Re: [9front] quick tmdate question

[lyndsay <lyndsay@vivaldi.net>]

awesome, that should be fine for me; all of the times being processed=20
have to be put in by the user and the users probably won't want to use a=20
calendar other than gregorian so almost all input, even before 1582,=20
would probably end up being in proleptic gregorian format anyway. at=20
least, that's what the mathematical equations i put in are assuming!=20
i'll also put a note in the help and manual for the software juuuust to=20
make sure.

thanks!

lyndsay

On 9/30/21 23:15, ori@eigenstate.org wrote:
> Quoth lyndsay <lyndsay@vivaldi.net>:
>> can't find a clear answer in tmdate(2)=C2=A0 - i'm writing a program (=
in c)
>> that needs a way to store and parse any arbitrary time AD. will those =
be
>> properly dealt with if i store times in Tm and parse user input with
>> tmparse() or should i write my own function?
> If the proleptic gregorian calendar is enough, then
> tmdate should work for you.  It does not handle the
> (locale-specific!) calendar changeovers.  Our timezone
> information simply doesn't encode calendar changes.
>
> I'm not even sure if the IANA timezone database encodes
> calendar changes.
>
> I'd suggest using the proleptic gregorian calendar, and
> converting from it as late as possible.  Ideally never.
>
> https://en.wikipedia.org/wiki/List_of_adoption_dates_of_the_Gregorian_c=
alendar_by_country
>