On Fri, 24 Jan 2003, Peter J. Acklam wrote: > [EMAIL PROTECTED] (Dave Rolsky) wrote: > > > Ok, having accepted the fact that day of week numbering will be > > 1-7, now we get to argue about which day is #1. > > > > Time::Piece (interface by Larry Wall) says Sunday, as does > > localtime, which comes from C's localtime function > > > > Date::Calc says Monday, and refers us to ISO 8601, which does > > indeed say day #1 is Monday. > > > > I'm inclined to go with ISO rather than backwards compatibility > > with C. > > I believe this thread proves that what is considered the first day > of the week must be customizable. If it isn't, then someone > (perhaps not someone on this mailing list, but someone) will be > dissatisfied and consider the "DateTime" modules useless and write > their own modules -- and much of the point with the "DateTime" > modules will be lost. > > In relation to week numbers and day of week numbers, there are > at least four parameters that must be customizable: > > 1) What is the first day of a week (Sunday, Monday, ...)? > 2) What is the number of this day (0, 1)? > 3) How many days of a week must be in the new year before > the week is considered the first week in the new year? > (ISO: 4, US: 1 and 7) > 4) What is the number of this week (0, 1)? > > As for the default behaviour, I think ISO is the best thing, since > it aims at being a world-wide standard.
Sure, but customizable at what level? Having an API that behaves differently in different installations is a BAD thing. The foo_0 methods, I think, already address this matter. -- And everyone said, "If we only live, We too will go to sea in a Sieve - To the hills of the Chankly Bore!" (The Jumblies, by Edward Lear)
