Dockerfile

# Base image with Python and common dependencies
FROM python:3.10-slim

WORKDIR /app

ENV DEBIAN_FRONTEND=noninteractive \
    PYTHONUNBUFFERED=1 \
    PYTHONDONTWRITEBYTECODE=1 \
    TRANSFORMERS_CACHE=/app/cache \
    HF_HOME=/app/cache \
    NLTK_DATA=/app/nltk_data \
    MPLCONFIGDIR=/app/.config/matplotlib

COPY requirements.txt requirements.txt 
COPY extract_img_pdf.py extract_img_pdf.py
COPY live_streaming_flask.py live_streaming_flask.py
COPY templates/ /app/templates 
COPY reference/ /app/reference
COPY .env .env
COPY test_streaming.py test_streaming.py 
COPY app_main.py app_main.py

# Install system dependencies
RUN apt-get update && apt-get install -y \
    build-essential \
    libglib2.0-0 \
    libsm6 \
    libxext6 \
    libxrender-dev \
    tesseract-ocr \
    poppler-utils \
    libgl1 \
    ffmpeg \
    libopencv-dev \
    curl \
    ca-certificates \
    && apt-get clean \
    && rm -rf /var/lib/apt/lists/*

RUN pip install --upgrade pip && pip install -r requirements.txt

# Create writable paths
RUN mkdir -p /app/nltk_data /app/.config/matplotlib \
 && mkdir -p /app/cache /app/data /app/OUTPUTS /app/outputs \
 && mkdir -p /app/outputs/DETECTED_IMAGE /app/outputs/SCANNED_IMAGE /app/outputs/EXTRACTED_JSON \
 && chmod -R 777 /app

EXPOSE 7860

CMD ["python", "app_main.py"]