Rules for rounding are underspecified w.r.t. handling ties and most probably they are inconsistent between languages.

Crystal tests were tricky to update because default rounding mode changed from half-up to half-even between Crystal 0.24 and 1.0, and hardcoded values for fixed tests were not correct anymore.

Instructions don't specify how break time should be distributed between breaks

python/js reference solutions don't produce same answer for day_plan_solution(1, 3, 1) due to differences in language-built-in round

python: missing fixed tests that test the rounding behaviour

example of incorrect (sometimes) passing solution: https://www.codewars.com/kata/reviews/581e4b3fed2e5b6e3d000021/groups/6573479cce1f7d00010b1f59

Terrible kata. Your examples and tests are incorrect. Take this example:

dayPlan(8, 5, 30) == [ 30, 83, 30, 83, 30, 83, 30, 83, 30 ]

8 hours is 480 minutes. If you add up all durations in your "correct" answer, you get 482 minutes.

Approved

python new test framework is required. updated in this fork

• Number.isNaN() is more reliable than isNaN() as isNaN('') is coerced to 0, while a boolean is coerced to 0 or 1; both values are intuitively "not numbers", but they don't evaluate to NaN, so isNaN() returns false.

• Unary plus (+) converts things into numbers

#11b2r/3px:0z:vV
still feeling the burn
nice kata

Ruby 3.0 should be enabled.

Kata passes in 0 tasks occasionally in random tests, behavior for which is undefined. (JavaScript)

(Also, yes, I know that the tests cannot be changed. It's important to have an open issue in case that circumstance changes in the future, and to make others more aware of the issue that encounter it.)