gitcommit

Tooling to create smart git commit messages.

Getting started

Checkout the repository and create a virtual environment a venv folder.

python3 -m venv venv

Everytime you interact with the project, activate your virtual environment.

source venv/bin/activate

Install development requirements with:

pip install -r requirements/development.txt

Take a look at the cli at src/main.py

In order to run it, use:

python -m gitcommit

Tests

To run the tests use pytest in the repo-root.

pytest --cov gitcommit

This runs all tests and prints the current coverage to the terminal.

Building

To build and install the programm run pip install -e .

You can access it via igit

Motivation

Finding the right commit message is hard. Maybe it's even tempting to ignore them and add a placeholder instead. Every git user did the git commit -m '.' at some point in their career.

What it does

igit commit takes information from your issue tracker, the git repository and the staged code. Those are used to render a smart commit message template, helping the developer to fill them with meaningful and explaining content.

How we built it

We built igit as a python package with focus on extensibility of information sources, e.g. the issue tracker or language support. To interface with git, we used pygit2. Template rendering is done via jinja2.

Information are gathered in dedicated modules and rendered with a template that can be configured for different git repositories similar to a .gitignorefile.

Challenges we ran into

• missing or incomplete documentation for a package
• working on a subset of files with multiple people (happy merging!)

Accomplishments that we're proud of

• working command line interface
• structured repository

What we learned

We learned that collaborative work, given only a short amount of time, is challenging. But working in a small (local) team makes prioritizing issues and solving problems as they arise a lot easier. Someone with an idea is around to help.

What's next for igit commit

The next steps involve building a igit command line interface, supporting:

• other programming languages
• different issue trackers (e.g. GitHub, Atlassian Bitbucket)
• more subcommands, helping the developers with changelog creation or merge request processing