Please check out my excercise and help me understand why I get an error on the 1e test (see more detail below)

[marcothijs]

My excercise https://dodona.be/nl/activities/9016521/ gets the first test wrong while the output "looks" the same as the test outcome. Can anyone explain why?
I have my solution code in the solution map, the test in the evaluation map.

Please see the screenprint below:

Thanks!

[IngeVerbeek]

Ik heb geen toegangsrechten tot de oefening dus ik kan niet kijken naar de opgave of je bestanden, maar zou het stomweg een "newline" op het einde van de regel kunnen zijn? Het lijkt alsof de output 2 regels verwacht.

[marcothijs]

Ik heb de oefening nu gepubliceerd (met nog steeds de fout in de 1e test) en zou nu wel zichtbaar moeten zijn in Dodona. De lege lijn is volgens mij niet het probleem, de andere tests werken wel. Het blijft mij een raadsel...

[TomNaessens]

Hi @marcothijs,

There's hidden characters present in both your solution, as in the test plan. When I download the testplan yaml and run it through hexdump, there's one/a few hidden characters present before the "Login succesvol!" of the second test case:

00000b0                       N   o   g       4       p   o   g   i   n
00000c0   g   e   n       o   v   e   r   .  \n
00000d0   �   � 217   L   o   g   i   n       s   u   c   c   e   s   v
00000e0   o   l   !  \n           -       s   t   d   i   n   :       |
00000f0  \n                           g   t   i   l   1   2   3   4  \n

That's a bit weird, because that second test was passing. It was the first one that's failing. The reason for that is that in your submitted solution, the hidden characters are also present:

0000110   j   u   i   s   t   _   w   a   c   h   t   w   o   o   r   d
0000120   :  \n                                                   p   r
0000130   i   n   t   (   "   �   � 217   L   o   g   i   n       s   u
0000140   c   c   e   s   v   o   l   !   "   )  \n
0000150                               c   o   r   r   e   c   t   e   _

If I adjust the code to not contain the hidden characters, the first test passes, but the second one starts to fail. When the hidden characters are removed from the test plan, that one should start to pass too.

The reason that this isn't immediately visible on the Dodona interface is that we process and sanitise the results to be able to create a meaningful diff. In this case however, it would be nice to have been able to spot this more easily. We'll give that a think and see what we can come up with.

Hope that helps!

Kind regards,

Tom

[marcothijs]

Hi Tom,

Thanks for explaining the reason but I still have no idea how to get rid off the hidden characters. I've changed the "" notition on the stdout with | and no "" on the following line + copied the same text from below... (just hoping to see a difference) and yes, now I don't have the test failing anymore. But I'm still not convinced this is the way to fix this problem... I don't like to be dependant on things I can't see ;-).

Thanks again.
Marco

[TomNaessens]

Hi @marcothijs,

The safest way to get rid of the hidden characters, is to remove the whole line, and retype it manually. Copy-pasting will also include those invisible characters, so that doesn't work to fix those.
You can also replace only part of the line where they're present, but that's harder since you can't exactly see where they start and end, so just doing the whole line is safer.

You mentioned that you now copied the line from the second testcase to the first one, but since the second testcase contains those visible characters, by copy-pasting the line, they are also present in the first testcase now. The tests succeed, since the same invisible characters are still present in the code solution too.

So while everything matches up now, that's only because the invisible characters are present everywhere, and students that will "manually" write the code (without copy pasting), will run into the original issue here, so to fix it fully, you'll have to remove all following lines and retype them (manually, to avoid copy-pasting the hidden characters) on lines 5, 15 and 22, and also on line 14 of the first example solution, and line 14 of the second example solution.

Hope that helps!

Kind regards,

Tom

[marcothijs]

Hi Tom,

Thanks for pointing out the issue once more, and indeed, it was still a mess. I've now manually typed the code in the test script as well in the 2 .py solutions. It should be fine now.
As I copy/paste regularly when creating exercises and I never had this issue before, I wonder where it did go wrong this time. It is good to know that when I see this test behavior in the future, that is getting cause by invisible characters and that I need to stretch my fingers :-).
Thanks again.
Marco