[DOCKER] Use OpenSeesPy regardless of OS or Python version

(motivation)

When I tried to use OpenSeesPy, the environment construction method differs depending on the OS, it can not be used depending on the Python version, or even if it can be used, the VS Code interpreter does not interpret the OpenSeesPy method. I had a hard time before I started. Therefore, I thought of a way to easily build an environment regardless of them.

About this article

• Use Docker to build an environment regardless of OS
• Use Remote Development in VSCode Easy environment construction without using Docker command
• OpenSeesPy? I wrote the code so that it will be installed, so you can use it

What is OpenSees Py?

OpenSees is an abbreviation for Open System for Earthquake Engineering Simulation, which is an open source framework used for structural analysis of structures and ground during earthquakes. OpenSeesPy is an interpreter for using it in Python.

What is Docker

Docker is a tool that allows you to build the same environment in different environments by container virtualization. If you describe the construction method in a file, it will create and execute not only the OS but also the installation of the specified version of Python and other libraries with docker. With Dockern.

procedure

Things necessary

Docker Desktop https://www.docker.com/products/docker-desktop

VS Code https://code.visualstudio.com/

Remote Development (installed in VS Code) https://marketplace.visualstudio.com/items?itemName=ms-vscode-remote.vscode-remote-extensionpack

Prepare Dockerfile

Docker prepares a file named Dockerfile and builds the environment based on it. This time, I prepared as follows. The main thing I'm doing is

1. Python installation (version 3.8, including linux environment)
2. Install pip
3. Install the required packages for pypi

# Dockerfile
FROM python:3.8

RUN mkdir -p /usr/src/app
WORKDIR /usr/src/app

RUN apt-get update
RUN apt-get install -y --no-install-recommends \
    python3-dev \
    python3-pip

ADD . /usr/src/app/
RUN pip install -r requirements.txt

CMD bash

Prepare requirements.txt

As described in the Dockerfile, pip install -r requiremennts.txt will install the package, so write the required package in advance. This time I want to start OpenSeesPy, so don't forget openseespy. (Well, it's no longer limited to OpenSeesPy)

# requrements.txt (example)
matplotlib==3.3.2
numpy==1.19.2
openseespy==3.2.2.5
pylint==2.6.0

Create some folders for Dockerfile and requirements.txt and put them in it.

Dockern

Open it in VS Code and press the blue button at the bottom left of VS Code (see figure below). Then select "Remote-Containers: Open Folder in Container ..." and select the folder with the Dockerfile and requirements.txt to complete the environment construction!

I'd like to say that VS Code still doesn't understand the OpenSeesPy methods. After opening it, you should see a folder called .devcontainer and a file called devcontainer.json in it. Plus, add the following description and you're done!

# devcontainer.json
{
    ...
	"settings": { 
		...
		"python.linting.pylintArgs": ["--extension-pkg-whitelist=openseespy.opensees"] <-add to
	},
	"extensions": [
		"ms-python.python" <-add to
	]

}

Completed code

I have already prepared the finished product, so I will replace it. (Cooking show style) https://github.com/kakemotokeita/openseespy-docker-vscode

Please use it if you like. I have included the ones I personally like, so please delete the ones you don't need.

Please kindly let me know if there are any mistakes or problems.

By the way, OpenSees can analyze the structure of a solid, but if you only want to analyze the vibration of a mass system, we created a package that can be used with simple settings. If you like, please use this as well. https://github.com/adc21/asva

Thank you.