Dockerfile

# # Use an official Python runtime as a parent image
# FROM python:3.11-slim

# WORKDIR /app      #add by P

# # Set environment variables for Hugging Face cache
# 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 templates/ /app/templates 
# COPY utils/ /app/utils
# COPY app.py app.py
# COPY blocks/ /app/blocks/ 
# COPY generated_projects/ /app/generated_projects/ 
# COPY blocks/sprites/ /app/blocks/sprites/ 
# COPY blocks/Backdrops/ /app/blocks/Backdrops/ 
# COPY blocks/sound/ /app/blocks/sound/ 
# # COPY OUTPUTS/ /app/OUTPUTS
# COPY . /app/
# COPY . .

# # 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 necessary directories with correct permissions
# RUN mkdir -p /app/nltk_data /app/.config/matplotlib \
#  && mkdir -p /app/cache /app/data /app/logs /app/outputs /app/blocks \
#  && mkdir -p /app/outputs/DETECTED_IMAGE /app/outputs/SCANNED_IMAGE /app/outputs/EXTRACTED_JSON \
#  && chmod -R 777 /app/cache /app/blocks /app/data /app/logs /app/outputs /app/outputs/SCANNED_IMAGE /app/outputs/EXTRACTED_JSON /app/outputs/DETECTED_IMAGE \
#  && chmod -R 777 /app

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

# # Expose port and start application
# EXPOSE 7860
# # CMD ["python", "app.py"]
# CMD ["gunicorn", "--bind", "0.0.0.0:7860", "--workers", "2", "app:app"]
# #CMD ["gunicorn", "--bind", "0.0.0.0:7860", "--workers", "", "--timeout", "360", "app:app"]
# Dockerfile: Hugging Face CPU-optimized
# Base: slim Python image for smaller footprint (CPU-only)
FROM python:3.11-slim

WORKDIR /app

# Environment: Hugging Face cache + force CPU behavior
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 \
    # Force CPU-only for PyTorch/transformers
    CUDA_VISIBLE_DEVICES= \
    PYTORCH_ENABLE_MPS=0 \
    XDG_CACHE_HOME=/app/.cache \
    PYTORCH_NO_CUDA=1


# Copy only what's needed (avoid duplicate COPY . /app)
COPY requirements.txt ./requirements.txt
COPY app.py ./app.py
COPY templates/ ./templates/
COPY utils/ ./utils/
COPY blocks/ ./blocks/
COPY generated_projects/ ./generated_projects/

# Install system dependencies (minimal for CPU inference + common tools)
RUN apt-get update && apt-get install -y --no-install-recommends \
    fontconfig \
    fonts-dejavu-core \
    build-essential \
    curl \
    ca-certificates \
    libglib2.0-0 \
    libsm6 \
    libxext6 \
    libxrender-dev \
    libgl1 \
    tesseract-ocr \
    poppler-utils \
    && apt-get clean && rm -rf /var/lib/apt/lists/*

# Python deps
RUN pip install --upgrade pip && pip install --no-cache-dir -r requirements.txt

RUN python -m nltk.downloader -d /app/nltk_data punkt averaged_perceptron_tagger wordnet || true

# Create necessary directories with correct permissions
RUN mkdir -p /app/nltk_data /app/.config/matplotlib \
 && mkdir -p /app/cache /app/data /app/logs /app/outputs /app/blocks \
 && mkdir -p /app/outputs/DETECTED_IMAGE /app/outputs/SCANNED_IMAGE /app/outputs/EXTRACTED_JSON \
 && chown -R root:root /app \
 && chmod -R 755 /app/cache /app/data /app/logs /app/outputs
 
 RUN mkdir -p /app/.config/matplotlib \
    /app/cache /app/nltk_data /nltk_data \
    /app/.cache/fontconfig /root/.cache/fontconfig \
    && chmod -R 777 /app/.config/matplotlib /app/.cache /app/cache /app/nltk_data /nltk_data /root/.cache/fontconfig

RUN fc-cache -f -v || true

# Set Flask env
ENV FLASK_APP=app.py \
    FLASK_ENV=production

# Expose port
EXPOSE 7860

# Healthcheck (lightweight endpoint; change if your app uses a different path)
HEALTHCHECK --interval=30s --timeout=5s --start-period=10s --retries=3 \
  CMD curl -f http://localhost:7860/healthz || exit 1

# Run Gunicorn with single worker (CPU-friendly) and threads; no timeout
# Single worker avoids multiple processes duplicating model memory on CPU-heavy workloads.
CMD ["gunicorn", "app:app", "-b", "0.0.0.0:7860", "-w", "1", "--threads", "4", "-k", "gthread", "--timeout", "0", "--graceful-timeout", "0"]