Dockerfile

# Use an official Python runtime as a parent image
FROM python:3.9-slim

# Set environment variables for Python
ENV PYTHONDONTWRITEBYTECODE 1
ENV PYTHONUNBUFFERED 1

# Set environment variables for Hugging Face cache to /tmp (which is writable)
ENV TRANSFORMERS_CACHE=/tmp/huggingface
ENV HF_HOME=/tmp/huggingface

# Ensure permissions are correct for /tmp/huggingface (optional, but just in case)
RUN mkdir -p /tmp/huggingface && chmod -R 777 /tmp/huggingface

# Create cache directories for Hugging Face and Matplotlib
RUN mkdir -p /app/.cache/huggingface /app/.cache/matplotlib

# Set environment variables for Hugging Face and Matplotlib cache
ENV TRANSFORMERS_CACHE=/app/.cache/huggingface
ENV HF_HOME=/app/.cache/huggingface
ENV MPLCONFIGDIR=/app/.cache/matplotlib

# Adding permission for the cache
RUN chmod -R 777 /app/.cache

# Give write permissions to the /app directory
RUN chmod -R 777 /app

# Create /app/logs directory and set permissions for logging
RUN mkdir -p /app/logs && chmod -R 777 /app/logs

# Set the working directory
WORKDIR /app

# Install system dependencies, including libgomp
RUN apt-get update && apt-get install -y \
    libgl1-mesa-glx \
    libgomp1 \
    libglib2.0-0 \
    && rm -rf /var/lib/apt/lists/*

# Set environment variables for Matplotlib and Hugging Face cache directories
ENV MPLCONFIGDIR=/tmp/matplotlib
ENV TRANSFORMERS_CACHE=/tmp/transformers_cache

# Copy the requirements file into the container at /app
COPY requirements.txt /app/

# Install any needed packages specified in requirements.txt
RUN pip install --no-cache-dir --upgrade -r requirements.txt

# Create directories for session storage, uploads, and cache
RUN mkdir -p /app/flask_sessions /app/uploads /tmp/matplotlib /tmp/transformers_cache && chmod -R 777 /app/flask_sessions /app/uploads /tmp/matplotlib /tmp/transformers_cache

# Set permission for the Paddle OCR
ENV PADDLEOCR_MODEL_DIR=/tmp/.paddleocr
RUN mkdir -p /tmp/.paddleocr && chmod -R 777 /tmp/.paddleocr

# Set environment variable for PaddleOCR to use /tmp/.paddleocr
ENV PADDLEOCR_HOME=/tmp/.paddleocr
RUN mkdir -p /tmp/.paddleocr && chmod -R 777 /tmp/.paddleocr

# Copy the rest of the application code to /app
COPY . /app/

# Expose the port that the app runs on
EXPOSE 7860

# Set environment variables for Flask
ENV FLASK_APP=app.py
ENV FLASK_ENV=production

# Command to run the Flask app using Gunicorn with 1 worker
CMD ["gunicorn", "--workers=1", "--bind=0.0.0.0:7860", "--timeout=120", "app:app"]