Dockerfile

# Use the official Python 3.11 slim image for better compatibility
FROM python:3.11-slim

# Set environment variables for optimal Python and Gradio behavior
ENV PYTHONUNBUFFERED=1
ENV PYTHONDONTWRITEBYTECODE=1
ENV PIP_NO_CACHE_DIR=1
ENV PIP_DISABLE_PIP_VERSION_CHECK=1
ENV DEBIAN_FRONTEND=noninteractive

# Create app user for security (but run as root for HF Spaces compatibility)
RUN useradd --create-home --shell /bin/bash app

# Set the working directory inside the container
WORKDIR /app

# Install system dependencies required for multi-user AI application
RUN apt-get update && apt-get install -y --no-install-recommends \
    # Build tools
    build-essential \
    gcc \
    g++ \
    make \
    cmake \
    pkg-config \
    # Network and download tools
    curl \
    wget \
    git \
    # Development libraries
    libffi-dev \
    libssl-dev \
    # Image processing libraries
    libjpeg-dev \
    libpng-dev \
    libfreetype6-dev \
    libtiff5-dev \
    libopenjp2-7-dev \
    # Document processing libraries
    libxml2-dev \
    libxslt1-dev \
    zlib1g-dev \
    # OCR and PDF processing
    tesseract-ocr \
    tesseract-ocr-eng \
    poppler-utils \
    # SQLite for session storage
    sqlite3 \
    libsqlite3-dev \
    # Cleanup
    && apt-get clean \
    && rm -rf /var/lib/apt/lists/* \
    && rm -rf /var/cache/apt/*

# Upgrade pip to latest version
RUN python -m pip install --upgrade pip setuptools wheel

# Set pip configuration for better performance and reliability  
RUN pip config set global.trusted-host "pypi.org files.pythonhosted.org pypi.python.org" \
    && pip config set global.no-cache-dir true \
    && pip config set global.disable-pip-version-check true

# Copy the requirements file first to leverage Docker's build cache
COPY requirements.txt .

# Install Python dependencies with optimizations for concurrent usage
RUN pip install --no-cache-dir --upgrade -r requirements.txt \
    && pip install --no-cache-dir \
        # Additional packages for multi-user support
        gunicorn \
        uvloop \
        # Performance monitoring
        psutil \
    && pip list --outdated

# Copy the rest of the application code
COPY . .

# Create comprehensive directory structure for multi-user application
RUN mkdir -p \
    # Core application directories
    temp logs uploads downloads cache \
    # Multi-user session directories (will be created dynamically)
    /tmp/data_extractor_temp \
    # WebSocket and terminal directories
    static \
    # Database directory for session storage
    data \
    && chmod -R 755 /app

# Set optimized permissions for multi-user concurrent access
RUN chmod -R 777 temp logs uploads downloads cache /tmp \
    && chmod -R 755 static \
    && chmod 755 app.py \
    && chmod -R 755 config utils workflow models

# Create non-root user but keep root permissions for HF Spaces
RUN chown -R app:app /app \
    && chown -R app:app /tmp/data_extractor_temp

# Set comprehensive environment variables for multi-user application
ENV PYTHONPATH=/app
ENV GRADIO_SERVER_NAME=0.0.0.0
ENV GRADIO_SERVER_PORT=7860
# ENV GRADIO_SHARE=false
# ENV GRADIO_DEBUG=false

# Matplotlib configuration for headless operation
ENV MPLBACKEND=Agg
ENV MPLCONFIGDIR=/tmp/mpl_cache

# Optimize for multi-user concurrent access
ENV GRADIO_QUEUE_DEFAULT_CONCURRENCY=10
ENV GRADIO_MAX_THREADS=20

# WebSocket and networking configuration  
ENV WEBSOCKET_HOST=0.0.0.0
ENV WEBSOCKET_PORT=8765

# Session and temporary file configuration
ENV TEMP_DIR=/tmp/data_extractor_temp
ENV SESSION_TIMEOUT=1800
ENV MAX_FILE_SIZE_MB=50

# AI model configuration (will be overridden by user env vars)
ENV COORDINATOR_MODEL=gemini-2.5-pro
ENV DATA_EXTRACTOR_MODEL=gemini-2.5-pro
ENV DATA_ARRANGER_MODEL=gemini-2.5-pro
ENV CODE_GENERATOR_MODEL=gemini-2.5-pro

# Security and performance settings
ENV PYTHONSAFEPATH=1
ENV PYTHONHASHSEED=random

# Expose the port that the Gradio application will run on
EXPOSE 7860
EXPOSE 8765

# Health check for container monitoring
HEALTHCHECK --interval=30s --timeout=30s --start-period=60s --retries=3 \
    CMD curl -f http://localhost:7860/ || exit 1

# Run as root for Hugging Face Spaces compatibility
USER root

# Create startup script for better error handling and logging
RUN echo '#!/bin/bash\n\
set -e\n\
echo "🚀 Starting Data Extractor Multi-User Application..."\n\
echo "📊 Python version: $(python --version)"\n\
echo "🌐 Server: 0.0.0.0:7860"\n\
echo "👥 Multi-user concurrency: Enabled"\n\
echo "🔒 Session isolation: Active"\n\
echo "💾 Temp directory: $TEMP_DIR"\n\
\n\
# Create runtime directories\n\
mkdir -p "$TEMP_DIR"\n\
mkdir -p /tmp/mpl_cache\n\
chmod 777 "$TEMP_DIR" /tmp/mpl_cache\n\
\n\
# Start the application\n\
exec python app.py\n\
' > /app/start.sh && chmod +x /app/start.sh

# The command to run when the container starts
CMD ["/app/start.sh"]