Very smart!

understood the mistake
of the julian calendar year % 100 != 0 - superfluous
really don't understand why...

Hello my code passes tests with years > 1752
and where < 1752 are lost for several days (1, 3 ... somewhere around 10)
in tests, the rule of the julian calendar?
(year % 4 == 0) and (year % 100 != 0) ?
Or am I misunderstanding something, please tell me

i would like at some time in The Future whith god help to be able to come with this kind of simple solutions

there was no year 0 in either the Gregorian and the Julian calendar. the years go from -1 BC (before Christ) to +1 AD (anno Domini). I would suggest removing the mention of year 0 in the description, and updating the random tests accordingly

fixed

C random tests can generate dates from September 3 - September 13 1752

that should be an issue, i checked the RNG in C and it can indeed generate such dates. raising

no, this is the point, there was no divisible-by-100 exception in the Julian calendar

Ah maybe I get it, you check if l - w != 0, and then add those numbers and multiply them by bitshifting them once. otherwise you're returning the square. Cool solution!

Can you perhaps explain your answer?

Description says : "leap years were on every year divisible by 4, without the exception of years divisible by 100."
is either bad English, or should be: "leap years were on every year divisible by 4, with the exception of years divisible by 100."

HOWEVER, the true Julian calendar had leap years every 4 years, except for some historic period where it was irregular.
Furthermore, it's stated that year 0 (which technically didn't exist at all on the Julian calendar!) is a leap year, although it's both divisible by 4 and 100.

Beautifully Simple!

Fixed ;-)

C version of this kata needs to be updated per https://github.com/codewars/content-issues/wiki/List-of-C-Kata-to-Update