# Embedding Service Dockerfile
# Handles ML-heavy operations: embeddings, re-ranking, PDF extraction

FROM python:3.11-slim

WORKDIR /app

# Install system dependencies for PDF extraction
RUN apt-get update && apt-get install -y --no-install-recommends \
    libmagic1 \
    poppler-utils \
    tesseract-ocr \
    tesseract-ocr-deu \
    && rm -rf /var/lib/apt/lists/*

# Install PyTorch CPU-only (smaller image)
RUN pip install --no-cache-dir torch --index-url https://download.pytorch.org/whl/cpu

# Copy and install requirements
COPY requirements.txt .
RUN pip install --no-cache-dir -r requirements.txt

# Note: Models are downloaded on first startup (not during build)
# This makes the build faster but first startup slower
# To pre-download models, mount a persistent volume for /root/.cache/huggingface

# Copy application code
COPY . .

# Health check
HEALTHCHECK --interval=30s --timeout=10s --start-period=60s --retries=3 \
    CMD python -c "import httpx; httpx.get('http://localhost:8087/health').raise_for_status()"

# Run the service
EXPOSE 8087
CMD ["uvicorn", "main:app", "--host", "0.0.0.0", "--port", "8087"]