Dockerfile

# ===============================
# Base image
# ===============================
FROM python:3.10-slim

WORKDIR /app

# ===============================
# System dependencies + OCR libraries
# ===============================
RUN apt-get update && apt-get install -y --no-install-recommends \
    sqlite3 \
    ca-certificates \
    libgomp1 \
    tesseract-ocr \
    tesseract-ocr-eng \
    libglib2.0-0 \
    libsm6 \
    libxext6 \
    libxrender-dev \
    libgl1 \
    build-essential \
    gcc \
    g++ \
    python3-dev \
    && rm -rf /var/lib/apt/lists/*

# Verify Tesseract installation
RUN tesseract --version

# ===============================
# Copy requirements and upgrade pip
# ===============================
COPY requirements.txt .

RUN pip install --no-cache-dir --upgrade pip setuptools wheel cython

# ===============================
# Install NumPy first (prebuilt wheel)
# ===============================
RUN pip install --no-cache-dir numpy==1.24.4

# ===============================
# Install Pandas from prebuilt wheel (avoid --no-binary)
# ===============================
RUN pip install --no-cache-dir pandas==2.0.3 --only-binary=:all:

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

# ===============================
# Optional: Gemini SDK, EasyOCR
# ===============================
RUN pip install --no-cache-dir --upgrade google-generativeai google-ai-generativelanguage
RUN pip install --no-cache-dir easyocr opencv-python-headless Pillow pytesseract

# ===============================
# Copy application code
# ===============================
COPY . .

# ===============================
# Create necessary directories
# ===============================
RUN mkdir -p /app/data/logs /app/data/docs /app/backend/app/db && chmod -R 777 /app/data

# ===============================
# Create __init__.py files
# ===============================
RUN touch backend/__init__.py \
    && touch backend/feature_builder/__init__.py \
    && touch backend/app/__init__.py \
    && touch backend/app/api/__init__.py \
    && touch backend/app/agent/__init__.py \
    && touch backend/app/wrappers/__init__.py \
    && touch backend/app/db/__init__.py \
    && touch backend/ingest/__init__.py

# ===============================
# Verify agent files exist
# ===============================
RUN test -f backend/app/agent/agent_orchestrator.py || \
    (echo "ERROR: agent_orchestrator.py not found! Add it before building." && exit 1)

# ===============================
# Initialize database
# ===============================
COPY backend/app/db/db_init.py backend/app/db/db_init.py
RUN echo "🗄️  Initializing database during build..." && \
    python backend/app/db/db_init.py && \
    echo "✅ Database initialized successfully!"

# ===============================
# Expose port
# ===============================
EXPOSE 7860

# ===============================
# Startup script
# ===============================
RUN echo '#!/bin/bash\n\
echo "🔍 Checking database..."\n\
python backend/app/db/db_init.py\n\
echo "✅ Database ready"\n\
echo "🚀 Starting application..."\n\
exec uvicorn app:app --host 0.0.0.0 --port 7860 --timeout-keep-alive 75\n\
' > /app/start.sh && chmod +x /app/start.sh

CMD ["/app/start.sh"]