Django project of the labeling process of UML fragments

A UML fragment is a portion of a UML class diagram. The label is a English text description. The app is hosted on Heroku!

Getting Started

This will run the website on your own computer.

Installation

To run this locally, you need:

• django : Version 3+
• Python 3+
• A Bash interpreter like Git Bash or Cygwin
• wget: a GNU command for downloading files

Collect data

To collect the data into the repo, use download.sh.

./download.sh

This will download a zip from another repo and unzip it in labeling/static/fragments/. About 9000 images totaling 200 MB will be available.

Run a local server

At the root of the repo, run in the command line

python manage.py migrate
python manage.py runserver

A message like this will print. Access the website via the IP address http://127.0.0.1:8000/.

$ python manage.py runserver
Watching for file changes with StatReloader
Performing system checks...

System check identified no issues (0 silenced).
July 09, 2021 - 10:15:20
Django version 3.2.4, using settings 'umllabels.settings'
Starting development server at http://127.0.0.1:8000/
Quit the server with CTRL-BREAK.

You only need to migrate once. After that, you can just use runserver.

Non-django files

• images.py: Copies relevant images from the raw fragment folder to the django static folder
• fragment.db: Original SQLite database of the file names of models and fragments
• download.sh: Downloads the archive of images from the internet and extracts it in the correct folder.