@Adám ]box on gives a weird error

file SALTUtils, line 28:

⍝ 2017 07 19 Adam: [14756] trap EditorFix errors and exceptions; added DYALOG (install folder)

that's a comment right

file UCMD, line 17:

 ur←(dcl,⍨dcl≤notQ×⍴nma)⎕SE.SALTUtils.Spice in

hmm

looks like I did something, because new session fixed it

@EriktheOutgolfer So everything ok now?

@Adám um yeah

Sup guys! @Adám thanks for golfing this answer further, could you give it a read to see if it's all correctly explained?

@J.Sallé "which element(s) of G belong to each substring of the input" should probably say "which element(s) of G are members of the substrings of the input"

@J.Sallé I liked your old explanation It takes the length of the (right) argument and mixes the vector aeiou to itself that many times, yielding every possible combination of [2, length(right arg)] vowels.

@Adám okay great, I'll edit that c:

@Adám asking for a bit of opinion: is the top bar (i.e. the primitive list) in RIDE 4.0 a bit broken?

I fell like I should be able to add more "spaces"

CMQ: ^^ I'd appreciate others' opinions too, just pinged Adám in case he knows a reason for that.

@EriktheOutgolfer Adam must be already in Sabbath mode at this time, so I'll try to answer. This is either a regression or a conscious design decision, I don't know which. Previously, after rearranging the squiggles, you were guaranteed to have a space after the last squiggle which you could move to create new groups. You can log an issue here: github.com/Dyalog/ride/issues/new or I can tell you about a very hacky way to fix it.

well, it's not even Friday yet, but I'll respect his religious beliefs

@EriktheOutgolfer goodness, my mistake, I lost track of weekdays...

@ngn that only adds one space or infinite spaces?

btw I don't mind much

@EriktheOutgolfer I tried to type a few but the chat reformatted it

@EriktheOutgolfer a finite number

added issue

@EriktheOutgolfer thanks

a dyalog employee would probably reply to that

@EriktheOutgolfer yeah, or someone affiliated

@ngn Heh, I just commented with that idea. Great minds…

@Adám if my memory isn't failing me, I think it was already like that at some point - a space always appears after the last squiggle

@ngn Why don't you comment that it was like that? Maybe refer them to the old code?

@Adám I'm trying to find it

@J.Sallé I can't help noticing, the APL expression at the end of Adam's "QuadS" answer is <53 chars and still golfable.

@ngn Yeah, but J. Sallé is clearly going for a "pure" APL solution there, not using RegEx. An appealing approach. Also, my "equivalent" is intended to be just that – equivalent – not golfed.

@Adám @ngn what was the last time you two had written a non-golfed APL? :P

@Uriel "a non-golfed APL" what? Program? Together?

@Adám whatever APL codes are called, program, script, de-texified math notebook

@Uriel 3.5 hours ago.

@Uriel Non-golfed? Yuck. Why would anybody write that :)

@Uriel Seriously: I'm not an APL programmer. I use APL only for golfing.

@Adám I don't think he meant a non-golfed implementation :)

@ngn so what are you using on daily job? JS?

@Uriel it depends on the job

@Uriel js and python have proved the most practical scripting languages for me

@Uriel from the APL family I like k most

@Uriel and sometimes I have to hold my nose and do things in Java or C#

@ngn c# is actually pretty OK. I don't really like it because it's only .NET and needs VS to run (slowly), but syntatically it's OK

Java is a nighmare of OOPification

K looks abit more readable than J

And Python and js are pretty awesome

@Uriel C# is not ok (in my opinion), it's only a M$-focused and slightly less retarded Java

@Uriel K is reduced instruction set, and has only depth, not rank.

@Adám with respect, I don't think this is an accurate description of k

@Adám "reduced instruction set APL" is a misleading phrase

@Adám indeed, k shares a small number of primitives with APL, but it also borrows important concepts from Lisp that are not present in APL

@ngn I didn't mean that as a criticism. I'm somewhat bothered by APL and J's OTT features.

@Adám also, it does have rank: only 0 and 1

@ngn No, that's depth 0 and 1.

@Adám it has atoms(i.e. scalars) and lists(vectors)

@ngn True, I think K is even more ingeniously put together than APL. And Arthur's starting over from scratch for every version is the right way to get purity, consistency, and the right selection of primitives (for the desired target audience).

⎕←≡4 ⋄ ⎕←≡4 2

@Adám
0
1

@ngn ^ Depth≡Rank for simple scalars and simple vectors.

⎕←⍴⍴4 ⋄ ⎕←⍴⍴4 2

@Adám
0
1

@Adám k can also do (0 1;2 3)

@ngn Does K have a depth/rank primitive/idiom?

@Adám (0 1;2 3) has depth 2 and rank 1 (in APL's terminology)

@Adám only count - monadic #

@Adám you can distinguish between atom and list by other means: @x is the type of x; types of atoms are negative

@ngn Yes. But afaict, rank is a pretty useless concept in k, as it is equal to depth for simple scalars and vectors, and does not change with nested vectors. Only depth actually carries meaning.

@Adám yeah, it's simpler :)

@ngn 0 3⍴0 in K?

@ngn simpler = reduced (instruction set)

@Adám probably 0#,3#0

@ngn That's the same as (). Just tried in latest K.

@Adám old versions supported "prototypical elements" but I don't know about the latest one

@Adám so he's decided to do away with prototypes... there were rumours

@ngn Actually, the result looks the same as (). What is K's ≡?

@Adám ~

@ngn Yup, 1b :-)

@Adám if it looks the same, it is the same; k doesn't have APL's stupid output formatting...

@ngn well, Microsoft still is a huge portion of Enterprise, private PC and IoT computing

@ngn Another upside of K. I'd want to make a new APL, taking the best from APL, K, and J. Inbreeding ftw.

@Adám the output is formatted as the (reasonably) simplest expression that would produce it

@Uriel there is no strength in numbers when it comes to designing programming languages

@ngn No, I think it is more like the expression which uses the fewest functions, other than , and ;

@Adám ; is not a function