dj4e-samples: a shelf of small Django projects to read and copy

One repository holding many small applications

dj4e-samples is described as a set of Django 5.2 projects that provide free sample code to support the dj4e.com web site, and some of the samples run live at samples.dj4e.com.

What that means in practice is a directory of small Django applications side by side. The top level lists names like hello, cats, gview, many, forums, chat, rest, getpost, session, favsql, favwc, favs, form, route, tagme, tracks, autos, bookone, bookmany, myarts, tenant, maintenance, usermodel, authz, menu, users, views, util, pics, zip and more. Each is a small project that does one thing.

That granularity is the point. A learner working through a course on sessions or on class based views can open one folder, read a models file, a views file and a template, and see the concept in isolation without wading through an application that also does authentication, file uploads and caching. For a developer who already knows Django and wants to fill a specific gap, the same property makes the repository a catalogue of patterns to copy.

There is a single manage.py at the root and one requirements52.txt, so the whole collection shares one environment and one migration history. That arrangement is convenient for the course, and it is the main structural limitation for reuse: the projects are siblings in one Django installation rather than independent, installable apps.

Branch per Django version, and how the port is done

Versioning is handled with branches rather than tags, and the README explains both the mechanism and the maintenance ritual.

The default branch is django52, and the title of the project is DJ4E Samples Django 5.2. If you want to work with Django 4.2, the README points you to the django42 branch and its own README file. The setup instructions even tell you how to detect the wrong environment: a shell prompt showing a django42 tag means you are in the old environment and should switch or set up the 5.2 one.

The porting procedure is described under the notes on new Django versions. When a new Django release arrives, likely once a year, you create a branch with the new version name, for example django62, and modify a short list of files. The README names them: README.md, home/templates/home/main.html and tools/checkup.sh. You then rename requirements52.txt to requirments62.txt and edit it as needed, with the misspelling carried over as written in the original instructions.

This is a small, readable upgrade policy, and it explains the shape of the repository. One branch per Django generation, one requirements file per branch, and a checkup script that presumably verifies the environment matches the branch. The repository is not archived and the last push was on 2026-10-19, so the branch scheme is still in active use, and there are no published GitHub releases because branches are the unit of distribution here.

Setting up the environment and running a sample locally

The README walks through the setup in careful order, aimed at a student on a shared hosting console.

First, deactivate any existing virtual environment and create a new one with a Python version that supports Django 5.2. The README says the version should be 3.10 to 3.13 with a preference for the later releases:

    cd ~
    deactivate  # May fail - this is OK
    python3.13 -m venv .ve52

If that command reports that python3.13 is not found, the README tells you to try 3.12, 3.11 and 3.10 until one works. Then activate it and check the interpreter:

    source ~/.ve52/bin/activate
    python --version

Next comes the dependency install, which the README warns can take a long time on a slow server and suggests running one command at a time:

    pip install --upgrade pip
    pip install django==5.2 ## this may take a couple of minutes
    python -m django --version

The version check should report at least 5.2. The README also shows how to configure git so your commits are attributed, which matters if you are working through the assignments that use git:

    git config --global user.email "[email protected]"
    git config --global user.name "Your name"

With the environment ready, the checkout and first run are:

    git clone https://github.com/csev/dj4e-samples.git
    cd dj4e-samples
    git checkout django52
    pip install -r requirements52.txt
    python -m django --version

    python manage.py migrate
    python manage.py createsuperuser --username dj4e-samples

On a local Mac, Linux or WSL machine you then start any of the sample applications with:

    cd dj4e-samples
    source .ve52/bin/activate   # If needed
    python manage.py runserver

and open http://localhost:8000.

Deploying on PythonAnywhere with a WSGI snippet

The README keeps a section on running the samples on PythonAnywhere, and says outright that you can always explore the code at samples.dj4e.com so you do not really need to run it there yourself. It is included as the maintainer's own setup notes.

The mechanics are still instructive. Once the code is checked out under your home directory and the migrations have run, you go to the Web tab and point two settings at your new application, with the source code and working directory both set to your home path. The WSGI configuration file then rewrites the path handling so the app can be imported:

    import os
    import sys

    path = os.path.expanduser('~/dj4e-samples')
    if path not in sys.path:
        sys.path.insert(0, path)
    os.environ['DJANGO_SETTINGS_MODULE'] = 'dj4e-samples.settings'
    from django.core.wsgi import get_wsgi_application
    from django.contrib.staticfiles.handlers import StaticFilesHandler
    application = StaticFilesHandler(get_wsgi_application())

Two things stand out. The settings module is named dj4e-samples.settings, which is the shared root project rather than any of the individual sample folders, and the static files handler is wrapped around the WSGI application so that assets are served correctly in that hosting model.

The README also notes that you can edit these files and settings in the Web tab to switch between projects, and that you must reload under the Web tab after every file or configuration change. That last point is the classic shared hosting constraint: no daemon to restart, so a configuration change only takes effect when you ask for the reload.

Development tools, assignments and the code you should not copy

The repository includes a tools directory described as helpful development tools for managing the development environment, with details in tools/README.md, and the porting notes reference tools/checkup.sh as one of the three files to update when moving to a new Django version. The top level also holds scripts/ and a TODO.md.

The git configuration step in the setup exists because some of the course assignments use git as the version control system, so the collection doubles as the assignment workspace. That context explains several things that would otherwise look odd in a general purpose sample repository: the specific superuser name in the commands, the shared home directory assumptions, and the emphasis on PythonAnywhere behaviour rather than on container deployment.

It also explains what should not be reused. The README prints a sample password next to the createsuperuser command, commented out with the changepassword line above it. That string is published in the repository, so it is not a credential to use anywhere. More generally, the settings in the shared root project are tuned for a teaching server, and the individual applications are demonstrations of one idea each rather than complete, hardened products.

The licence is MIT, with the licence text in LICENSE.md at the top of the repository, so the code can be reused and modified freely as long as the notice is kept.

Reading samples versus starting your own project

There is a real alternative to reading this collection, and it is the one most experienced Django developers would choose for production work: generate a fresh project with the Django tooling and build from there. The difference in approach is straightforward.

A generated project gives you one application, a clean settings module, a test structure and a dependency list that matches what you actually need, with nothing else in the tree. The dj4e-samples repository gives you dozens of working examples that share a single root project and a single settings module, which is ideal for reading and awkward to lift code out of, because the useful parts are entangled with the shared configuration.

The middle path is to use these samples as a reference library rather than a base. When a course assignment or a tutorial covers class based views, forms, sessions or authentication, look at the matching folder, copy the pattern into a project you created yourself, and adapt it. You get the pedagogic value without inheriting a settings module tuned for somebody else's server.

One more difference is worth naming. The samples are pinned to a Django generation and updated roughly yearly, which means a pattern you copy may reflect an idiom from Django 4.2 or 5.2 that has since changed. Check the branch you are reading against the Django version you are running, and treat the older code as history rather than as current best practice.

Editorial conclusion

Adopt dj4e-samples as a reading library and a set of starting points if you are working through the Django for Everybody course or you want small, self-contained Django applications to compare against your own code. Do not adopt it as an application to deploy, since each project is a teaching artefact with a shared root manage.py and no production settings, and do not reuse the superuser credentials printed in the README, since the password shown there is published in the repository. Before you rely on it, check that the branch you clone matches the Django version you intend to install, since the default branch is django52 and the Django 4.2 lessons sit on a separate django42 branch, and run python -m django --version after activating the environment to confirm what you actually have.

Frequently asked questions

Official sources

1. csev/dj4e-samples on GitHub
2. Issues
3. License: MIT
4. Project website
5. README