A long time ago, in a century far away...

#1

I wonder if the Pick (or at least UniData) ICONV function correctly
calculates leap years in antiquity. Running the following program
on UniData 6.0 gives an interesting result.

FOR Y = 1 TO 12000
LY1 = (ICONV('3-1-':Y,'D') - ICONV('2-28-':Y,'D')) - 1
LY2 = (REM(Y,4)=0) AND ((REM(Y,100)#0) OR (REM(Y,400)=0))
IF LY1 # LY2 THEN CRT Y:' ':LY1:' ':LY2
NEXT Y

The result is interesting in that it is not null; it indicates that UniData
thinks the years 200 and 600 are leap years, but thinks that neither
400 nor 800 are leap years.

200 1 0
400 0 1
600 1 0
800 0 1

Comments?

--
frosty

#2

Goo'day,

Found your treatise amongst all the tripe.... Wish he'd stay off
whatever he's on..... or at least lose the cdp address...... The early
Greeks and Romans (and a few others since then) had the right
idea......

Go, Genghis!

On Wed, 5 Oct 2005 15:07:23 -0700, "frosty" <frostyj (AT) bogus (DOT) tld> wrote:

Quote:

I wonder if the Pick (or at least UniData) ICONV function correctly
calculates leap years in antiquity. Running the following program
on UniData 6.0 gives an interesting result.

FOR Y = 1 TO 12000
LY1 = (ICONV('3-1-':Y,'D') - ICONV('2-28-':Y,'D')) - 1
LY2 = (REM(Y,4)=0) AND ((REM(Y,100)#0) OR (REM(Y,400)=0))

Should LY2 really have the bit about divisible by "100"???? Isn't it
only 4 and 400???

Somewhere in the fog.....

Quote:

IF LY1 # LY2 THEN CRT Y:' ':LY1:' ':LY2
NEXT Y

The result is interesting in that it is not null; it indicates that UniData
thinks the years 200 and 600 are leap years, but thinks that neither
400 nor 800 are leap years.

200 1 0
400 0 1
600 1 0
800 0 1

Comments?
Regards,

Bruce Nichol
Talon Computer Services
ALBURY NSW Australia

http://www.taloncs.com.au

If it ain't broke, fix it until it is....

#3

Goo'day,

Found your treatise amongst all the tripe.... Wish he'd stay off
whatever he's on..... or at least lose the cdp address...... The early
Greeks and Romans (and a few others since then) had the right
idea......

(yawn)

#4

uv 10.x:
100 1 0
200 1 0
300 1 0
500 1 0
600 1 0
700 1 0
900 1 0
1000 1 0
1100 1 0
1300 1 0
1400 1 0
1500 1 0
Chandru
"frosty" <frostyj (AT) bogus (DOT) tld> wrote

Quote:

I wonder if the Pick (or at least UniData) ICONV function correctly
calculates leap years in antiquity. Running the following program
on UniData 6.0 gives an interesting result.

FOR Y = 1 TO 12000
LY1 = (ICONV('3-1-':Y,'D') - ICONV('2-28-':Y,'D')) - 1
LY2 = (REM(Y,4)=0) AND ((REM(Y,100)#0) OR (REM(Y,400)=0))
IF LY1 # LY2 THEN CRT Y:' ':LY1:' ':LY2
NEXT Y

The result is interesting in that it is not null; it indicates that
UniData
thinks the years 200 and 600 are leap years, but thinks that neither
400 nor 800 are leap years.

200 1 0
400 0 1
600 1 0
800 0 1

Comments?

--
frosty

#5

frosty wrote:

Quote:

I wonder if the Pick (or at least UniData) ICONV function correctly
calculates leap years in antiquity. Running the following program
on UniData 6.0 gives an interesting result.

FOR Y = 1 TO 12000
LY1 = (ICONV('3-1-':Y,'D') - ICONV('2-28-':Y,'D')) - 1
LY2 = (REM(Y,4)=0) AND ((REM(Y,100)#0) OR (REM(Y,400)=0))
IF LY1 # LY2 THEN CRT Y:' ':LY1:' ':LY2
NEXT Y

The result is interesting in that it is not null; it indicates that UniData
thinks the years 200 and 600 are leap years, but thinks that neither
400 nor 800 are leap years.

200 1 0
400 0 1
600 1 0
800 0 1

Comments?

--
frosty

The rules for determining leap years are actually -

Once every 4 years
Except every 100 years
Except every 400 years

So UNIDATA have it right..

We used to use it as a 'how many lines of code do you need to determine
a leap year' in Databasic ( 1 as it happens )

Regards

Godric

#6

<godric (AT) trapezesoftware (DOT) com> wrote

Quote:

frosty wrote:
I wonder if the Pick (or at least UniData) ICONV function correctly
calculates leap years in antiquity. Running the following program
on UniData 6.0 gives an interesting result.

FOR Y = 1 TO 12000
LY1 = (ICONV('3-1-':Y,'D') - ICONV('2-28-':Y,'D')) - 1
LY2 = (REM(Y,4)=0) AND ((REM(Y,100)#0) OR (REM(Y,400)=0))
IF LY1 # LY2 THEN CRT Y:' ':LY1:' ':LY2
NEXT Y

The result is interesting in that it is not null; it indicates that
UniData
thinks the years 200 and 600 are leap years, but thinks that neither
400 nor 800 are leap years.

200 1 0
400 0 1
600 1 0
800 0 1

Comments?

--
frosty

The rules for determining leap years are actually -

Once every 4 years
Except every 100 years
Except every 400 years

So UNIDATA have it right..

We used to use it as a 'how many lines of code do you need to determine
a leap year' in Databasic ( 1 as it happens )

Regards

Godric

#7

<godric (AT) trapezesoftware (DOT) com> wrote

Quote:

The rules for determining leap years are actually -

Once every 4 years
Except every 100 years
Except every 400 years

So UNIDATA have it right..

We used to use it as a 'how many lines of code do you need to
determine
a leap year' in Databasic ( 1 as it happens )

Unidata is right. But since Feb 29,2000 existed. I did a little
research.

http://h71000.www7.hp.com/openvms/pr...2000/leap.html

Rule is once every 4 years except in years that are not divisible by
400.

rog

#8

godric (AT) trapezesoftware (DOT) com wrote:

Quote:

frosty wrote:
I wonder if the Pick (or at least UniData) ICONV function correctly
calculates leap years in antiquity. Running the following program
on UniData 6.0 gives an interesting result.

FOR Y = 1 TO 12000
LY1 = (ICONV('3-1-':Y,'D') - ICONV('2-28-':Y,'D')) - 1
LY2 = (REM(Y,4)=0) AND ((REM(Y,100)#0) OR (REM(Y,400)=0))
IF LY1 # LY2 THEN CRT Y:' ':LY1:' ':LY2
NEXT Y

The result is interesting in that it is not null; it indicates that
UniData thinks the years 200 and 600 are leap years, but thinks that
neither 400 nor 800 are leap years.

200 1 0
400 0 1
600 1 0
800 0 1

Comments?

--
frosty

The rules for determining leap years are actually -

Once every 4 years
Except every 100 years
Except every 400 years

So 400 and 800 should be leap years; UniData thinks they are not.
Also, 200 and 600, which are divisible by 100 but not by 400,
should not be leap years; UniData thinks they are.

Quote:

So UNIDATA have it right..

I would disagree.

Quote:

We used to use it as a 'how many lines of code do you need to
determine a leap year' in Databasic ( 1 as it happens )

Regards

Godric

#9

Mark Brown wrote:

Mark... do you have something to add? =`:^>

I'm hoping you agree with me on this; seems nobody else does.

--
frosty

Quote:

godric (AT) trapezesoftware (DOT) com> wrote in message
news:1128678267.914647.31120 (AT) o13g2000cwo (DOT) googlegroups.com...

frosty wrote:
I wonder if the Pick (or at least UniData) ICONV function correctly
calculates leap years in antiquity. Running the following program
on UniData 6.0 gives an interesting result.

FOR Y = 1 TO 12000
LY1 = (ICONV('3-1-':Y,'D') - ICONV('2-28-':Y,'D')) - 1
LY2 = (REM(Y,4)=0) AND ((REM(Y,100)#0) OR (REM(Y,400)=0))
IF LY1 # LY2 THEN CRT Y:' ':LY1:' ':LY2
NEXT Y

The result is interesting in that it is not null; it indicates that
UniData
thinks the years 200 and 600 are leap years, but thinks that neither
400 nor 800 are leap years.

200 1 0
400 0 1
600 1 0
800 0 1

Comments?

--
frosty

The rules for determining leap years are actually -

Once every 4 years
Except every 100 years
Except every 400 years

So UNIDATA have it right..

We used to use it as a 'how many lines of code do you need to
determine a leap year' in Databasic ( 1 as it happens )

Regards

Godric