A Django web application for storing and browsing bilingual (Hebrew / English) classical Jewish texts, powered by Sefaria data and ChatGPT-assisted translation.
Private Sources lets you build a personal digital library of classical Jewish texts with parallel Hebrew and English translations. You can organise texts into categories and works, then browse individual sections paragraph-by-paragraph in a clean bilingual layout.
New sections are added by pasting the output of a ChatGPT prompt into the built-in upload form. The app parses the numbered paragraphs automatically and stores them in the database.
- 📖 Browse texts organised by Category → Work → Section
- 🔤 Side-by-side Hebrew (RTL) / English (LTR) paragraph view
- ⬆ Upload new sections by pasting ChatGPT bilingual output
- ⬅ ➡ Previous / Next navigation between sections of a work
- 🔒 Upload restricted to authenticated users
- 🛠 Full Django admin interface for managing all content
- 📱 Responsive Bootstrap 5 design
Browse all categories at a glance.
See all works in a category.
List of sections within a work.
Read Hebrew and English paragraphs side by side.
Paste ChatGPT output to add a new section in seconds.
- Python 3.10 or newer
pipandvenv(included with Python)
# 1. Clone the repository
git clone https://github.com/binyominzeev/private-sources.git
cd private-sources
# 2. Create and activate a virtual environment
python -m venv venv
source venv/bin/activate # Windows: venv\Scripts\activate
# 3. Install dependencies
pip install -r requirements.txt
# 4. Apply database migrations
python manage.py migrate
# 5. Create an admin account
python manage.py createsuperuser
# 6. Start the development server
python manage.py runserverOpen http://localhost:8000/ in your browser.
The admin panel is at http://localhost:8000/admin/.
- Visit the homepage to see all categories.
- Click a category to see its works.
- Click a work to see its sections.
- Click a section to read the bilingual paragraphs.
- Use the ← Prev / Next → buttons to move between sections.
- Log in at
/admin/. - Create a Category (e.g. Torah).
- Create a Work inside that category (e.g. Genesis).
- Optionally create Sections and Paragraphs directly.
-
Ask ChatGPT to translate a passage and format the output like this:
1. עברית: Hebrew text of paragraph one. English: English translation of paragraph one. 2. עברית: Hebrew text of paragraph two. English: English translation of paragraph two. -
Log in and click ⬆ Upload in the navbar (or visit
/upload/). -
Select the Category, Work, and enter the Reference and Slug.
-
Paste the ChatGPT output into the text area.
-
Click Upload & Parse — the paragraphs are saved immediately.
Override these in your production environment (e.g. in a .env file loaded by your process manager):
| Variable | Default | Description |
|---|---|---|
DJANGO_SECRET_KEY |
(insecure default) | Django secret key — must be changed in production |
DJANGO_DEBUG |
True |
Set to False in production |
DJANGO_ALLOWED_HOSTS |
localhost 127.0.0.1 |
Space-separated list of allowed host names |
# Collect static files
python manage.py collectstatic
# Run with a production WSGI server (example: gunicorn)
pip install gunicorn
gunicorn private_sources.wsgi:application --bind 0.0.0.0:8000- Database: SQLite works fine for a personal library. For higher traffic, switch to PostgreSQL or MySQL in
private_sources/settings.py. - Static files: Serve the
staticfiles/directory via Nginx or a CDN. - HTTPS: Use a reverse proxy (Nginx / Caddy) with a TLS certificate from Let's Encrypt.
server {
listen 80;
server_name yourdomain.com;
return 301 https://$host$request_uri;
}
server {
listen 443 ssl;
server_name yourdomain.com;
ssl_certificate /etc/letsencrypt/live/yourdomain.com/fullchain.pem;
ssl_certificate_key /etc/letsencrypt/live/yourdomain.com/privkey.pem;
location /static/ {
alias /path/to/private-sources/staticfiles/;
}
location / {
proxy_pass http://127.0.0.1:8000;
proxy_set_header Host $host;
proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
proxy_set_header X-Forwarded-Proto $scheme;
}
}