Kattis style check

[ghamerly]

This is a quick review based on an automated style checking script, and some comments from me. Please fix those things that are reasonable to fix. (Some problems, for example, don't really need TLE submissions.)

• general

  • [x] general: multiple values for source: {'Will Code for Drinks F2019': 1, 'Will Code for Drinks, Fall 2019': 1, 'Kattis': 4}

• askmarilyn

  • [x] askmarilyn: has no TLE submissions
  • [x] askmarilyn: since this problem uses custom output validation, test the validator by adding some wrong_answer submissions that: print nothing, that spam the validator with lots of correctly-formatted output (e.g. more rounds than 1000), and that spam the validator with lots of incorrectly-formatted output (e.g. strings when numbers are expected, long lines, etc.).

• barshelf

  • [x] barshelf: has no WA submissions
  • [x] barshelf: has no TLE submissions
  • [x] barshelf/problem_statement/problem.en.tex: bad includegraphics width; use a multiplier (e.g. 0.9) or HTML layout can break
  • [x] barshelf/problem_statement/problem.en.tex: use \cdot instead of \times

• cheaperdrink

  • [x] cheaperdrink/problem_statement/problem.en.tex: bad includegraphics width; use a multiplier (e.g. 0.9) or HTML layout can break

• drinkingsong

  • [x] drinkingsong: has no TLE submissions
  • [x] drinkingsong: the output says "counting down to zero", but I think
    this should say "counting down to one".

• elderscrollbar

  • [x] elderscrollbar/problem.yaml: this feels like a problem where you should use "validator_flags: space_change_sensitive"

• tgif

  • [x] tgif: has no TLE submissions
    ~

[thorehusfeldt]

Regarding TLE for scoring problems, I am actually confused. Maybe @simonlindholm can chip in here as well: What is the difference between AC(x) for x less than the maximum score and TLE? For barshelf, a submission that starts by (say) constructing all permutations of the input would receive AC(0), and should reside in – where? PAC or TLE?

[thorehusfeldt]

Regarding includegraphics, “use multiplier”… multiplier of what? Of a LaTeX length macro? I can write, for instance,

\includegraphics[width= 0.4\textwidth]{img/messy_bottles.png}

It is not clear to me that the HTML layout will make more sense of that. I couldn’t find good best-practice examples to imitate.

[simonlindholm]

If it's AC(0) it should go in partially_accepted/; however, I updated testdata_tools/ to a more recent version which never gives AC(0) but instead gives as overall verdict the first error, so that shouldn't happen. Edit: but it is indeed true that for scoring problems, partially_accepted/ to large extent replaces time_limit_exceeded/; it's rare that solutions too slow for even the first test group are worth coming up with.

[thorehusfeldt]

Well, for barshelf I was able to concoct a not-completely-insane exponential-time solution, and added some size-20 problems to the smallest test group to give it something to work on. So now there is a TLE submission.

[ghamerly]

Yes, multiplier on \textwidth, exactly like you suggest. Sorry for not being clear.

On Mon, Oct 14, 2019, 03:33 Thore Husfeldt notifications@github.com wrote:

Regarding includegraphics, “use multiplier”… multiplier of what? Of a LaTeX length macro? I can write, for instance,

\includegraphics[width= 0.4\textwidth]{img/messy_bottles.png}

It is not clear to me that the HTML layout will make more sense of that.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/thorehusfeldt/will-code-for-drinks-F2019/issues/12?email_source=notifications&email_token=AA2CQHPXXVBGWS4KXGMTJATQOQVGNA5CNFSM4JAIQDM2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEBDXI7I#issuecomment-541553789, or unsubscribe https://github.com/notifications/unsubscribe-auth/AA2CQHOO2HQBV3BX53R74ETQOQVGNANCNFSM4JAIQDMQ .

[thorehusfeldt]

Oh, because it’s sent through plasTeX, which indeed does recognise this macro. This makes sense.

[thorehusfeldt]

Fixed in 369996866d2184863c1f689d2b43e3ef4968f590 to d2d611991b8ccdbf0be5afd5b11b8be97490707a