Create Dockerfile

Author: FJiayang

Dockerfile

@@ -0,0 +1,71 @@
+#
+# Dockerfile for a feature-rich code-server with Conda and UV
+# Pre-configured for Asia/Shanghai timezone and Alibaba Cloud PyPI mirror.
+#
+
+# Step 1: Start from the official code-server base image.
+FROM codercom/code-server:latest
+
+# Step 2: Set arguments for tool versions.
+ARG MINIFORGE_VERSION=23.11.0-0
+ARG PYTHON_VERSION=3.11
+
+# Step 3: Define environment variables for tool paths and configurations.
+ENV CONDA_DIR=/opt/conda
+ENV UV_DIR=/home/coder/.local
+# Add bin directories to the system PATH.
+ENV PATH=${CONDA_DIR}/bin:${UV_DIR}/bin:${PATH}
+# CRITICAL: Set the timezone environment variable.
+ENV TZ=Asia/Shanghai
+# CRITICAL: Configure UV to use the Alibaba Cloud mirror by default.
+ENV UV_INDEX_URL=https://mirrors.aliyun.com/pypi/simple
+
+# Step 4: Switch to the ROOT user for system-level installations.
+USER root
+
+# Step 5: Install system dependencies, set timezone, and install Miniforge.
+RUN \
+    # Update package lists and install necessary tools
+    apt-get update && apt-get install -y --no-install-recommends \
+        wget \
+        curl \
+        git \
+        build-essential \
+        tzdata \
+    # Set the timezone non-interactively
+    && ln -snf /usr/share/zoneinfo/$TZ /etc/localtime && echo $TZ > /etc/timezone \
+    # Download and install Miniforge
+    && wget "https://github.com/conda-forge/miniforge/releases/download/${MINIFORGE_VERSION}/Miniforge3-${MINIFORGE_VERSION}-Linux-x86_64.sh" -O miniforge.sh \
+    && /bin/bash miniforge.sh -b -p ${CONDA_DIR} \
+    && rm miniforge.sh \
+    # Give the 'coder' user ownership of the conda directory
+    && chown -R coder:coder ${CONDA_DIR} \
+    # Clean up apt cache to keep the image size down
+    && apt-get clean \
+    && rm -rf /var/lib/apt/lists/*
+
+# Step 6: Switch back to the standard, non-root 'coder' user.
+USER coder
+
+# Step 7: As the 'coder' user, install user-specific tools and set up the default environment.
+RUN \
+    # Install 'uv' using its official script
+    curl -LsSf https://astral.sh/uv/install.sh | sh && \
+    \
+    # Create a default conda environment
+    conda create -n py${PYTHON_VERSION} python=${PYTHON_VERSION} -y && \
+    \
+    # Pre-install common packages into the new environment using 'uv'.
+    # This will now automatically use the Alibaba Cloud mirror due to the ENV var.
+    uv pip install --python=${CONDA_DIR}/envs/py${PYTHON_VERSION}/bin/python \
+        numpy \
+        pandas \
+        matplotlib \
+        scikit-learn \
+        jupyterlab && \
+    \
+    # Configure the shell to automatically activate this environment on login
+    echo "conda activate py${PYTHON_VERSION}" >> ~/.bashrc
+
+# Step 8: The base image's entrypoint will start code-server.
+# All environment variables (TZ, UV_INDEX_URL, PATH) will be inherited.