在 Docker 容器外运行 python 文件，该文件在 Docker 容器外导入 python 文件

Question

So I have a simple Dockerfile:

FROM ubuntu:latest
WORKDIR /home
RUN apt-get -y update 
RUN apt install -y software-properties-common
RUN add-apt-repository ppa:deadsnakes/ppa 
RUN apt install -y python3

It runs good with:

sudo docker build -t sample-image .

First I run my docker:

sudo docker run -t -d --name mycontainername sample-image:latest

If I run:

sudo docker exec -i mycontainername python3 < helloworld.py 

It works fine:

hello from docker

And then, I want to run a script a.py that imports a class_b from b.py. Note that a.py, b.py and helloworld.py are only in mi local computer, so they does not exist inside the container . a.py, b.py and helloworld.py are in the same folder

sudo docker exec -i mycontainername python3 < a.py

It says

ModuleNotFoundError: No module named 'b'

a.py:

import b
print("Hello world")

I think that the module 'b' is searched in the directory /home inside the docker. But putting the module inside the docker is not the solution I want, since I'm constantly editing that file

I'm constantly modifying a.py and b.py so I would want a command to run a.py with a command line like the following (I don't want to create the entire container over and over whenever I make a little change in the scripts):

sudo docker exec -i mycontainername python3 < a.py

Is there any command to do this? Thanks in advance

Answer 1

I agree with David that maybe you don't need to involve docker here, but if you really do then volumes will do what you need. You can map a directory in your local machine to a directory in the running container.

For example, if I create a directory called src and in it I create two files - a.py:

import b
print("hello from a")

and b.py

print("hello from b")

Now run a container from your image with local src mapped to your container working directory:

docker run -it -d --name mycontainername -v $(pwd)/src:/home sample-image:latest

now you can run a.py in the container:

docker exec -i mycontainername python3 a.py

you should see

hello from b
hello from a

All good so far. Now on your local machine, edit src/b.py or src/a.py and try that docker exec on the same running container, and the changes will have taken effect in the container.