It could be great to have and. and or. in if. and while. conditions, like
if. a~:0 and. b wrote:
> ...not forgetting that *. is least common multiple for integer arguments.
>
> eg
>
>12 *. 16
> 48
>
> Mike
>
> Please reply to mike_liz@tiscali.co.uk.
> Sent from my iPad
>
>> On 6 Nov 2017,
t;
> Nevertheless, illogical as your way is, I would rather have it your way. I
> have often wanted (v0`v1`v2)} y to mean
> (v0 y) (v1 y)} (v2 y), and never wanted it to mean what it does.
>
> However, I don't think the improvement would justify the effort and the
> inco
ote:
> You are using the verb (monad }). You need (dyad }).
>
> Henry Rich
>
>
> On 1/5/2017 8:16 PM, Moon S wrote:
>>
>> Hm, and v0`v1`v2 has some quirks...
>>
>> v0=:_1 1&*@{. NB. head item, multiplied by _1 1
>> v1=:_2&{.
ng
wrote:
> (+/)`%`# 1 : ' ''`f g h'' =. m label_. (f g h) f.'
> +/ % #
>
>
>
>
> - Original Message -
> From: Moon S
> To: programming
> Sent: Thursday, January 5, 2017 7:26 PM
> Subject: Re: [Jprogramming] Tacit definition
>
> - Original Message -
> From: Moon S
> To: programming
> Sent: Thursday, January 5, 2017 6:36 PM
> Subject: [Jprogramming] Tacit definition involving adverb argument
>
> Is there a tacit definition of such a verb? Is there a way to pass a
> value to the adverb?
Is there a tacit definition of such a verb? Is there a way to pass a
value to the adverb?
f =: 3 : '0 (<:y)} 11 22 33'
f 1
0 22 33
f 2
11 0 33
PS. Also, I couldn't suspect that besides a usual definition of dyad
adverb such as
1 : ('';':';'...x...m...y...') NB. explicit defs inside
the
> > In brief, load debug with:
> >
> >load 'debug'
> >
> > and press Ctrl-K to show the debugger.
> >
> > On Mon, Jan 2, 2017 at 2:11 PM, Moon S wrote:
> >
> > > 1) JQt (8.05, j64, windows, all latest) can't debug a progr
TurboPascal... 1983... no, we know better :)
dbstop was the key. Thank you Chris!
On Tue, Jan 3, 2017 at 1:07 AM, chris burke wrote:
> See the lab: Help|Studio|Labs|Debug
>
> In brief, load debug with:
>
>load 'debug'
>
> and press Ctrl-K to show the deb
1) JQt (8.05, j64, windows, all latest) can't debug a program. I mean, run
it line by line, entering or stepping over functions. At least I couldn't
find how to do it. Moreover, it can't run a program: file > open > latest
... someprogram.ijs --> new edit window with green background pops up, then
I won't lie, first I wrote that solution in Python. It's much easier and
faster. Then the program was translated to J. The biggest problem was that
J didn't have dictionaries. While even K has! And sparse arrays don't
support strings or boxed values. Argh. And we are at the end of year 2016!
Anyway
t; # n&>) t will do.
On Sun, Dec 4, 2016 at 9:26 PM, Henry Rich wrote:
> (p#n)&> applies (p#n) to each atom of t.
>
> (p&> # n&>) applies # to the results of p&> and n&> on the entirety of t.
>
> Both forms are valid J, but they do different thin
(p&> # n&>) t NB. is ok too. So... why (p#n)&>t is not? And how to deal
with this kind of problems?
On Sun, Dec 4, 2016 at 8:18 PM, Moon S wrote:
> OK, simple case
>
>t NB. three boxed strings
>
> ┌──┬──┬──┐
>
> │22│33│44│
>
> └──┴──
20
0
0
dissect doesn't show anything for p and n in the second case; and what's "3
1(1)" for '#'?
http://mas.orgfree.com/adventofcode2016/diss1.png
On Sun, Dec 4, 2016 at 6:14 PM, Henry Rich wrote:
> I have no time to help, other than to suggest you use d
I have this code:
t =: cutLF CR-.~fread '04.dat'
a =: a.{~97+i.26NB. 'abc...z'
c =: 5{.a\:[:+/"1 a=/[:/:~_10}.'-'-.~] NB. calculate checksum string
v =: c -: _1}._6&{. NB. checksums match = good string
n =: [:".3{._10{.] NB. extract
bug, the size of window is not
> > > > QT_DEFSIZE but the minwh, and the plot window
> > > > failed to resize.
> > > >
> > > > Пт, 09 сен 2016, Robert Herman написал(а):
> > > > > Bill
> > > > >
> > >
As a workaround, this works:
pd 'qtc c:\tmp\plot.png 1880 360'
exit 0
It creates an image file with the graphics, shows it with the default image
viewer, and exits jqt.
On Fri, Sep 9, 2016 at 3:21 PM, Moon S wrote:
> Yes it should have been something like pd 'qt 1880
window still defaults to 400 386. Where is the QT_DEFSIZE var defined,
> or how can I change it in a program?
>
> Thanks.
>
> Rob
>
>
>
> On 9 September 2016 at 17:47, bill lam wrote:
>
> > It should in plot/commands > output commands.
> > eg,
In JQt (8.05 Win 64) I do something like this:
require'plot'
pd'reset'
pd (i.1000);?1000#100
pd'show'
I get the chart on my screen with the client area sized 480x360 pixels.
I have read wiki pages on plot, pd options and commands, but I could not
find a way to make it bigger, e.g. 1880x360. Ho
andle. And we are fast enough that we can
> switch to arbitrary precision integer arithmetic and still be faster
> than the brute force approach:
>
> (3) 6!:2 'F x:i.'
> 0.042063
>
> I hope this helps...
>
> Thanks,
>
> --
> Raul
>
> On Wed,
9)'
>
> 0.169901
>
> 6!:2 '(+/@:(*:@>:@i.))"0 (i.)'
>
> 13.4185
>
> 6!:2 '(+/@:(*:@:>:@i.))"0 (i.)'
>
> 0.156401
>
> 6!:2 '(+/@:(*:@>:@:i.))"0 (i.)'
>
> 13.49
>
>
> Note that there
I was searching integer solutions (k,m) for
1^2 + 2^2 + ... + k^2 = m^2
and I found that one expression runs much faster than the other:
(#~(0=1|[:%:+/@:([:*:1+i.))"0) 2+i.
24
(#~(0=1|[:%:+/@:(*:&>:&i.))"0) 2+i.
24
The first one is ~100 times faster, and moreover, the expression
It seems that there's no extended hexadecimal literals in J.
The decimal ones work ok. Let's convert a long integer to its hex
representation
'0123456789abcdef'{~16#.^:_1 (1111111x)
53973c0482de06eddf68d4531e9062c7
16b53973c0482de06eddf68d4531e9062c7 N
Here's another code golf challenge I found:
http://www.blonde.net/blog/2016/01/18/blonde-code-golf-january-16
In a string, swap a-z with z-a, A-Z with Z-A. Assume that you will always
get a valid string and that spaces and special characters should remain
unchanged.
alphaSwap('abcxyz') // -> zyx
I had nearly the same for the last day. Then remembered that J has modular
exponentiation optimized. And I have 64-bit J, so no need for extended
numbers.
3 timex '(33554393&|@*&252533)^:(2947(<.&-:@(*:@+-3&*@[+]))3029)20151125'
5.12972
3 timex
'33554393|20151125*(33554393&|@^&(2947(<.&-:@(*
±1 or ±½ of course :)
Sorry didn't read all, just saw that -n/2 ... 0 ... n/2.
For odd n it's i:<.-:n
i:<.-:5
_2 _1 0 1 2
On Tue, Dec 22, 2015 at 8:20 PM, Moon S wrote:
> > NB. Creat a rotation vector from -n/2 ... 0 ... n/2
>
> Try i: -: y
>
> On Tue
> NB. Creat a rotation vector from -n/2 ... 0 ... n/2
Try i: -: y
On Tue, Dec 22, 2015 at 7:50 PM, Richard Donovan
wrote:
> Thanks Thomas, works very well!
>
> > From: tmcguir...@gmail.com
> > Date: Tue, 22 Dec 2015 12:14:31 -0500
> > To: programm...@jsoftware.com
> > Subject: Re: [Jprogramming
I found the reason why convert/misc/md5 didn't work on the 64-bit
architecture.
The failures came from the rot function -- it was ok for 32-bit ints, but
wrong for 64-bit.
This is what works ok:
rot=: 16b and sh or ] sh~ 32 -~ [ NB. (y << x) | (y >> (32 - x))
I saved it in https://github.
I changed those keys to numbers... and cheated a bit with < and >
t=: ;: > cutLF (':,',CR)-.~ fread {:ARGV NB.
|Sue|440|pomeranians|1|samoyeds|9|children|4|
c=: ;: 'children cats samoyeds pomeranians akitas vizslas goldfish trees
cars perfumes' NB. categories
a=: ".@>@{: + (100 * c i. {.) NB. ana
; happPair=: +&({&Tbl)&< |. NB. happPair 2 3
> happ=: +/@:(2 happPair\ (, {.)) NB. happ i.8
> smoutput happMax=: >./ happ"1 (i.@! A. i.) # Tbl
>
> NB. part 2
> Tbl=: Tbl {.~ >: 2##Tbl
> happPair=: +&({&Tbl)&< |.
> happ=: +/@:(2 happPair\ (, {.
So far these were simple tasks with clear algorithms, so it's interesting
to play with the programs -- keep everything in verbs and be as lazy as
possible (i.e. build one final function, that just gives the solution of
the task being given the data; especially funny if the functions are
tacit), or
For part1 you don't need json actually ;)
part1 =: 3 : 0 NB. replace all non-digits to spaces, replace '-' to '_', do
+/".
a=.'abcdefghijklmnopqrstuvwxyz"'',[]{}:',CRLF
+/". ('-';'_')rplc~ ' '((i.#y)#~+./a=/y)}y
)
On Sat, Dec 12, 2015 at 10:47 PM, Joe Bogner wrote:
> somewhat tricky as I h
Such a pity J is not too popular on github.
Anyway, here's my solution too. In general, I don't like to use tacit defs
everywhere. Procedural logic is much more clear and easier to read and
understand and not always slower.
NB. next good password http://adventofcode.com/day/11
ilo =: (+./)@(8 11
1|.o;/|.l is a bit slower than the for (1.5s).
G.P.
On Tue, Feb 3, 2015 at 7:41 PM, Moon S wrote:
> I see. Thank you!
> Still it's a pity there's no way to tell '/' an initial value without
> boxing/unboxing.
>
>
>(1) 6!:2 '>(v~&.>)
'q=:q v y'' "0 l[q=:o'
0.0038912
On Tue, Feb 3, 2015 at 7:30 PM, 'Pascal Jasmin' via Programming <
programm...@jsoftware.com> wrote:
> (,: 'xx')4 : 'x,''_'',":y'~ reduce 1 2 3 4
> xx_4_3_2_1
: ((&.>)/)(>@:)
>
> 100([:-~boxscan ,~)&(<"_1) i.5
> 90
>
>
> 100([:(- +/"1)~ boxscan ,~)&(<"_1) i.5 2
> 55
>
>
> then as an adverb,
>
> reduce =: 1 : '([: u boxscan ,~)&(<"_1)'
>
>
0
> > +---+
> > |_45 _44 _43 _42 _41|
> > +---+
> >
> >
> > R.E. Boss
> >
> >
> > > -Original Message-
> > > From: programming-boun...@forums.jsoftware.com
> > [mailto:programming-
> > >
I have a list (l), some object (o) and a verb (v) modifying the object:
l = l_0 l_1 ... l_n
o_new = o_old v l_i
The list items and the object have different types. I want to apply all the
list items to the object:
o_new = (...((o_old v l_0) v l_1) ... v l_n) NB. the order of items
doesn't ma
37 matches
Mail list logo
