On Mon, 13 Jan 2003, Rich Bowen wrote:

> So the internals would become ... what exactly? Right now, there are two
> integers stored - the "julian day", which is not actually the julian
> day, and the seconds, which is seconds since midnight (although I tend

Sorry, I wasn't clear.  I meant that since we are _already_ storing rata
die days plus seconds past midnight, we should change the _names_ of the
various methods and hash elements in the code, so that they don't
misleadingly say julian.  That's all.

> to get confused all the time about what timezone this is actually stored
> in - your local one, or GMT). We also store the offset (ie,
> hours offset from GMT) which is a 5-character string that looks like
> "-0100" or "+0430". Timezones are hard.

We'll need to store an object, but we can't do that until we have
DateTime::TimeZone done ;)

> Changing the "julian day" bit to rata die is goodness, and this is a
> good time to do it. Moreover, if the base object has rata die stored,
> rather than julian, I could finally finish all the DateTime::Calendar
> classes that I have half finished.

Yep.  I'm thinking we'll offer a method:

 my ($days, $secs, $nanosecs) = $datetime->rata_die

and declare that the standard method for inter-operability, so _all_
calendars must implement it.


-dave

/*=======================
House Absolute Consulting
www.houseabsolute.com
=======================*/

Reply via email to