local_setup.py.example

import datetime

import mara_pipelines.config
import mara_acl.config
import mara_db.config
import mara_db.dbs
from mara_app.monkey_patch import patch

import app.config


@patch(mara_db.config.databases)
def databases():
    return {
        # the project requires two databases: 'mara' for the app itself, and 'dwh' for the etl
        'dwh': mara_db.dbs.PostgreSQLDB(user='root', host='localhost', database='example_project_dwh'),
        'mara': mara_db.dbs.PostgreSQLDB(user='root', host='localhost', database='example_project_mara')
    }

# Disable http header based authentication
patch(mara_acl.config.require_email_http_header)(lambda: False)


# How many cores to use for running the ETL, defaults to the number of CPUs of the machine
# On production, make sure the ETL does not slow down other services too much
patch(mara_pipelines.config.max_number_of_parallel_tasks)(lambda: 4)

# The first day for which to download and process data (default 2017-01-01).
# Locally, a few days of data is enough to test a pipeline.
# On production, size of days that can be processed depends on machine size.
# One year of data amounts to roughly 50GB database size
patch(app.config.first_date)(lambda: datetime.date.today() - datetime.timedelta(days=5))

# Whether it is possible to run the ETL from the web UI
# Disable on production
patch(mara_pipelines.config.allow_run_from_web_ui)(lambda: True)