A Web based frontend that enables a user to evaluate the correctness of a source of site information.
The evaluator runs as a series of docker containers. There are three docker-compose files for running the system in 3 different modes:
Use docker-compose.yml. Will build all containers from source for local development and allows for live changes of the UI. Exposes intermediate services like postgres and graphiql to the host network for debugging.
Use production.yml. Build all containers from source in production mode. Only exposes UI and goaccess ports. Use .env to configure what port these services run on. All services will be reverse proxied through Nginx.
Use aws.yml. Identical to production, but uses prebuilt containers pulled from Docker Hub. Use .env to configure what ports to use as well as what image tag to use. Image tags match the git branch they were built from. The only required files are aws.yml and postgres/schema.sql on first run to initalize the database, if you choose to run the database in docker-compose instead of externally (ie. with AWS Fargate).
All secret env variables should be put in .env file in the root directory. Additionally there are variables to configure docker compose based on what mode you've chosen above.
The file should look like:
# mode specific
TAG= # aws only
UI_PORT= # aws or production
GOACCESS_PORT= # production only
# always used
DB_USER=
DB_PASSWORD=
DB_HOST=
JWT_SECRET=
GEOCODIO_API_KEY=
RDWATCH_HOST=
RDWATCH_USERNAME=
RDWATCH_PASSWORD=Run in psql:
SELECT smart_private.register_user('my_username', 'my_password');
edit permission in the smart.user table
Find the source ID's you want to give the user access to and add them to the smart_Private.users_sources table:
INSERT INTO users_sources('user_id','source_id') VALUES(999,999);
Sources can be imported using the "import" button in the navigation bar. In python/manual_import, there are scripts for manually importing from a CSV file. See manual import instructions here
Follow these instructions.