Dockerfile

FROM python:3.11-slim

# Set environment variables
ENV PYTHONUNBUFFERED=1 \
    PYTHONDONTWRITEBYTECODE=1 \
    HF_HOME=/tmp/huggingface_cache \
    PORT=7860

# Install system dependencies
RUN apt-get update && apt-get install -y --no-install-recommends \
    build-essential \
    git \
    ffmpeg \
    libsndfile1 \
    && rm -rf /var/lib/apt/lists/*

# Set working directory
WORKDIR /app

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

# Create necessary folders and grant open permissions (required for Hugging Face non-root user 1000)
RUN mkdir -p /app/data /app/data/users /app/data/qdrant /tmp/huggingface_cache && \
    chmod -R 777 /app/data /tmp/huggingface_cache

# Copy the rest of the application code
COPY . .

# Run a python script to pre-download the ML models to the cache folder during Docker build
# This makes container startup immediate and avoids download timeouts.
COPY scripts/preload_models.py /tmp/preload_models.py
RUN python /tmp/preload_models.py

# Set permissions for the app folder again
RUN chmod -R 777 /app

# Expose port 7860 for Hugging Face Spaces
EXPOSE 7860

# Command to run uvicorn
CMD ["uvicorn", "app.main:app", "--host", "0.0.0.0", "--port", "7860"]