Fedora Patch Tracker

A prototype for the Fedora Patch Tracker Project in GSoC 2015. WIP

Usage

1. Install the dependencies with pip install requirements.txt
2. Install redis and start it
3. Open the shell by running python shell.py and create the database by executing db.create_all().
4. Run the development server python run.py
5. Run python worker.py
6. Access the project by visiting http://localhost:5000 on your browser

Limitations

The current design goes through all the branches of the package's git repository and stores the patches in the database. This poses a few problems:

1. All patches are decoded using UTF-8 charset and characters are replaced in case there's a decode error
2. Packages with a large number of patches (e.g. kernel) end up taking a lot of space in the database and the retrieval GET request for them is large. This may be solved by storing the reference of the patches in the database instead and retrieving them individually on demand instead of having a single GET request to fetch them all.