Hi Leslie,
I did say it was a heavy handed method to establish a leap year, so I think we
are in agreement. :-)
The method that I ended up using was Raul's isLeap which is the J version of
your algorithm with a clever use of 2 | to ensure that divisible by 4, not 100,
400 is the result
isleap=. 2 | +/ 0= 4 100 400 | yt NB. yt is the year of the current date.
By the way if you want to see todayno in J, all you need to do is type in its
name without an argument and it will return the definition
todayno
3 : 0
0 todayno y
:
a=. y
if. 1=x do. a=. 0 100 100 #: a end.
a=. ((*/r=. }: $a) , {:$a) $,a
if. 1<x do.
'y m d h mm s'=. <"_1 |: a
else.
'y m d'=. <"_1 |: a
end.
y=. 0 100 #: y - m <: 2
n=. +/ |: <. 36524.25 365.25 *"1 y
n=. n + <. 0.41 + 0 30.6 #. (12 | m-3),"0 d
if. 1<x do.
n=. n + (24 60 60#.h,.mm,.s)%24*3600
end.
0 >. r $ n - 657378
)
Also, I believe Linda was correct stating that it has day 0 as being January
1st, 1800. This makes it more useful as a way of determining dates relative to
each other rather than a Gregorian-Julian conversion verb.
todayno 1800 1 1
0
Cheers, bob
On Sep 27, 2014, at 9:25 PM, 'Leslie S Satenstein' via Beta
<[email protected]> wrote:
> leapyear = ((year mod 400) =0 ) or ((year mod 4 = 0) and not ((year mod 100)=
> 0).
>
> It seems to me to be excessive use of todayno. Not having seen the code to
> todayno, I would
>
> presume it uses a gregorian to Julian day number conversion, The julian
> day is defined as
>
> (jan 1 year 1) = zero
>
> and todayno calculates the days since.
>
> I would presume that the correction includes the 10 day jump in year 1580
> where the Pope corrected the gregorian calendar by declaring the new leapyear
> rule and .... October 21 1580 was deemed to be Nov 1st of that year.
>
> Furthermore, if you want to know what day of the week the the 31th of the
> month is, you would realize that it would be 3 days after February 28, for
> non leap years and three days after Feb 29th on leapyears. A similar rule
> applies for months with 30 days.
> This was an established convention.
>
>
>
>
> Regards
>
>
> Leslie
>
>
>
>
>
>> ________________________________
>> From: robert therriault <[email protected]>
>> To: Beta forum <[email protected]>
>> Sent: Wednesday, September 24, 2014 10:28 AM
>> Subject: Re: [Jbeta] Concern about todayno
>>
>>
>> Hi Henry,
>>
>> ...and because it doesn't exist Feb 29, 2003 has the same todayno value as
>> March 1, 2003. That seems pretty reasonable to me since checking if 2003 2
>> 29 =&todayno 2003 3 1 becomes a way (albeit heavy handed) to discern leap
>> years.
>>
>> Cheers, bob
>>
>> On Sep 24, 2014, at 2:14 AM, Henry Rich <[email protected]> wrote:
>>
>>> 2003 2 29 doesn't exist
>>>
>>> Henry Rich
>>>
>>> On 9/24/2014 5:13 AM, Linda Alvord wrote:
>>>> Shouldn't the difference be 366 every four years. Bob's comments caused me
>>>> to worry about leap year, and I am getting incorrect answers but they may
>>>> be
>>>> based on this.
>>>>
>>>>
>>>>
>>>> diff=: 13 :'(todayno x)-todayno y'
>>>>
>>>>
>>>>
>>>> (2005 3 1) diff 2004 2 29
>>>>
>>>> 366
>>>>
>>>> (2004 3 1) diff 2003 2 29
>>>>
>>>> 366
>>>>
>>>> (2003 3 1) diff 2002 2 29
>>>>
>>>> 365
>>>>
>>>> (2002 3 1) diff 2001 2 29
>>>>
>>>> 365
>>>>
>>>> (2001 3 1) diff 2000 2 29
>>>>
>>>> 366
>>>>
>>>> (2000 3 1) diff 1999 2 29
>>>>
>>>> 366
>>>>
>>>> (1999 3 1) diff 1998 2 29
>>>>
>>>> 365
>>>>
>>>> (1998 3 1) diff 1997 2 29
>>>>
>>>> 365
>>>>
>>>> (1997 3 1) diff 1996 2 29
>>>>
>>>> 366
>>>>
>>>> (1996 3 1) diff 1995 2 29
>>>>
>>>> 366
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>> ----------------------------------------------------------------------
>>>> For information about J forums see http://www.jsoftware.com/forums.htm
>>
>>
>>
>>
>>>>
>>> ----------------------------------------------------------------------
>>> For information about J forums see http://www.jsoftware.com/forums.htm
>>
>> ----------------------------------------------------------------------
>> For information about J forums see http://www.jsoftware.com/forums.htm
>>
>>
>>
> ----------------------------------------------------------------------
> For information about J forums see http://www.jsoftware.com/forums.htm
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm
