Skip to content

Latest commit

 

History

History
131 lines (86 loc) · 3.83 KB

README.md

File metadata and controls

131 lines (86 loc) · 3.83 KB
Raw

Aid Transparency Tracker

A data quality measurement tool for international aid data.

Installation

N.B. If you would like to develop locally with vagrant follow these instructions

Run the following commands to bootstrap your environment:

git clone --recursive https://github.com/pwyf/aid-transparency-tracker.git
cd aid-transparency-tracker

Setup a virtual environment, and install dependencies:

python3 -m venv .ve
source .ve/bin/activate
pip install "setuptools<58"
pip install -r requirements.txt

Install and run a postgres database. If you have docker-compose, you can do this:

docker-compose -f docker-compose-postgres.yml up -d

(You may need to run docker compose instead of docker-compose if you have docker v2 installed).

You can login to the PostgreSQL server (which might be useful to check things are working) using psql with the following command (using password from dockerfile; or using port and password you set, if not using the docker compose setup):

psql -h localhost -p 5433 -U postgres

Copy and if necessary edit the config.py.tmpl. (If you installed postgres using docker-compose you shouldn't need to edit anything.)

cp config.py.tmpl config.py

If you're running this for the first time, edit tests/organisations_with_identifiers.csv to have less organisations. e.g. For just one:

cp tests/organisations_with_identifiers_sample_1_org.csv tests/organisations_with_identifiers.csv

Finally, run the setup script to populate your database:

flask setup

This will prompt you to create a new admin user (this will be the username & password you use to login to the ATT web app, once it is up and running--see 'Running' section below).

Fetching and testing data

  1. You can download a dump of today’s IATI data with:

    flask download_data

    The data will be downloaded to the __iatikitcache__ directory by default, or you can add an iatikit.ini file to specify a different location.

  2. The relevant data (according to the organisations in your database) should then be moved into place using:

    flask import_data

    This will move files into the IATI_DATA_PATH specified in your config.py

  3. Tests are run on this data using:

    flask test_data

    The complete output of this is stored as CSV files in the IATI_RESULT_PATH specified in your config.py

  4. Finally, you can refresh the aggregate data shown in the tracker using:

    flask aggregate_results

    This step will destructively populate the aggregateresult table of your database.

Running

You can run a development server with:

flask run

Survey component

Reinitialise

If at any time you need to reset, you can drop all tables using:

flask drop_db

Then follow the installation instructions to reinitialise.

Code formatting

We are currently incrementally adding files to black and isort code formatting.

To ensure the relevant files are formatted correctly, run:

black conftest.py config_test.py unittests/test_web.py integration_tests/
isort conftest.py config_test.py unittests/test_web.py integration_tests/

These will be checked by GitHub Actions.

Data files required by the tests

The PWYF tests which assess the quality of the data use various files which are stored in the tests/ folder. Some of these files need to be updated from the following locations:

orgid.json: http://org-id.guide/download.json

publishers.json: https://iatiregistry.org/publisher/download/json

CRSChannelCode.json: https://iatistandard.org/reference_downloads/203/codelists/downloads/clv3/json/en/CRSChannelCode.json (see https://iatistandard.org/en/iati-standard/203/codelists/crschannelcode/ for more information)