Dockerfile.python

# Dockerfile for Python Sentiment Analysis Service
FROM python:3.10-slim

# Set working directory
WORKDIR /app

# Install system dependencies
RUN apt-get update && apt-get install -y \
    gcc \
    g++ \
    git \
    curl \
    && rm -rf /var/lib/apt/lists/*

# Copy requirements first for better caching (using CPU-only version to avoid CUDA)
COPY requirements-cpu.txt requirements.txt

# Install Python dependencies
RUN pip install --no-cache-dir -r requirements.txt

# Download transformer models during build (not runtime)
# Commented out to prevent OOM on t3.micro during build
# RUN python -c "from transformers import pipeline; \
#     pipeline('sentiment-analysis', \
#     model='distilbert-base-uncased-finetuned-sst-2-english', \
#     device=-1)"

# Copy application code
COPY *.py ./

# Copy config directory
COPY config/ ./config/

# Copy PDF generation module
COPY pdf_generation/ ./pdf_generation/

# Copy Images directory for PDF diagrams
COPY Images/ ./Images/
# Copy tests directory
COPY tests/ ./tests/
COPY CANDIDATE_KNOWLEDGE_BASE.md ./CANDIDATE_KNOWLEDGE_BASE.md

# Create output directories
RUN mkdir -p my_volume/sentiment_analysis \
    my_volume/hf_model \
    extracted_text \
    cache \
    /tmp/hf_cache

# Set environment variables
ENV TRANSFORMERS_CACHE=/tmp/hf_cache
ENV HF_HOME=/tmp/hf_cache
ENV PYTHONUNBUFFERED=1

# Expose port for FastAPI
EXPOSE 8000

# Health check - give more time for ML models to load
HEALTHCHECK --interval=30s --timeout=10s --start-period=120s --retries=5 \
    CMD curl --fail http://localhost:8000/health || exit 1

# Run FastAPI wrapper with single worker (needed for in-memory job storage)
# Use --workers 1 or remove --workers flag entirely
CMD ["uvicorn", "main_api:app", "--host", "0.0.0.0", "--port", "8000"]