Merge pull request geekan#29 from voidking/pr

Optimize Dockerfile and README

Dockerfile

@@ -1,28 +1,34 @@
+# This Dockerfile is friendly to users in Chinese Mainland :) 
+# For users outside mainland China, feel free to modify or delete them :)
+
+# Use a base image with Python 3.9.17 slim version (Bullseye)
 FROM python:3.9.17-slim-bullseye
 
-#Special gifts for mainland China users :) 
-#Change to your own preferenced mirrors if you wish, just uncomment them is fine too.
-    #Below is apt mirror setup.
+# Install Debian software needed by MetaGPT
 RUN sed -i 's/deb.debian.org/mirrors.ustc.edu.cn/g' /etc/apt/sources.list &&\
-    apt update &&\ 
+    apt update &&\
     apt install -y git curl wget build-essential gcc clang g++ make &&\
     curl -sL https://deb.nodesource.com/setup_19.x | bash - &&\
-    apt install -y nodejs 
+    apt install -y nodejs &&\
+    apt-get clean
 
-    
+# Set the working directory to /app
 WORKDIR /app
-    #This sets npm mirror url.
+
+# Install Mermaid CLI globally and clone the MetaGPT repository
 RUN npm config set registry https://registry.npm.taobao.org &&\
     npm install -g @mermaid-js/mermaid-cli &&\
+    npm cache clean --force &&\
     git clone https://github.com/geekan/metagpt
+
+# Install Python dependencies and install MetaGPT
 RUN cd metagpt &&\
     mkdir workspace &&\
-    #This sets the pip mirror url.
     pip config set global.index-url https://mirrors.aliyun.com/pypi/simple/ &&\
-    pip install -r requirements.txt --no-cache-dir &&\
+    pip install -r requirements.txt &&\
+    pip cache purge &&\
     python setup.py install
 
-RUN pip cache purge &&\
-    apt autoclean
-
+# Running with an infinite loop using the tail command
+CMD ["sh", "-c", "tail -f /dev/null"]
 

README.md

@@ -36,7 +36,7 @@ For example, if you type `python startup.py "Design a RecSys like Toutiao"`, you
 It requires around **$0.2** (GPT-4 api's costs) to generate one example with analysis and design, around **$2.0** to a full project.
 
 ## Installation
-
+### Traditional Installation
 ```bash
 # Step 1: Ensure that NPM is installed on your system. Then install mermaid-js.
 npm --version
@@ -51,6 +51,38 @@ cd metagpt
 python setup.py install
 ```
 
+### Installation by Docker 
+```bash
+# Step 1: Download metagpt official image and prepare config.yaml
+docker pull metagpt/metagpt:v0.1
+mkdir -p /opt/metagpt/config && docker run --rm metagpt/metagpt:v0.1 cat /app/metagpt/config/config.yaml > /opt/metagpt/config/config.yaml
+vim /opt/metagpt/config/config.yaml # Change the config
+
+# Step 2: Run metagpt image
+docker run --name metagpt -d \
+    -v /opt/metagpt/config:/app/metagpt/config \
+    -v /opt/metagpt/workspace:/app/metagpt/workspace \
+    metagpt/metagpt:v0.1
+
+# Step 3: Access the metagpt container
+docker exec -it metagpt /bin/bash
+
+# Step 4: Play in the container
+cd /app/metagpt
+python startup.py "Write a cli snake game"
+```
+
+The command `docker run ...` do the following things: 
+- Start metagpt container with default command `tail -f /dev/null`
+- Map host directory `/opt/metagtp/config` to container directory `/app/metagpt/config`
+- Map host directory `/opt/metagpt/workspace` to container directory `/app/metagpt/workspace` 
+
+### Build image by yourself
+```bash
+# You can also build metagpt image by yourself.
+cd metagpt && docker build --network host -t metagpt:v0.1 .
+```
+
 ## Configuration
 
 - Configure your `OPENAI_API_KEY` in any of `config/key.yaml / config/config.yaml / env`
@@ -66,24 +98,6 @@ cp config/config.yaml config/key.yaml
 | OPENAI_API_KEY # Replace with your own key | OPENAI_API_KEY: "sk-..."                  | export OPENAI_API_KEY="sk-..." |
 | OPENAI_API_BASE # Optional                            | OPENAI_API_BASE: "https://<YOUR_SITE>/v1" | export OPENAI_API_BASE="https://<YOUR_SITE>/v1"   |
 
-## Docker Setup
-
-You can also use docker to setup MetaGPT.
-```bash
-cd metagpt
-docker build --network host -t metagpt:<version> .
-```
-There are some changes of mirrors in the dockerfile, for users outside mainland China, feel free to modify or delete them :)
-
-You can also pull the image from dockerhub via `docker pull sablin39/metagpt:<TAG>`.
-
-To run the docker image, you can use the following command.
-```bash
-docker run -it -v <MetaGPT-config-dir>:/app/metagpt/config -v <Workspace-dir>:/app/metagpt/workspace metagpt:<version> <command>
-```
-This command mounts the `config` and `workspace` folder in the host machine. You should use absolute directory of these folders.
-
-
 ## Tutorial: Initiating a startup
 
 ```shell