ETL Data warehouse
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
scripts
sift-core
sift-odk-importer
terraform
.gitignore
.travis.yml
LICENSE
README.md
docker-compose.yml
setup.cfg

README.md

Sift

Survey collection and analytics

Setup

Dependencies

Installation

$ git clone git@github.com:readevalprint/sift.git
$ cd sift

$ docker-compose build
$ docker-compose run core setuplocaldb
$ docker-compose run odk-importer setuplocaldb

Setup

The only thing that you need now is a superuser!

docker-compose run core manage createsuperuser
docker-compose run odk-importer manage createsuperuser

Usage

This will start the sift-core 0.0.0.0:8000 and the odk-importer on 0.0.0.0:8443

$ docker-compose up

Development

All development should be tested within the container, but developed in the host folder. Read the docker-compose.yml file to see how it's mounted.

$ docker-compose run core bash
root@localhost:/#

Deployment

Set the GATHER_CORE_TOKEN and GATHER_CORE_URL environment variables when starting the gather2-odk-importer to have ODK Collect submissions posted to Gather2 Core. If a valid GATHER_CORE_TOKEN and GATHER_CORE_URL combination is not set, the server will still start, but ODK Collection submissions will fail.

When deploying set the env var DJANGO_S3_FILE_STORAGE to True. See settings.py for the other available settings.

Infrastructure deployment is done with AWS CloudFormation, which configuration files are stored in cloudformation directory.

Application deployment is managed by AWS Beanstalk and is being done automatically on the following branches/environments:

All the logs are forwarded to CloudWatch Logs to be easily accessible for developers