly on their local system an alternate workflow is available.

Create and activate a virtual environment.

$ virtualenv ~/.virtualenvs/crowdsourcing
$ . ~/.virtualenvs/crowdsourcing/bin/activate

Install the package dependencies and (optionally) the test dependencies.

$ pip install -e .
$ python setup.py test_requirements

Start a Postgres container in the background. This will expose Postgres on localhost port 5432. Exclude the -v ... argument to avoid persisting data over multiple container runs.

$ docker run -v /tmp/postgres:/var/lib/postgresql/data -dp 5432:5432 postgres

Set environment variables to target postgres container. The four commands below and any other development config can be added to the virtual environment's bin/activate script to automate this moving forward.

$ export DB_HOST=localhost
$ export DB_NAME=postgres
$ export DB_USER=postgres
$ export DB_PASSWORD=

Execute the Django initialization commands.

$ ./manage.py migrate
$ ./manage.py createsuperuser
$ ./manage.py runserver 0.0.0.0:8080

Configuration

Configuration should be set via environment variables.

Config Value	Description
`AWS_ACCESS_KEY_ID`	Set via boto3 config or environment variable.
`AWS_DEFAULT_REGION`	Set via boto3 config or environment variable.
`AWS_SECRET_ACCESS_KEY`	Set via boto3 config or environment variable.
`AWS_S3_CUSTOM_DOMAIN`	Cloudfront domain alias for static files.
`AWS_STORAGE_BUCKET_NAME`	S3 bucket for Django's storage backend.
`DB_HOST`	IP or hostname of the database.
`DB_NAME`	Database name (for app, not tests).
`DB_PASSWORD`	Database password (for app, not tests).
`DB_USER`	Database user (for app, not tests).
`DJANGO_URL_PREFIX`	URL route to service (/crowdsourcing in prod).
`DJANGO_SECRET_KEY`	Securely generated key for internal Django use.
`SENTRY_DSN`	URL for reporting uncaught exceptions.

nypublicradio / crowdsourcing