I'm starting to write a few questions for an assignment, where each question builds on the previous one: they first have to write 2 independent functions A and B, then a third function C that calls A and B, then have to run a program that calls C.
Without CodeRunner, I'd give them a code file to fill in and hope for the best, i.e. that they don't get stuck with A and B. With CodeRunner, each can be a separate question, so they can do C even if were unable to do A or B, because I can put the solutions to A and B as hidden support files for C. So far so good.
But for the last question, I really just want them to run the program with their own input and then report whether the output makes sense - it's an information retrieval type of program. I can put the program in a support file and have them write as the answer
result = process('...')
where process is the main function of the program, and '...' is their chosen input. The single test can then just do print(result), but since I don't know what is the expected output, the test will always fail. Can the grader be sidestepped so that the output is printed without being graded?
I could of course hand out the program as a .pyc file so that the answers to A, B, C are not revealed, and hope nobody finds out about decompilers, but having it all in CodeRunner to avoid switching contexts/media would be better.