Currently, when a new docker container is initialized, a new terminal window is used in RStudio. When the docker container is exited, the container is removed from docker (it's a temporary container) but the terminal window remains open in RStudio. This isn't a huge issue, but it due to the way dockerterm labels terminal windows, if a user starts a new Docker container with the same image without previously closing the open terminal an error will occur.
Handling the closing of the terminal window within dockerterm may entail keeping the app running until explicitly told to exit, at which point the container is shut down and the terminal window is closed. This solution would freeze the user's host R session while the application was being run, but the user would still be able to run R code against the docker container. The question is, is this desirable functionality or is it better to allow the user to run R code against the host R session and the docker R session?
Currently, when a new docker container is initialized, a new terminal window is used in RStudio. When the docker container is exited, the container is removed from docker (it's a temporary container) but the terminal window remains open in RStudio. This isn't a huge issue, but it due to the way dockerterm labels terminal windows, if a user starts a new Docker container with the same image without previously closing the open terminal an error will occur.
Handling the closing of the terminal window within dockerterm may entail keeping the app running until explicitly told to exit, at which point the container is shut down and the terminal window is closed. This solution would freeze the user's host R session while the application was being run, but the user would still be able to run R code against the docker container. The question is, is this desirable functionality or is it better to allow the user to run R code against the host R session and the docker R session?