Dredd

Automated quiz and code grader.

Running Dredd

To run dredd, you can do the following:

$ python3 dredd --logging=debug

This will start the dredd server with debug level logging.

Docker Containers

Internally, dredd uses docker to execute submitted code in an isolated container. The precise docker image is defined in the scripts/sandbox.sh script (ie. IMAGE).

Provided in this repository are two Dockerfiles:

• Dockerfile.submit: This provides an example of a simple docker container that can be used to submit quizzes and code to dredd.

• Dockerfile.code: This provides an example of a simple docker container that will be used to execute code in isolation. As such, it should contain the interpreters and compilers required by dredd and the scripts/run.py script.

Build:

$ docker build -t pbui/dredd:20170814      - < Dockerfile.submit     # GitLab CI
$ docker build -t pbui/dredd-code:20170825 - < Dockerfile.code       # Code run.py

Push:

$ docker push pbui/dredd:20170814
$ docker push pbui/dredd-code:20170825

Programming Languages

To add support for a programming language, you must do the following:

1. Update Dockerfile.code to include the interpreter or compiler.

2. Update scripts/run.py to include a rule for the programming language. For instance, to support C and Python, you need to specify the following:

   Language('C',                                       # Language
       'gcc -std=gnu99 -o {executable} {source} -lm',  # Compilation
       './{executable}',                               # Execution
       ('.c',)                                         # Extension
   ),
   Language('Python',                                  # Language
       '',                                             # Compilation
       'python3 {source}',                             # Execution
       ('.py',)                                        # Extension
   

3. Update tests/test_code_echo.sh to include a simple echo test for that language (ie. the program reads in the input and echos it back).