Client Data Delivery Validator

Overview

This is a small Python data-quality exercise for validating synthetic client-delivery CSV files before review.

What it validates

• Required columns
• Required values
• Unique product IDs
• Allowed categories
• Positive numeric prices
• Allowed statuses

Design

The command-line workflow follows a simple path:

CSV
  -> pandas DataFrame
  -> validation functions
  -> structured ValidationIssue records
  -> ValidationResult
  -> CLI summary

Each validation function checks one aspect of the delivery. Issues are collected as structured records, and the combined result provides a clear pass-or-review summary.

Project structure

.
├── data/
│   ├── invalid_delivery.csv
│   └── valid_delivery.csv
├── delivery_validator/
│   ├── __init__.py
│   └── validator.py
├── tests/
│   └── test_validator.py
├── .gitignore
├── README.md
└── requirements.txt

Setup

On macOS or Linux with Python 3 installed:

python3 -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt

Run

python -m delivery_validator.validator data/valid_delivery.csv
python -m delivery_validator.validator data/invalid_delivery.csv

Example behaviour

valid_delivery.csv:

• 20 records
• 0 issues
• PASS

invalid_delivery.csv:

• 20 records
• 9 issues
• REVIEW REQUIRED

Tests

python -m pytest -q

12 pytest tests cover the core validation rules and regression behaviour.

Data

All included example products, brands, and records are synthetic.

Scope

This repository is a bounded engineering exercise rather than a production client-data-delivery system.