Re: timestamp

[tomas]

On Sat, Feb 12, 2022 at 12:49:10PM +0100, Ricardo Wurmus wrote:
> 
> Hi adriano,
> 
> I’ve got no good answers as to “why” things are the way they are, but
> the manual explains the range of these values:
> 
> > It seesm to be
> >
> > (tm:mon %3)
> >
> > This returns
> >
> > 11
> >
> > I expected 12 but ok, I recognize this kind of weirdness
> 
>  -- Scheme Procedure: tm:mon tm
>  -- Scheme Procedure: set-tm:mon tm val
>      Month (0-11).
> 
> > I'm unhappy with 
> >
> > (tm:year $3)
> >
> > This returns
> >
> > 120
> >
> > it's 2020
> >
> > Why would 120 represent 2020 ?
> 
>  -- Scheme Procedure: tm:year tm
>  -- Scheme Procedure: set-tm:year tm val
>      Year (70-), the year minus 1900.
> 
> I don’t know why this would be useful, but that’s what it is ¯\_(ツ)_/¯

Oh. That is old POSIXy tradition :-)

Guile actually only passes along what libc provides, and I think it is a
good idea, in general. See the man page of localtime(3) on your next Linux-ish
box for all the details. If you prefer an online version, perhaps [1], [2]
convey an idea.

The year... those interfaces were developed well before the year 2000. It was
usual, Back Then (TM) to write down the year as a two-digit number (the 19
prefix was implied). Nobody could predict that the year 2000 was about to
arrive, honestly ;-)

As to the month... zero based numberings ease computations and setting up
name arrays (as in: 0: January, 1: February and so on).

Don't ask me why, though, the day of the month (1..31) is one-based. It is
the only exception :-)

But it makes for an easy calculation where you can add up the year (times
year length, taking leap years into account), the days-in-month so far for
full months and the current month day and things to arrive at the current
Epoch day. Guess one has to be one-based, otherwise you'll have to add 1 at
the end.

(This is all somewhat tongue-in-cheek ;-P

But it illustrates some other point: for folks coming from C and POSIX, (I
do), all that stuff feels "pretty natural", although it isn't, of course.

Cheers

[1] https://en.cppreference.com/w/c/chrono/localtime
[2] https://en.cppreference.com/w/c/chrono/tm

-- 
t

signature.asc
Description: PGP signature