Refactor Dockerfile to implement multi-stage builds for cleaner and more efficient setup

Dockerfile

@@ -1,66 +1,46 @@
-FROM python:3.13.5-slim
+# ---- Builder Stage ----
+FROM python:3.13.3-slim-bookworm AS builder
 
-WORKDIR /app
+WORKDIR /virtualenvs
 
-RUN apt-get update && apt-get install -y \
-    build-essential \
-    curl \
-    git \
- && rm -rf /var/lib/apt/lists/*
-
-# Install uv by copying from the official Docker image
+# Install uv
 COPY --from=ghcr.io/astral-sh/uv:latest /uv /uvx /bin/
 
-# Copy the project metadata for dependency resolution
-COPY pyproject.toml uv.lock /app/
+ENV PYTHONUNBUFFERED=1 \
+    PYTHONDONTWRITEBYTECODE=1 \
+    UV_COMPILE_BYTECODE=1 \
+    UV_LINK_MODE=copy \
+    PATH="/virtualenvs/.venv/bin:$PATH"
+
+# Copy dependency files
+COPY pyproject.toml uv.lock /virtualenvs/
 
-# Install dependencies into a virtual environment
-# Use cache mount for better build performance
+# Install dependencies (excluding project itself)
 RUN --mount=type=cache,target=/root/.cache/uv \
-    cd /app && \
+    cd /virtualenvs && \
     uv sync --frozen --no-install-project
 
-# Set up the virtual environment
-ENV VIRTUAL_ENV=/app/.venv
-ENV PATH="/app/.venv/bin:$PATH"
+# ---- Development Stage ----
+FROM python:3.13.3-slim-bookworm AS development
 
-# Route all caches/configs to /app and set HOME for Streamlit and others
-ENV HOME=/app \
-    STREAMLIT_HOME=/app/.streamlit \
-    STREAMLIT_CACHE_DIR=/app/.streamlit/cache \
-    STREAMLIT_USER_PATH=/app/.streamlit \
-    STREAMLIT_DISABLE_TELEMETRY=1 \
-    PYTHONUNBUFFERED=1 \
+WORKDIR /src
+
+ENV PYTHONUNBUFFERED=1 \
     PYTHONDONTWRITEBYTECODE=1 \
-    UV_COMPILE_BYTECODE=1 \
-    UV_LINK_MODE=copy \
-    HF_HOME=/app/.cache/huggingface \
-    TRANSFORMERS_CACHE=/app/.cache/huggingface/transformers \
-    TORCH_HOME=/app/.cache/torch \
-    XDG_CONFIG_HOME=/app/.config \
-    XDG_CACHE_HOME=/app/.cache \
-    EASY_OCR_CACHE_DIR=/app/.EasyOCR
+    VIRTUAL_ENV=/virtualenvs/.venv \
+    PATH="/virtualenvs/.venv/bin:$PATH"
 
-# Create writable dirs for all caches/configs, including EasyOCR
-RUN mkdir -p \
-    /app/.streamlit/cache \
-    /app/.cache/huggingface/transformers \
-    /app/.cache/torch \
-    /app/.config \
-    /app/.cache \
-    /app/.EasyOCR \
- && chmod -R 777 /app/.streamlit /app/.cache /app/.config /app/.EasyOCR
+# Install system dependencies
+RUN apt-get update && apt-get install -y netcat-openbsd && rm -rf /var/lib/apt/lists/*
+
+# Install uv in the runtime stage
+COPY --from=ghcr.io/astral-sh/uv:latest /uv /uvx /bin/
 
-# Fallback for Marker: redirect any site-packages "static" to /app/static (writable)
-RUN mkdir -p /app/static && \
-    if [ -e /usr/local/lib/python3.13/site-packages/static ]; then \
-      rm -rf /usr/local/lib/python3.13/site-packages/static; \
-    fi && \
-    ln -s /app/static /usr/local/lib/python3.13/site-packages/static && \
-    chmod -R 777 /app/static
+# Copy virtual environment from builder stage
+COPY --from=builder ${VIRTUAL_ENV} ${VIRTUAL_ENV}
 
-# Copy app
-COPY src/ ./src/
+# Copy application code
+COPY . /src
 
 EXPOSE 8501
 HEALTHCHECK CMD curl --fail http://localhost:8501/_stcore/health || exit 1