Dockerfile

FROM python:3.9-slim

WORKDIR /app

# Install system dependencies
RUN apt-get update && apt-get install -y \
    build-essential \
    gcc \
    python3-dev \
    cython3 \
    libopenblas-dev \
    liblapack-dev \
    tesseract-ocr \
    libtesseract-dev \
    libleptonica-dev \
    pkg-config \
    zlib1g-dev \
    && rm -rf /var/lib/apt/lists/*

# Set environment variables
ENV CFLAGS="-Wno-error=implicit-function-declaration"
ENV PYTHONUNBUFFERED=1
ENV PIP_NO_CACHE_DIR=1
ENV PIP_PREFER_BINARY=1
ENV FLASK_RUN_HOST=0.0.0.0
ENV FLASK_RUN_PORT=5001

# Copy requirements first for better caching
COPY requirements.txt .

# Remove zlib from requirements.txt as it's a system package
RUN sed -i '/zlib==1.0/d' requirements.txt

# Install Python packages in specific order
RUN pip install --no-cache-dir --upgrade pip && \
    pip install --no-cache-dir numpy==1.22.1 && \
    pip install --no-cache-dir scipy==1.7.3 && \
    pip install --no-cache-dir scikit-learn==1.0.2 && \
    pip install --no-cache-dir -r requirements.txt && \
    python -m spacy download en_core_web_sm && \
    python -m nltk.downloader punkt

COPY . .

# Expose the port
EXPOSE 5001

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