search

blockpy-edu / blockpy

Blockly that's more Python than JavaScript, powered with Skulpt
Apache License 2.0
398 stars 133 forks source link

Feedback Engine Error while running the sample problems #40

Closed xwying closed 4 years ago

xwying commented 5 years ago

Hi, I've been encountering some errors while I was trying to run the sample problems provided here: https://canvas.instructure.com/courses/1134562

When I run the problem 3 - 9, it always shows an error that was in give_feedback.py, however, the errors are different for each problem.

For example, when I run the problem 3, it throws:

Feedback: Internal Error 
NameError: name 'calls_function' is not defined on line 36
Feedback Engine Error
Critical Error in BlockPy's feedback generation. Please show the above message to an instructor so they can contact a developer!
The error was in give_feedback.py.

Any idea about how to fix this?

acbart commented 5 years ago

Looks like it's using the old style API. That example site needs to be updated to use the new API from Pedal. Unfortunately, the semester just started here at UD and I don't have time to fix them up. I'll keep this issue open and hopefully we can get around to it sometime soon.

xwying commented 5 years ago

Looks like it's using the old style API. That example site needs to be updated to use the new API from Pedal. Unfortunately, the semester just started here at UD and I don't have time to fix them up. I'll keep this issue open and hopefully we can get around to it sometime soon.

Thanks for your quick reply. I was also thinking about the same thing.

Actually, I posted this issue because I'm TA-ing a course in this semester and the professor integrated the BlockPy into her Canvas course. So this problem seems to affect all the assignments in our course that use BlockPy. Basically, the students will see the correct outputs on the console but will not get any credit for it. Do you have any suggestion about this? For example, is there any way to temporarily disable the feedback mechanism? Thank you.

acbart commented 5 years ago

It all depends on various factors. If your instructor is using BlockPy, and wants to modify a problem, they will need to make new assignments. If they're using our assignments, then they will need to get in touch with us. I would suggest that your professor get in touch with me directly to explain their situation. I cannot offer much support, however - this is an exceptionally busy time for me. Usually people who want to use our curriculum plan such things out with us. Using BlockPy on its own, sans our questions, doesn't require such planning of course.

On Mon, Feb 11, 2019, 4:34 PM Xiaowen(Shawn) Ying <notifications@github.com wrote:

Looks like it's using the old style API. That example site needs to be updated to use the new API from Pedal https://github.com/acbart/pedal. Unfortunately, the semester just started here at UD and I don't have time to fix them up. I'll keep this issue open and hopefully we can get around to it sometime soon.

Thanks for your quick reply. I was also thinking about the same thing.

Actually, I'm posting this issue because I'm TA-ing a course in semester and the professor integrated the BlockPy into her Canvas course. So this problem seems to affect all the assignments in our course that use BlockPy. Basically, the students will see the correct outputs on the console but will not get any credit for it. Do you have any suggestion about this? For example, is there any way to temporarily disable the feedback mechanism? Thank you.

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/RealTimeWeb/blockpy/issues/40#issuecomment-462503384, or mute the thread https://github.com/notifications/unsubscribe-auth/AA2wy0jcnuKn5JFy55FyJ5tiJKCWKQ6mks5vMeHJgaJpZM4a1FZe .

xwying commented 5 years ago

Thanks for your reply. The assignment was created by our previous instructor and we didn't directly use the sample problems. I've been looking for how to modify the feedback code to fix the error, and your reply inspired me that maybe I don't have enough permission to see the options.

After contacting the professor to grant me permission I can finally see and modify the on_run code. So I changed the old APIs to the new pedal APIs and it worked. Thanks!