# Use a slim Python 3.9 base image
FROM python:3.9-slim-buster@sha256:320a7a4250aba4249f458872adecf92eea88dc6abd2d76dc5c0f01cac9b53990

# Set environment variables
ENV PYTHONDONTWRITEBYTECODE=1 \
    PYTHONUNBUFFERED=1 \
    HF_HOME=/app/cache \
    HUGGINGFACE_HUB_DISABLE_XET=1 \
    PIP_NO_CACHE_DIR=1

# Set working directory
WORKDIR /app

# Install system dependencies
RUN apt-get update && apt-get install -y --no-install-recommends \
    build-essential \
    gcc \
    g++ \
    && rm -rf /var/lib/apt/lists/*

# Upgrade pip to the latest version
RUN pip install --upgrade pip

# Copy requirements first for better caching
COPY requirements.txt .

# Install dependencies from requirements.txt
RUN pip install --no-cache-dir -r requirements.txt || { echo "Failed to install requirements"; exit 1; }

# Pre-download models
RUN python -c "from sentence_transformers import SentenceTransformer; SentenceTransformer('all-MiniLM-L6-v2', cache_folder='/app/cache')" || { echo "Failed to pre-download SentenceTransformer model"; exit 1; }
RUN python -c "from transformers import AutoModelForCausalLM, AutoTokenizer; AutoModelForCausalLM.from_pretrained('distilgpt2', cache_dir='/app/cache'); AutoTokenizer.from_pretrained('distilgpt2', cache_dir='/app/cache')" || { echo "Failed to pre-download transformer model"; exit 1; }
# Verify cache contents
RUN ls -l /app/cache && \
    find /app/cache -type f || { echo "Cache verification failed"; exit 1; }

# Create directories, user, and set permissions
RUN mkdir -p /app/data /app/cache && \
    useradd -m myuser && \
    chown -R myuser:myuser /app && \
    chmod -R u+rw /app/data

# Copy all project files
COPY . .

# Set non-root user
USER myuser

# Expose port for Gunicorn
EXPOSE 8000

# Start Gunicorn with a single worker to reduce memory usage
CMD ["gunicorn", "--bind", "0.0.0.0:8000", "--workers", "1", "--timeout", "120", "app:app"]