Since #406 brings up the issue of mid-autumn festival, I figured that I should open a tracking issue for handling non-trivial variable_dates generally stemming from other calendar systems. A list I can think of are:
[ ] Chinese and related lunisolar systems. Commonly tabulated due to pain of computation. But the CN/HK/TW tabulation will be off due to Viet Nam and Korea using their respective timezones.
[ ] Hindu and Buddist lunisolar calendars. No idea, but first-look seems simpler than Chinese due to more fixed leap rules. Except for the Burmese correction, which depends on observation.
Each of these calendar systems will likely require an external dependency with loads of tabulation, so expect discussions about optional dependencies and porting. (Even with a system as "simple and native" as Gregorian, it took some work to figure out how to find Easter!)
Dependency on observation may require updates every couple of years, and also increase the amount of data required to ship. Unless you try to predict the future. (Well, the dependency should handle it...)
Since #406 brings up the issue of mid-autumn festival, I figured that I should open a tracking issue for handling non-trivial
variable_date
s generally stemming from other calendar systems. A list I can think of are:Each of these calendar systems will likely require an external dependency with loads of tabulation, so expect discussions about optional dependencies and porting. (Even with a system as "simple and native" as Gregorian, it took some work to figure out how to find Easter!)
Dependency on observation may require updates every couple of years, and also increase the amount of data required to ship. Unless you try to predict the future. (Well, the dependency should handle it...)