If you can keep the general exclusive nature of i. then first function is
easier
5 ([ + i.@-~) 20
5 6 7 8 9 10 11 12 13 14 15 16 17 18 19
then
steppedinterval =: ([ + i.@-~)/ : (-@[ {.\ ([ + i.@-~)/@]) NB. monadic = step
1
3 steppedinterval 5 20
5 8 11 14 17
On Sunday, March 29,
Since I use the debugging every now and then, I thought it would be easy to
answer Ruda's question but it was not. My first thought is that the name
of the verb in question is not "gettplog" but "gettplot_jzplot_". However
there appears to be both "gettplog_jzplot_" and "gettplog_jwplot"
which
Hello to All,
CORRECTION to my previous post:
I produced a horrible bug in the display of time/space measurements at the end
of the script.
With the bugfix, the time/space display, the use of ^: does not appear very
fast any more.
Please accept my apologies,
JoHo
I do apologise to all involved:
There was an inexcusable bug in the display of the test data
at the end of my script.
Verb rt in my script has been fixed, and consequently
the display of time/space data does show a completely
different picture now.
My versions using ^: are actually significantly
those nouns are decommitted, and replaced by verbs,
jpathsep
hostpathsep
winpathsep
On Mon, Mar 30, 2020, 3:35 AM Rudolf Sykora wrote:
> Dear list,
>
>
> I am running j901 and have tried to use the gnuplot addon on Windows.
> Following
> https://code.jsoftware.com/wiki/Addons/graphics/gnuplot
You can first isolate problem be related to pop up box by setting
wdinfo_z_=: echo
so there won't be any pop-up info box.
If crash gone away then it is a jqt issue.
On Mon, Mar 30, 2020, 12:43 AM Rudolf Sykora wrote:
> Dear list,
>
>
> if I do
>
>load 'plot'
>'ylog 1' plot 1 2 0 3
>
>
...a bugfix is in the works
there is a horrible bug in the display of the testing table
--
For information about J forums see http://www.jsoftware.com/forums.htm
Have I missed the Fold step?
--
(B=)
--
For information about J forums see http://www.jsoftware.com/forums.htm
Yet anotherone featuring ^: ...
]JVERSION
Engine: j807/j32/windows
Release-e: commercial/2019-11-04T10:14:12
Library: 8.07.26
Qt IDE: 1.7.9s/5.9.6
Platform: Win 32
NB. steps from a to b, using ^:
step=: fst=: 0&{
endv=: snd=: 1&{
srtv=: trd=: 2&{
tail=: {:
There's jpath, and nowadays, every recent OS will let you use '/' as a
path separator in most or all contexts.
But, looking at the gnuplot implementation, it's using jhostpath with
GNUPLOTBIN as an argument, and I don't see any definition for
GNUPLOTBIN, so I'd just replace that entire expression
Dear list,
I am running j901 and have tried to use the gnuplot addon on Windows.
Following
https://code.jsoftware.com/wiki/Addons/graphics/gnuplot
I tried the setup that could automatically run gnuplot from J. This
setup, however, uses some words which are apparently missing in j901,
namely
Brian Schott writes:
> I think you mean to use
> dbss'gettplog * : *'
I believe that's essentially equivalent to
dbstop 'gettplog'
Anyhow, using dbss does not help.
Thanks
Ruda
--
For information about J forums see
Using the wavefile.ijs that contains the VOSS routine for generating random
note sequences:
mynotes =: 30 voss 3
mynotes
7 8 6 11 13 14 6 9 7 12 10 9 12 12 8 11 11 7 11 13 6 8 12 11 15 11 10 12 6 10
NB. Then using the NumToName function in MusicalSyntax.ijs
NB. I get a boxed list of note names
as for the trailing 5s:
i: needs to fill with trailing 0s so you intermediately get
_2 _1 0 1 2
_2 0 2 0 0
adding 5 leads to your result
I once wrote this (here adjusted to Ruda’s example)
in case stepping is used really frequently
throughout the code — but there is too much
(two
Ruda,
I think you mean to use
dbss'gettplog * : *'
--
(B=)
--
For information about J forums see http://www.jsoftware.com/forums.htm
Dear list,
if I do
load 'plot'
'ylog 1' plot 1 2 0 3
then I get an error message about NaN in 'gettplog'. Something like that
is almost expected to happen (but not necessarily, __ coming from log 0
still isn't NaN, or is it?). I wanted to edit gettplog to see what could
be done, but
I have used this pair for a very long time,
to=. [ + 1 i.@+ <.@-~
inc=. (%&) ((<,'&.')`) (`:6) NB. replacing <'&.' to make it "future proof"
0 to 10
0 1 2 3 4 5 6 7 8 9 10
0 to (10 inc) 100
0 10 20 30 40 50 60 70 80 90 100
0j2 to (0.1j_0.2 inc) 1j0
0j2 0.1j1.8 0.2j1.6 0.3j1.4
As your examples illustrate, this is something of an underspecified problem.
But let's try a few things...
First, there's your basic i. which gets us a sequence:
i.1+14
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14
And adding step 3 means taking fewer steps and making them longer
3*i.<.1+14%3
0 3 6
Wrong forum...
But, also, you're kind of missing a point: when you are a programmer,
the skills and knowledge you pick up in re-implementation will help
you in programming contexts even if you end up working in a completely
different environment.
Yes, if you have an immediate deadline, heading
5+3*i.5
5 8 11 14 17
On 03/29/2020 05:52 AM, Rudolf Sykora wrote:
Dear all,
how do you usually generate a sequence of integers from 'a' to 'b' with
a step 's'?
If I ignore the step, yesterday I wrote
interval=.{.+(i. @ >: @ ({:-{.))
interval 5 19
5 6 7 8 9 10 11 12 13 14 15 16 17
To me the First big distinction between @s and is the fact that v is
executed dyadically with @ and monadically with &. This is likely suggested
by the width of the rectangles especially at the bottom of the diagram, but
it is not stated there, in favor of the phrase about each cell. But the
It also allows, compared to /, x (items) having a different shape than y
(accumulated result)
On Saturday, March 28, 2020, 07:19:13 p.m. EDT, Henry Rich
wrote:
You can do whatever you need without Fold, but it has these features:
* iteration can be forward or backward through items
I was thinking that i: with a complex argument would get you there, eg:
5+i:2 2j2
3 4 5 6 7
3 5 7 5 5 NB. why the trailing 5s?
but bending it to your wishes looks a bit tricky.
I vaguely recalled there was a utility verb, "steps." I've found it in
various addons;
I think it's in
Rudolf Sykora wrote:
> Dear all,
>
>
> how do you usually generate a sequence of integers from 'a' to 'b' with
> a step 's'?
>
> If I ignore the step, yesterday I wrote
>
>interval=.{.+(i. @ >: @ ({:-{.))
>interval 5 19
> 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19
>
> but that seems
Dear all,
how do you usually generate a sequence of integers from 'a' to 'b' with
a step 's'?
If I ignore the step, yesterday I wrote
interval=.{.+(i. @ >: @ ({:-{.))
interval 5 19
5 6 7 8 9 10 11 12 13 14 15 16 17 18 19
but that seems overly complicated...
Then I can include the step
could one as easily have a lilypond file created instead?
since converting from MusicXML usually doesn’t work that well
else I’d rather re-implement the music generation in scheme
(guile) in order to use it in lilypond directly
Am 29.03.20 um 05:14 schrieb Thomas McGuire:
I have a couple of
26 matches
Mail list logo
