dayofyear is not great when going into a new year

[Eli the Bearded]

In comp.lang.python, Chris Angelico  <rosuav at gmail.com> wrote:
> There are multiple definitions for "day of year", depending on how you
> want to handle certain oddities. The simplest is to identify Jan 1st
> as 1, Jan 2nd as 2, etc, to Dec 31st as either 365 or 366; but some
> libraries will define the year as starting with the week that contains
> the Thursday, or something, and then will define days of year
> accordingly.

That sounds like some weird off-shoot of the ISO-8601 calendar. That
document primarily concerns itself with weeks. Week 1 of a year is the
first week with a Thursday in it. The last week of a year will be either
52 or 53, and you can have things like days in January belonging to the
week of the previous year. Wikipedia gives examples:

https://en.wikipedia.org/wiki/ISO_week_date

If you are operating on that, then it might indeed make sense to number
the days from YYYY-W1-1. I can't say I've ever encountered that. Since
W1-1 is always a Monday on the ISO calendar, it would have the neat
property that you could always turn day of year into day of week with a
mod operation. It would have the quirk that years are either 364 or 371
days, neither of which most people would answer when asked "How many
days are there in a year?"

I've only used ISO dates for by-week graphs, because they have the nice
property of "all weeks are seven days", so you don't get oddball weeks
screwing up your plots.

> If you want an easy way to graph day-by-day data and the exact day
> numbers are irrelevant, what I'd recommend is: Convert the date into
> Unix time, divide by 86400, floor it. That'll give you a Julian-style
> date number where Jan 1st 1970 is 0, Jan 2nd is 1, etc, and at the end
> of a year, it'll just keep on incrementing. That would get you past
> the 2020/2021 boundary pretty smoothly.

That works well. The previous suggestion using January 1st 2020 as an
epoch start is also good.

Elijah
------
also finds "week starts on Monday" to be oddball about ISO-8601