SCT feedback on specific arguments: would be really good for beta testing 07/05

[hugobowne]

hey guys! we're trying to reproduce something like this is pythonwhat": i.e., being able to provide tailored feedback for each argument passed to a function

we can't get it to work yet. have tried the following + other stuff. How would you do it?

I will beta test without it but would be great because makes student experience far better ;-)

tagging @franciscastro

[vvnkr]

@hugo You are correct, this currently doesn't work:

test_function("open", index=1, args=[0], ...)
test_function("open", index=1, args=[1], ...)

I remember telling you you could give argument-specific feedback like that in pythonwhat but I forgot it is disabled. This is because each function call can only be tested once, it gets blacklisted afterwards because of some implementation technicalities. @filipsch and me will discuss if this is desirable behaviour and fix it/provide an alternative.

[filipsch]

@vincentvankrunkelsven either way, we need a way to fix this: either we want to be able to pass a list that has the same length as the args in incorrect_msg, either we want the blacklisting to work better that you can check the same function call with different test_function() calls (this is how it's done in testwhat, mainly for backwards compatibility reasons).

[filipsch]

@hugobowne I'll look into it. Can't tell when will be fixed.

[hugobowne]

thanks you guys are great 👍

as stated, i will beta test irrespective of this functionality. of course, the sooner the better as it improves user experience so very much ;-)

PEACE!

[filipsch]

I'm working on a fix for this, but I guess it'll only be possible to specify multiple incorrect_msg by simply writing multiple test_function() calls where you each time specifically mention a single argument or keyword. If we would make it possible to specify multiple incorrect messages, then it's hard to figure out which message refers to which argument or which keyword.

[filipsch]

@hugobowne This is currently on staging, will be deployed to master as soon as Robin updates the docker-container (with python autocomplete and bokeh)

This fix allows you to use the syntax vincent used for argument-specific feedback

test_function("open", index=1, args=[0], ...)
test_function("open", index=1, args=[1], ...)

I've also added a piece of documentation on the test_function() article: https://github.com/datacamp/pythonwhat/wiki/test_function#example-3-custom-feedback

Peace out.