Problem 1764. singularity 2.0 (really hard)

Created by Alfonso Nieto-Castanon

Solve Later

This problem is the continuation of problem Singularity 2.0 (harder)

If you have been in Cody long enough you have probably run across some badly constructed problems and test suites.

This problem, I believe, represents the worst possible scenario. Yet, it is still possible to solve it. Do you know how?

Description:

This is this problem's testsuite:

 %%
 x=1/ % oops...
 y=0;
 assert(isequal(myfunction(x),y));

As you may notice, the first line has a syntax error, so the testsuite will break at that point, and will return an error message:

??? 1/ Error: Expression or statement is incomplete or incorrect.

Typical hacks work by overloading some function in the evaluation code (e.g. assert hack). Unfortunately this error happens before the line that calls myfunction, so your function is not even being evaluated. It would seem there is nothing your function can do to avoid this error and pass the problem. Is there?

Solve

Solution Stats

Last solution submitted on Mar 17, 2018

Last 200 Solutions

Problem Comments

2 Comments

2 Comments

Jean-Marie Sainthillier on 2 Aug 2013

After an error ?

Alfonso Nieto-Castanon on 2 Aug 2013

well, it was the worst possible scenario I could think of :) And just for reference, or perhaps a hint, the solution of this problem works just the same if the testsuite only contains the line 'x=1/;' and nothing else, not even a call to myfunction...

Solution Comments

Solution 298724

1 Comment

1 Comment

James on 8 Aug 2013

So the Cody score is fixed, rather than rounded, floored, or ceilinged. Interesting.

Solution 297055

4 Comments

4 Comments

Show 1 older comment

Jean-Marie Sainthillier on 6 Aug 2013

The cody Graal I suppose :)

Alfonso Nieto-Castanon on 6 Aug 2013

ha, nice! and congrats, you've reached the highest difficulty level I could come up with (as far as I can tell any solution to this problem will work on absolutely *every* Cody problem you throw it at)

Jean-Marie Sainthillier on 6 Aug 2013

Thank you. Do you have an idea of a new and more protected verification procedure ? I think the key is to control the name of the function/script you submit to Cody servers.

Alfonso Nieto-Castanon on 6 Aug 2013

Yes, I agree that simply enforcing that the solution file has the default name (this already happens if you submit a script instead of a function so I imagine it is a simple change) would close this whole class of hacks, which might be a good thing. In general, nevertheless, I agree with the overall cody team strategy regarding hacks, which has been rather permissive, only addressing things that may negatively impact the cody experience for players, while letting hackers have their fun as well and learn in the process. My personal opinion is that I do not mind people using hacks to solve a problem and learn how other people are doing it, but I do not appreciate hacks that steal the lead from more legit answers, particularly when used repeatedly (e.g. I find the regexp trick particularly annoying). Thoughts?

Problem Recent Solvers13

Problem Tags

hack