py4phy / classroom-generator Goto Github PK

Output total regular points and total extra credit points.

At the moment, all points are added and reported.

BUILD_all.sh contains poor heuristics for copying starter code and README.md. etc. It does not copy example directories.

All problem assets should be handled by generate_tests.py and listed in the generate.yml file.

update use of https://github.com/emibcn/badge-action (getting a warning that it uses node.js 12 functions, maybe fixed in latest)

If a test is marked as extra then failing this test will always fail the whole workflow and thus lead to a workflow banner that indicates failure.

This can confuse students who know that they don't need the extra work to get full points (and indeed, the point banner shows that they have NN/NN points).

Failing extra tests should not fail the whole workflow or at least the "passing" banner should be shown if all regular tests passed. See also #3 .

I want to be able to generate tests that take a variable defined in a file and pass these to an (arbitrary) function that I define in generate.yml and then compare the output to a reference given in the generate.yml.

Something like "provide indices a and b so that you can extract the list ["fee"] from ["foo", "fee", "fum"]".

1. get a and b from student work (eg a=1, b=2)
2. define lambda a, b: ["foo", "fee", "fum"][a:b]
3. compare to desired output (e.g. ["fee"])

See

• https://github.com/Py4Phy/phy432-assignments/issues/1
• https://github.com/Py4Phy/phy432-assignments/issues/3

Create a Python package that can be installed with the standard mechanisms (such as pip).

• replace BUILD_all.sh with a proper Python script
• add standard package infrastructure
• put package on pypi