Dockerfile

# Base Image
FROM python:3.11-slim

ENV PYTHONDONTWRITEBYTECODE=1
ENV PYTHONUNBUFFERED=1
# What it does: 
    # Uses an official Python image with Python 3.11 preinstalled. 
    # slim = a stripped-down Debian Linux with only essentials.

# Why it matters:
    # ✅ Smaller image size → faster builds → faster deploys → less bandwidth
    # ✅ Still compatible with apt-get
    # ❌ Not Alpine (which causes many Python dependency nightmares)

# Mental model: Give me a tiny Linux computer that already has Python 3.11 inside.

# Working directory
WORKDIR /app

# What it does:
# Creates /app folder inside the container
# Sets it as the default directory for:
    # COPY
    # RUN
    # CMD
    # everything

# Why it matters:
    # Keeps your container clean and organized
    # No files scattered in / like a junk drawer

# Mental model: “From now on, pretend the container lives inside /app.”


# System deps
RUN apt-get update && apt-get install -y --no-install-recommends \
    build-essential \
    && rm -rf /var/lib/apt/lists/*

# What it does:
    # apt-get update → Updates Linux package list
    # apt-get install -y build-essential
        # → Installs: gcc, g++
        # make (needed to compile Python packages like numpy, cryptography, etc.)
    # rm -rf /var/lib/apt/lists/*
        # → Deletes cached package list to reduce image size

# Why it matters:
    # Some Python packages are not pure Python → they compile C/C++
    # Without this: ❌ pip install fails with cryptic errors

# Mental model: “Install a compiler so Python packages can build their muscles.”

# Copy only requirements first
COPY requirements.txt .
# Copies only requirements.txt into container
# If your code changes but req.txt stays same, then docker reuses the pip install layer and build becomes 10x faster

# Mental mode: Install dependencies only when dependency list changes.

# Install dependencies
RUN pip install --no-cache-dir -r requirements.txt

ENV PYTHONDONTWRITEBYTECODE=1
ENV PYTHONUNBUFFERED=1
# Creates log folder
RUN mkdir -p /app/logs

# Copy app code
COPY . .
# Copies everything from your project folder into /app

# Expose port
EXPOSE 5000
# The container listens on port 8000
# Mental mode: Hey Platform, traffic comes in on 8000

# Run the app using gunicorn (production server)
CMD ["gunicorn", "app:app", "--workers", "1", "--threads", "1", "--timeout", "120", "--bind", "0.0.0.0:5000"]
# Breakdown:
    # gunicorn: Production WSGI server
    # -w 2 : 2 worker processes
    # -b 0.0.0.0:8000: listen on all interfaces, port 8000
    # app:app: app = Flask(__name__)