[iClaudiusX]

Technically, the Zune bug didn't kill the device. The bug sent it into an infinite boot loop until the following day when the leap year calculation could return.

But yes, the difficulty in verifying a correct implementation of date and time calculation calls out for a standard test apparatus to "prove" correct results for a defined scope.

[avip]

I once stupidly implemented a "date math" feature, and after introducing numerous bugs in production opted to use https://svn.python.org/projects/sandbox/trunk/datetime/test_... as my ref. test (which exposed numerous other bugs). Likely not full coverage of all date-time related bizarrities, yet a useful resource.

[nkozyra]

> I once stupidly implemented a "date math" feature

I believe this to be one of those rites of passage for any programmer.

[karlp]

For date and time, Arrow is a lifesaver in python

https://arrow.readthedocs.io/en/latest/