The APL Orchard

LdBeth

00:07

if I get a binary pattern 0 0 1 1 0 1 1 and 1 2 3 4 5 6 7 how can I get (1 2)(3 4)(5)(6 7)?

Adám

@LdBeth So you want to group by runs of identical elements, i.e. cut on pair-wise inequality?

⋄ 0 0 1 1 0 1 1 {⍵⊂⍨1,2≠/⍺} 1 2 3 4 5 6 7

TryAPL

@Adám
┌───┬───┬─┬───┐
│1 2│3 4│5│6 7│
└───┴───┴─┴───┘

LdBeth

@Adám I thought about 2≠/ but forget the ⊂, thanks!

rabbitgrowth

03:05

Would it make sense to think of 1 1⍉m as extracting the diagonal of the matrix because if you "average out" the vertical and horizontal axes, you get a diagonal "axis"?

As in, the vertical and horizontal axes intersect at a 90° angle, and half of that is 45°?

PyGamer0

03:48

CMP (Chat Mini Poll): What do you find inconsistent / weird / irritating / want to change the language...with Dyalog APL?

LdBeth

04:24

@PyGamer0 I want do transformation on coordinate like image[x,y]←image[cos x, sin y] but figure it not easy to be done generically in APL

And I hope ⊥ and ⊤ can match J's behavior by operating on the first rank by default. but that's a minor one.

Perigord

04:57

How do you print large values without them turning into scientific notation?

LdBeth

@Perigord by modify ⎕pp

Adám

05:47

@Perigord Dyadic ⍕ will do that.

@LdBeth That would make them inconsistent with +.×

Brian BED

08:12

@PyGamer0 i wish you could do "⎕WC -?" to get a list of examples, similar to what you see in aplcart. you can do this with ] but not ⎕ which is pretty annoying imo. i dont think you can do it with ) either

xpqz

@PyGamer0 k-style dicts would be a neat addition

PyGamer0

what do yall think about not being able to use / in trains without doing something like ⊢⍤/

Adám

08:46

@PyGamer0 github.com/abrudz/dyalog_vision

@PyGamer0 ^ are all additions. I'd like the . operator split up into . for namespaces, a dyadic operator (maybe ˙) for inner product, and a monadic operator (not sure what glyph) for outer product.

Kamila Szewczyk

i'd love to see alternant natively in Dyalog.

-.× mat to compute the determinant, how cool is that?

Adám

That's a consistent extension (as in NARS2000 and SHARP APL).

@PyGamer0 I think I'd like APL2's inner product, though, and make reductions apply to major cells.

PyGamer0

09:10

@Adám yeah i have seen that and i wanted to ask about the depth operator

what does the depth operator do? is it similar to BQN's depth modifier?

Adám

Identical.

PyGamer0

hm ok

ill need to learn the rank operator and depth operator then

so depth operator is a generalized each

xpqz

09:31

@PyGamer0 rank is a force multiplier for your apl xpqz.github.io/learnapl/rank.html

PyGamer0

> So given a complex vector that contains a variety of scalars and higher-rank arrays, you can’t use the Rank operator to selectively apply a function to, say, just the vectors.

^ can the depth operator do that?

xpqz

yes

PyGamer0

so does rank pairs arguments uptill a specified rank?

like 1 2 3+⍤0 99⊢4 5 6 is (1 + 4 5 6)(2 + 4 5 6)(3 + 4 5 6)?

xpqz

Almost

     1 2 3+⍤0 99⊢4 5 6
5 6 7
6 7 8
7 8 9

PyGamer0

@xpqz what else does it do?

xpqz

09:39

You got the pairing correct, but not the rank of the result.

PyGamer0

oh

the result has the rank 2

xpqz

How about this:

      1 2 3+⍤¯1 1⊢3 3⍴⍳9
 2  3  4
 6  7  8
10 11 12

Negative rank is relative: it means "subtract from rank of argument"

PyGamer0

i think ↑(1+1 2 3)(2+4 5 6)(3+7 8 9)

xpqz

@PyGamer0 Here's what the Dyalog docs say: "The sub-arrays of R are the results of the individual applications of f. If these results differ in rank or shape, they are extended to a common rank and shape in the manner of Mix. See Mix."

help.dyalog.com/latest/index.htm#Language/Primitive%20Operators/…

You can test drive this: xpqz.github.io/cultivations/Rank.html

@PyGamer0 Correct

PyGamer0

@xpqz i got the the first rank drill right yay

although how will the rank's result's rank is decided?

xpqz

09:55

It's basically ↑ like you suspected

PyGamer0

ok

Brian BED

10:41

there seems to be a glitch with the ride editor. it seems to be adding and deleting spaces before the line when pasting stuff in

Kamila Szewczyk

it's been a thing for years now i think

i fixed it in my own fork

Brian BED

oo link?

Kamila Szewczyk

i didn't publish the patch

Brian BED

ah alright. i think ill just live with it then

Adám

@KamilaSzewczyk Did you log the bug? Maybe make a PR?

Kamila Szewczyk

10:45

i didn't

it's related to indentation width

notice how enter after brace gives you a bigger indent than what a paste results in

and when you paste on successive lines, the formatting doesn't occur

indentation in RIDE is in general broken beyond recognition.

e.g.

 f←{
     ⍝ ok

when I press enter after typing this comment, it doesn't indent the following line for me on stock RIDE.

Adám

11:37

@KamilaSzewczyk 4.4?

Kamila Szewczyk

IDE:
  Version: 4.4.3
  Platform: Linux x86_64
  Date: 2021-07-03 16:13:46 +0200
  Git commit: acb4b1d280380834a8021f04aec965fb4e486f68
  Preferences:{
    "colourScheme":"Dracula",
    "kbdLocale":"en_US",
    "lbarOrder":"← +-×÷*⍟⌹○!? |⌈⌊⊥⊤⊣⊢ =≠≤<>≥≡≢ ∨∧⍲⍱ ↑↓⊂⊃⊆⌷⍋⍒ ⍳⍸∊⍷∪∩~ /\\⌿⍀ ,⍪⍴⌽⊖⍉ ¨⍨⍣.∘⍤⍥@ ⍞⎕⍠⌸⌺⌶⍎⍕ ⋄⍝→⍵⍺∇& ¯⍬ ⍭√⍛⍢⍍…⍙⍗⍐φ⍦⍁⍮Φ⍡⌂ ",
    "otherExe":"/usr/bin/dyalog",
    "wse":"1"
  }

Interpreter:
  Version: 18.0.40684
  Platform: Linux-64
  Edition: Unicode/64
  Date: Jun 19 2021 at 00:08:29

Brian BED

11:53

      ⎕io
0
      1 1 2 2 2 3 3 3[1≡2]
1

can someone explain to me whats happening?

oh no wait

AHHHHhhhhhhh

ah nevermind i get it now

Brian BED

13:10

      {↑(∪⍵)(+/¨(1+⍳⌈/⍵)=⊆⍵)}1 1 2 2 2 3 3 3
1 2 3
2 3 3

there has to be a better way to get count of unique elements right?

my way is really inelegent :/

(this is with ⎕io←0)

dzaima

⎕←⍉ ,∘≢⌸ 1 1 2 2 2 3 3 3 ⊣ ⎕IO←0

TryAPL

@dzaima
1 2 3
2 3 3

Brian BED

wow thats awesome. thanks :O

ZtA0

A naive question: How could one merge two in parallel developed versions of a workspace (besides the new Link+git approach)?

LdBeth

14:40

Any reason for not using LINK?

Adám

14:57

@ZtA0 Maybe have a look at ]ws.compare?

LdBeth

16:46

would the method for matching parenthesis work well on things like " that are symmetric and no nesting allowed?

Adám

@LdBeth You could use it, but with alternating weights of 1 and ¯1, however, it is simply ≠\, e.g. ⋄ s↑⍤,⍥⊂(⊢∨≠\)'"'=s←'"this" is a "word"' and ⋄ s↑⍤,⍥⊂(⊢∨≠\)'"'=s←'is "this" is a "word"?'

TryAPL

@Adám
" t h i s "   i s   a   " w o r d "
1 1 1 1 1 1 0 0 0 0 0 0 1 1 1 1 1 1
i s   " t h i s "   i s   a   " w o r d " ?
0 0 0 1 1 1 1 1 1 0 0 0 0 0 0 1 1 1 1 1 1 0

LdBeth

17:14

@Adám thanks, it looks easier than what I though

ZtA0

@Adám Thanks, I will take a look.

ZtA0