Chalten might help, i remember it had representations for months of years (January of 1990), it might have something similar for weeks.
On 13 May 2016 at 05:40, Cédrick Béler <[email protected]> wrote: > Yes I agree. We may have a preference to set the week day start (fistDayOfWeek > ?). > > > After that we could agree on the fact the first week is the first complete > week (otherwise it can be a preference). > > Then we need to find the first day index of the first fistDayOfWeek. > (Date year: self year day: 1) weekdayIndex > -> gives the index according to the preference (for now Sunday -> 1) > > Then it’s quite straightforward to get the week number. > > Just have to check if the current date is in the last week of the previous > year (returns 52 or 53). > I did something but not sure yet (but no preference). > > There are been some works on time representation (Chronos if I remember > well ? but cannot load in Pharo 5, depends on OSProcess). > > An interesting read [1]. > > Cheers, > > Cédrik > > ps: > http://stephane.ducasse.free.fr/Teaching/CoursAnnecy/0506-M1-COO/A%20New%20Object-Oriented%20Model%20of%20the%20Gregorian%20Calendar.pdf > > > > Interesting to know. But I see this already seems hard coded... > > Date nameOfDay: 1 "--> #Sunday" > > Timespan subclass: #Date > poolDictionaries: 'ChronologyConstants' > > ChronologyConstants class >> initialize > DayNames := #(Sunday Monday Tuesday Wednesday Thursday Friday Saturday). > > Also seems to have changed at some point, with the historical > nameOfDay recorded in the Terse Guide [1] circa 2010 shown to be > different. Perhaps this is something we should make a regional > preference. Was there any ongoing works on other regional > preferences? > >
