Dockerfile

# PURPOSE:
# Build a clean Docker image for KnowFlow AI.
#
# THIS IMAGE SUPPORTS:
# - Phase 2 CLI backend testing
# - JupyterLab development
# - Future Streamlit app execution
#
# Docker solves common local issues:
# - broken .venv
# - wrong Python version
# - missing packages
# - Ubuntu pip restrictions
# - Jupyter kernel conflicts
# ============================================================


# ============================================================
# 1. BASE IMAGE
# ============================================================
# python:3.11-slim is lightweight but still suitable for this app.
# ============================================================

FROM python:3.11-slim


# ============================================================
# 2. PYTHON ENVIRONMENT SETTINGS
# ============================================================
# PYTHONDONTWRITEBYTECODE=1:
# Prevents Python from writing .pyc files.
#
# PYTHONUNBUFFERED=1:
# Makes logs appear immediately in Docker output.
#
# PIP_NO_CACHE_DIR=1:
# Reduces Docker image size by not storing pip cache.
# ============================================================

ENV PYTHONDONTWRITEBYTECODE=1
ENV PYTHONUNBUFFERED=1
ENV PIP_NO_CACHE_DIR=1


# ============================================================
# 3. SET WORKING DIRECTORY
# ============================================================
# All project files live inside /app in the container.
# ============================================================

WORKDIR /app
ENV PYTHONPATH=/app

# ============================================================
# 4. INSTALL SYSTEM DEPENDENCIES
# ============================================================
# build-essential:
# Needed by some Python packages during installation.
#
# curl:
# Useful for health checks and debugging.
#
# git:
# Useful if packages need git-based installation.
# ============================================================

RUN apt-get update && apt-get install -y \
    build-essential \
    curl \
    git \
    && rm -rf /var/lib/apt/lists/*


# ============================================================
# 5. COPY REQUIREMENTS FIRST
# ============================================================
# Docker layer caching:
# If requirements.txt does not change, Docker can reuse this layer.
# ============================================================

COPY requirements.txt /app/requirements.txt


# ============================================================
# 6. INSTALL PYTHON PACKAGES
# ============================================================

RUN python -m pip install --upgrade pip setuptools wheel \
    && pip install -r /app/requirements.txt


# ============================================================
# 7. REGISTER JUPYTER KERNEL
# ============================================================
# This allows JupyterLab inside Docker to use a clean kernel.
# ============================================================

RUN python -m ipykernel install \
    --sys-prefix \
    --name knowflow-ai-docker \
    --display-name "Python (KnowFlow AI Docker)"


# ============================================================
# 8. COPY PROJECT FILES
# ============================================================
# .dockerignore controls what should NOT be copied.
# .env must NOT be copied into the image.
# ============================================================

COPY . /app


# ============================================================
# 9. CREATE RUNTIME FOLDERS
# ============================================================

RUN mkdir -p /app/data/raw \
    && mkdir -p /app/vector_db \
    && mkdir -p /app/outputs \
    && mkdir -p /app/logs


# ============================================================
# 10. EXPOSE PORTS
# ============================================================
# 8888 = JupyterLab
# 8501 = Streamlit
# ============================================================

EXPOSE 8888
EXPOSE 8501


# ============================================================
# 11. DEFAULT COMMAND
# ============================================================
# Starts JupyterLab for development.
#
# SECURITY NOTE:
# Token is disabled here only for local development.
# Do NOT expose this container publicly with token disabled.
# ============================================================

CMD ["jupyter", "lab", "--ip=0.0.0.0", "--port=8888", "--no-browser", "--allow-root", "--NotebookApp.token=", "--NotebookApp.password="]