Skip to content

MartinStrobel/evaluator-images

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

77 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Coursemology Evaluator Images

These images are used to run code packages in controlled containers. The names of the images follow the namespacing as used in the polyglot Coursemology::Polyglot::Language subclasses.

The images are built on top of a base image, which installs make and configures the container command. Base images are tagged with the OS images they are built on: Debian 8 or 10. All other images install the required packages for a particular language.

Using the Images for Debugging

If your code cannot pass Coursemology's autograder, you can manually simulate the autograding process with these Docker images to help with debugging.

Install Docker to run the images on your own machine.

Pulling Images

Python

Python has separate tags for each version. The default tag latest does not exist.

You can see available versions in the python folder of this repository.

docker pull coursemology/evaluator-image-python:3.7

C++ and Java

There is only 1 image for each of these languages.

docker pull coursemology/evaluator-image-c_cpp
docker pull coursemology/evaluator-image-java

Preparing the Package Files

When editing a programming question on Coursemology, there is a link to a package zip file. Download and extract it.

Copy the code you want to debug into the files in the submission folder. More information about the package structure can be found at the evaluator-samples repository.

Getting a Shell

Use docker run to start a container with the same environment as Coursemology's autograder. Specify -it and --entrypoint=/bin/bash or the container will attempt to run the evaluation script and exit immediately.

For convenience when debugging, mount the host folder where the package files have been unzipped and prepared to /home/coursemology/package in the container.

docker run -it --entrypoint=/bin/bash --mount type=bind,source="$(pwd)",target=/home/coursemology/package coursemology/evaluator-image-java

Change to the /home/coursemology/package directory.

root@someid:/# cd /home/coursemology/package
root@someid:~/package#

Running the Tests

The coursemology-evaluate.sh script runs the sequence of steps for autograding.

In the package folder, run the following commands to run the public tests.

make prepare
make compile
make public

You can edit the files used by the test runner and run the tests again to debug. See the output from the prepare step to figure out the file is. Remember to run the compile step again for compiled languages.

Viewing the Test Report

Open the report.xml file generated in the package folder.

Other Makefile Commands

Running the private tests:

make private

Running the evaluation tests:

make evaluation

Cleaning up if the prepared file has been messed up:

make clean

About

Evaluator images for programming questions

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Dockerfile 60.0%
  • Shell 40.0%