Yes, thank you Roger for this explanation of dyadic & which i've recently
been thinking about how to incorporate into my j programming.
|Date: Sun, 16 Aug 2020 21:00:42 + (UTC)
|From: "'Pascal Jasmin' via Programming"
|To: "[email protected]"
|Subject: Re: [Jprogramming] conditional
Thanks, Henry. Back home now, so emailing from my laptop, where J902
resides,
is easier...
On startup, I get
allolddefs 1
++---+
|base|treerep|
++---+
+---++
|jsocket|ccdm|
+---++
+-++
|z|bind|
+-++
As far as I recall, these aren't my definitions
On the current implementation it is much preferable to use
bitmask operator@]^:["0 operand
rather than
bitmask 0&(] operator)"0 operand
They both execute at low rank, but the second version reinterprets the
internal form ( 0&(] operator)^:bitmask ) for each atom.
I have yet to find an app
dyadic m&f …
first I thought I should learn this construct
now I think I don’t need to
any suggestions?
is this considered an idiom that ought to be known and used?
Am 17.08.20 um 16:20 schrieb Henry Rich:
On the current implementation it is much preferable to use
bitmask operator@]^:["0 oper
You have to clean them up. Maybe you can beg someone else to do it, but
assume nothing. The last two are easy. On my system I see
bind_z_
2 : 'u@(v"_)'
ccdm_jsocket_
1 : '(''"wsock32" '',u)&(15!:0)'
These are OK. You will find they are different on your system, because
you haven't up
I think the primary benefit of that notation is instructional.
add=: 1&+
2 add 3
5
There's more efficient ways of achieving the same end, but efficiency
is not always the most important issue.
--
Raul
On Mon, Aug 17, 2020 at 10:25 AM Hauke Rehr wrote:
>
> dyadic m&f …
>
> first I though
Thanks Henry,
I woke up this morning realizing that the argument to & could be any value and
that cleared up a lot of my fuzziness on what was going on. Thank you to Pascal
as well for the explanation and Roger for twisting my brain for a few hours.
operand =. 2 2 2 $ 3 2 2 3 2 3 3 2
bitmask =
Late to the party - I had a look at the problem after the correspondence
closed, and have only
just got home to my own wifi.
I developed a few functions, tpxxxmd, not pretty in themselves, which
attempt a number of variations,
some using box characters, 19 22 25 26 { a.
Unfortunately, I not
With arguments that small, most of the time is spent in lexing and
parsing. To test the time of execution, you need arguments long enough
that the execution dominates. Or, you can make a tacit verb that has
done the parsing in advance.
To get serious measurements you need to delay the sampli
… the last version (meant to be raul’s) doesn’t quite look right …
can’t verify right now
Am 17.08.20 um 16:49 schrieb 'Michael Day' via Programming:
Late to the party - I had a look at the problem after the correspondence
closed, and have only
just got home to my own wifi.
I developed a few f
Tetration (I think it is called) is an impressive display of dyadic (m&)
You are allowed to keep adding (&m) bindings.
As a hybrid depth/breadth first (do depth 10 times in a breadth way), then
10 x&search^:(until)^:_ y
lets you make that chunk size a verb parameter.
With that said, there is
Thanks, Henry.
I thought I had run Package Manager on all my addons, but it seems
not. My excuse is
that we were away for a month, and snatching wifi when fitfully
available, so I might have
tried updating them and failed...
Anyway, for others who might be as silly as I, in case it's
Dear All
Please use the beta forum for any new messages on the 902 beta. Thanks.
Just a reminder - the programming forum is used for announcements,
and otherwise posts should be directed to the appropriate forum.
Eric announced the latest 902 beta to the programming forum, but as
his email stat
I took this as the definition of "raul" :
raul =: 13 : ';"1('' - '',&.>'' o+|''){~(+3*_1
1&E.)"1&.|:(*>./\.@:*)0,~2:`(}.@((i:,.]-1:) ~.))`((1,~]#_1:)"0)}~y'
I copied this from his email of 7/Aug/20 timed (here) at 23:21 . There
was a line break at *>./\.@: which
might have messed up the def
Now that I think of it, x m&v y is equivalent to x (m&v @ ] ^: [) y
so there is never going to be a case where dyadic m&v is necessary. The
question is how valuable the uses mentioned are. I know I have lost a
lot of time over the years by mistaken execution of dyadic m&v, and a
domain e
Thanks Henry,
I have increased argument sizes and defined using tacit. I am still not seeing
results that suggest the differences in the two approaches. Feel free to ignore
if I am just running down a rabbit hole, but if it is useful here are my
results.
JVERSION
Engine: j902/j64avx2/darwi
Yes.. I think mine needs a scan added, so that only the first + on a line
is retained.
This would add maybe a dozen characters to the line.
--
Raul
On Mon, Aug 17, 2020 at 10:49 AM 'Michael Day' via Programming <
[email protected]> wrote:
> Late to the party - I had a look at the probl
with nesting, I'm not sure of the equivalency
3 (2&(+&2@]^:[))^:[ 3
39
3 (2&(+&2)) 3
15
but the actual equivalent is
3 (2&(+&2@]^:[)@:])^:[ 3
15
+&2^:6 ] 3
15
+&2^:(18) ] 3
39
I'm not sure why the first is 39, but I understood why the short
expression/last 2 was 15.
But a short "verb
I think you are right. I see a persistent but small (<5%) difference in
my timings. Much less than I expected. The m&v version does
reinterpret (u^:n) for every atom, but it takes a fast path and does not
have to allocate memory.
Henry Rich
On 8/17/2020 11:56 AM, 'robert therriault' via Pr
The J program I'm currently working on uses Plot for showing multiple
versions of stock prices simultaneously. Hence, I need to
differentiate colors. I'm familiar with some minimal "standard"
colors such as red, green, blue, etc., but some of those shades are
too light or too dark for my purposes
Plot uses the graphics/color addon. The names are in
github.com/jsoftware/graphics_color/blob/master/colortab.ijs .
Plot uses uppercase names, like other Plot globals. For example:
load'plot'
ALICEBLUE_jzplot_
240 248 255
On Mon, Aug 17, 2020 at 6:02 PM HH PackRat wrote:
>
> The J program
On 8/17/20, chris burke wrote:
> Plot uses the graphics/color addon. The names are in
> github.com/jsoftware/graphics_color/blob/master/colortab.ijs .
Thanks, Chris! That's exactly what I was looking for!
Harvey
--
For informat
22 matches
Mail list logo
