On Mon, Feb 21, 2022 at 2:23 PM R.E. Boss wrote:
>round=: [ * [: <. 0.5 + %~
Looking closer at this, this was a bit different from my initial attempt.
Also, a version which rounds to the nearest integer is good up through
cyclotomic 70.
71 is a prime number, so we should expect (71#1) as th
round=: [ * [: <. 0.5 + %~
R.E. Boss
-Original Message-
From: Programming On Behalf Of Raul
Miller
Sent: maandag 21 februari 2022 17:42
To: Programming forum
Subject: Re: [Jprogramming] polynomial division
Ok, ... I'm tempted to put that into the rosettacode entry.
What definiti
I should have added, the workaround is to use +/"]@:=
Henry Rich
On 2/21/2022 7:57 AM, Igor Zhuravlov wrote:
(+/@:= S: 0 <"0@i.@#) '' ; 1
0 1
(+/@:= S: 0 <"0@i.@#) (0 $ 0) ; 1
4 1
(+/@:= S: 0 <"0@i.@#) (0 $ 0) ; 1
3 1
(+/@:= S: 0 <"0@i.@#) (0 $ 00) ; 1
1 1
(+/@:= S: 0 <"0@i.
Ok, ... I'm tempted to put that into the rosettacode entry.
What definition are you using for 'round'?
When I try
round=: [: <. 0.5 + ]
it works up through cyclotomicREB 69 but when I use
round=: [ * [: <. 0.5 + %~
it only works up through cyclotomicREB 65.
Thanks,
--
Raul
On Mon, Feb
Fixed for next beta. The comparison compounds +/@:comparison did not
correctly detect an empty left argument.
Henry Rich
On 2/21/2022 7:57 AM, Igor Zhuravlov wrote:
(+/@:= S: 0 <"0@i.@#) '' ; 1
0 1
(+/@:= S: 0 <"0@i.@#) (0 $ 0) ; 1
4 1
(+/@:= S: 0 <"0@i.@#) (0 $ 0) ; 1
3 1
(+/
There was a mistake in taskfmt, but cyclotomic 6 was correct.
cyclotomicREB=:{{{."1 [0.01 round +. p. 1;^j.+:pi* y%~ >:I.1=y +. >:i.y}}
taskfmt=:{{rplc&'_-' rplc&('^1)';')';'x^0';'1') '
'-.~}:,|.')+',~"1'(',"1('1'-.~"1 ":,.0-.~y),"1 'x^',"1":,.I.|y}}
(cyclotomic"0 -: cyclotomicREB"0) >:i.30
1
That was the first thing I tried.
cyclotomic 6 should return 1 _1 1
Thanks,
--
Raul
On Mon, Feb 21, 2022 at 8:55 AM R.E. Boss wrote:
>
> From the hip.
>
> cyclotomic=:{{{."1 [0.01 round +. p. 1;^j.+:pi* y%~ >:I.1=y +. >:i.y}}
> taskfmt=:{{rplc&'_-' rplc&('^1';'';'x^0';'1') '
> '-.~}:,|.')+',
I can reproduce the problem on the last 4 lines. The others seem OK.
I'll look into it.
Henry Rich
On 2/21/2022 7:57 AM, Igor Zhuravlov wrote:
(+/@:= S: 0 <"0@i.@#) '' ; 1
0 1
(+/@:= S: 0 <"0@i.@#) (0 $ 0) ; 1
4 1
(+/@:= S: 0 <"0@i.@#) (0 $ 0) ; 1
3 1
(+/@:= S: 0 <"0@i.@#) (0
From the hip.
cyclotomic=:{{{."1 [0.01 round +. p. 1;^j.+:pi* y%~ >:I.1=y +. >:i.y}}
taskfmt=:{{rplc&'_-' rplc&('^1';'';'x^0';'1') '
'-.~}:,|.')+',~"1'(',"1('1'-.~"1 ":,.0-.~y),"1 'x^',"1":,.I.|y}}
([: taskfmt cyclotomic)"0 >:i.30
(x)+(-1)
I am on an older version
(+/@:= S: 0 <"0@i.@#) '' ; 1
0 1
(+/@:= S: 0 <"0@i.@#) (0 $ 0) ; 1
0 1
(+/@:= S: 0 <"0@i.@#) (0 $ 0) ; 1
0 1
(+/@:= S: 0 <"0@i.@#) (0 $ 00) ; 1
0 1
(+/@:= S: 0 <"0@i.@#) (0 $ 00) ; , 1
0 1
(+/@:= S: 0 <"0@i.@#) (0 $ 00) ; 01
0 1
(+/@:= S: 0 <"0@i.@#) (0
(+/@:= S: 0 <"0@i.@#) '' ; 1
0 1
(+/@:= S: 0 <"0@i.@#) (0 $ 0) ; 1
4 1
(+/@:= S: 0 <"0@i.@#) (0 $ 0) ; 1
3 1
(+/@:= S: 0 <"0@i.@#) (0 $ 00) ; 1
1 1
(+/@:= S: 0 <"0@i.@#) (0 $ 00) ; , 1
0 1
(+/@:= S: 0 <"0@i.@#) (0 $ 00) ; 01
0 1
(+/@:= S: 0 <"0@i.@#) (0 $ 00) ; , 01
1 1
Apologies for a mistake in my previous post, below.
This was where I said
"This trick for my [function] div is analogous to allowing for
leading decimal zeros
in 1234 % 0.0001 in "ordinary" long division."
I'd forgotten that J's representation of polynomials is not the same as its
repr
12 matches
Mail list logo
