rctp_fcst/Dockerfile

39 lines
900 B
Docker

# Use a specialized uv image for faster dependency management
FROM ghcr.io/astral-sh/uv:python3.14-bookworm-slim AS builder
# Set the working directory
WORKDIR /app
# Enable bytecode compilation
ENV UV_COMPILE_BYTECODE=1
# Copy from the cache instead of linking since it's a multi-stage build
ENV UV_LINK_MODE=copy
# Install dependencies first (for caching)
COPY pyproject.toml uv.lock ./
RUN --mount=type=cache,target=/root/.cache/uv \
uv sync --frozen --no-install-project --no-dev
# Final stage
FROM python:3.14-slim-bookworm
WORKDIR /app
# Copy the environment from the builder
COPY --from=builder /app/.venv /app/.venv
# Copy the application code
COPY . .
# Set environment variables
ENV PATH="/app/.venv/bin:$PATH"
ENV PYTHONUNBUFFERED=1
ENV TZ=Asia/Taipei
# Create directory for downloads
RUN mkdir -p /app/Downloads
# The command to run the bot
CMD ["python", "bot/tg_bot.py"]