Overview

This util outputs the join of two datasets with client information: personal data and financial data.

This can be achieved by running the CLI program or spining up an API and calling the endpoint.

Requirements

To run the util or spin up the API locally python3.10 (or later) is required. To run the util or spin up the API via the containerized util docker is required.

Installation

System

NOTE: it's recommended to use a Virtual Environment. Read this.

• Install the requirements with: pip install -r requirements.txt.

Docker

• Build the image: docker build --network=host . -t abm-amro-exercise.

Tests

Run tests with the system installation or within the built docker image as follows:

System

Run tests with: python -m pytest tests. Run tests generating a coverage report with: python -m coverage run -m pytest tests. To show the coverage report, run python -m coverage report --omit 'tests/conftest.py'.

Docker

Run tests with: python -m pytest tests.

docker run \
    --rm \
    -it \
    -v <LOCAL PATH TO FINANCIAL AND PERSONAL DATA>:/home/abm/data \
    -v <LOCAL PATH TO STORE LOGS FROM DOCKER APP>:/home/abm/logs \
    --network=host \
    --entrypoint python \
    abm-amro-exercise -m pytest tests`

NOTE: Change pytest tests to coverage run -m pytest tests to generate a coverage file (remember to add another volume to persist it in local!)

Usage

System install

• Run util with python -B app/core.py (see options with flag -h), OR spin up the API (locally) with python -Bm uvicorn app.__main__:api --port 8000.

CLI usage example:

python -B app/core.py \
    --path-to-personal-info-file data/personal_info.csv
    --path-to-financial-info-file data/financial_info.csv
    --countries Netherlands 'United Kingdom'

Docker install

• Run util with:

docker run \
    --rm \
    -it \
    -v $(pwd)/data:/home/abm/data \
    -v $(pwd)/docker-logs:/home/abm/logs \
    --network=host \
    abm-amro-exercise

Once inside the container, run the util following the same instructions for System install.

NOTE: add/modify volumes to map to other datasets, write the output, etc. NOTE2: network=host allows reaching the API easily (after spining it up). NOTE3: Docker must use a different folder for logs (or add USERs and permissions to the image). Otherwise, it will cause write permission issues when used along the locall installation.

client_data/result.csv

System install

# Create the desired destination folder
mkdir -p client_data

# Pipe the output of the program to the deesired filepath
python -B app/core.py \
    --path-to-personal-info-file data/personal_info.csv
    --path-to-financial-info-file data/financial_info.csv
    --countries Netherlands 'United Kingdom' > client_data/result.csv

Docker install

# Create the desired destination folder
mkdir -p client_data

# Pipe the output of the program to the desired filepath.
docker run \
    --rm \
    -it \
    -v $(pwd)/data:/home/abm/data \
    -v $(pwd)/docker-logs:/home/abm/logs \
    --network=host \
    --entrypoint=python \
    abm-amro-exercise \
    --path-to-personal-info-file data/personal_info.csv
    --path-to-financial-info-file data/financial_info.csv
    --countries Netherlands 'United Kingdom' > client_data/result.csv