Exchanger Python Demo App

This repository contains a demo application meant to demonstrate CLI application development with Python based on type hints and OOP, including useful modules/libraries/syntax/etc.

Exchanger app specification

The Exchanger app has a simple purpose: it's a CLI app for converting amounts of money from one currency to another. The inputs are provided in a file, and the ouputs are printed to the console.

Input file format

<first line is "from" currency>
<second line is "to" currency>
<amount of money to convert (positive decimal > 0)>
<amount of money to convert (positive decimal > 0)>
<amount of money to convert (positive decimal > 0)>
...

An example input file converting from US Dollar to Euro is provided here, and the full list of currency codes can be found here. For simplicity sake it's assumed inputs are valid and any issues, e.g. money isn't positive decimal, will raise and propagate exceptions as-is.

Running the Exchanger app

1. Clone this repo

2. Setup a Python virtual environment with all the Python dependencies based on requirements.txt. Real Python has a useful article for virtual environments.

3. Run src/main.py and provide the file path to the input file. To use the example input file provided with repo run in your terminal (cmd/bash/etc.) from repo root:

   python src/main.py exchanger_conversion_spec.txt

Implementation references

Exchange rate from one currency to another is discovered by using the free API https://exchangerate.host/#/docs.

The Python implementation was written for Python 3.9.6, tested with CPython, and relies on the following:

Important built-in modules

• typing module - for type hints support, specialized strongly typed classes and using advanced typing features
• abc module - implementing Abstract Classes
• asyncio module - support for async IO, futures, coroutines and tasks
• pathlib module - utilities for file system paths, particularly useful Path type

Important Python libraries

• typer - create CLI applications with type hints support
• lagom - a DI container that enables DI design principle with auto-wiring of dependencies based on type hints
• pydantic - represent system entities as typed data models using type hints with many features that improve upon the built-in dataclasses module. Can also read application settings into data models from environment variables, secret files, and any other custom settings source, e.g. JSON files.
• orjson - fast library for working with JSON - serialization & deserialization
• aiohttp - used as HTTP client to send HTTP requests using async IO
• aiofiles - file system access using async IO
• asyncstdlib - async versions of built-in functions; this repo uses its tuple function to transorm the input file lines provided by aiofiles into a tuple.

Code Quality and Linting

The following code quality and linting solutions are used:

• MyPy - type hints checker. pydantic integrates with it to type check data models.
• black - code formatter.
• isort - sorting import statements.
• flake8 - Style guide enforcement. Here are some useful plugins:
  • flake8-builtins - checks for accidental use of builtin functions as names
  • flake8-comprehensions - checks for misuse or lack of use of comprehensions
  • flake8-mutable - checks for mutable default parameter values Python issue
  • pep8-naming - checks that names follow Python standards defined in PEP8
  • flake8-simplify - checks for general Python best practices for simpler code
  • flake8-pytest-style - check that pytest unit tests are written according to style; this plugin isn't relevant for this repo
  • flake8-logging-format - ensures logs use extra arguments and exception(); this plugin isn't relevant for this repo

Note MyPy and flake8 are configured with setup.cfg.

Contributions

This repo doesn't accept contributions since it's acting as a demo.