@user1027871 you should be able to post messages in this chatroom now.

Welcome to the APL Orchard :D

If you have any further problems with user1027871's write permissions feel free to ping me :) I may not be around for a while tho D:

I am here now if anyone needs anything.

Are there any APL linting tools?

@Razetime dzaima/paste?

that's just a storage/presentations space right

I'm talking about static code checking

I was mentioning dzaima/paste because I thought you could take a peek at the code there, but I confused linting with syntax highlighting

ah

I'd taken a peek already for aplgolf

@Razetime there is barely anything for static APL parsing, which is very much required to be able to analyze anything at all

hm interesting

so first there needs to be one of that

@Razetime but that's impossible if you're using ⍺⍺/⍵⍵, ⍎, namespaces, or reading globals. (fwiw, Dyalog has a compiler, which does need static parsing, but it doesn't do anything if the things listed (and some others) are used)

@Adám the C comment expression is broken

      '/*'∘(≠\⍷∨¯1⌽∘⌽⍷∘⌽) 'a /* world */ b c'
┌→────────────────────────────────┐
│0 0 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0│
└~────────────────────────────────┘
      '/*'∘(≠\⍷∨¯1⌽∘⌽⍷∘⌽) 'a /* world */'
┌→────────────────────────┐
│1 1 0 0 0 0 0 0 0 0 0 0 0│
└~────────────────────────┘

@dzaima i suppose that can sorta be ignored

how do other languages with an eval function do it, then?

they mark eval as a severe linter error :')

neat

@Razetime depending on the language, eval might not actually completely thrash all ability to statically parse - the type of a variable doesn't usually change its usage

(and, fwiw, i do plan on experimenting with some static analysis for BQN at some point)

@dzaima yes please, I need more places to steal from

Can the scope of ⎕RL be limited within a Dyalog Class namespace, so reproducible numbers are generated on each call without being effected by use of Roll etc outside of the class?

@user3821550 does just adding ⎕RL←1 in the namespace not work?

yeah, and then you can access ? from inside the namespace with ns.?

just tested it and it does work

I have set ⎕RL to 1 1 within the top function in the class namespace, (which is being being run by MiServer) but on each call from MiServer, I don't get the same results. I will give ns.? a try. Thanks

you have to set it within the namespace itself, not a function in it

alternatively, you could do something like this

Setting ⎕RL to 1 1 in the class script and using ##.? appears to do the trick. Many thanks.

don't see why ##. would be needed

i posted a writeup on metaprogramming in APL, it's probably a bit lame but i like it: palaiologos.rocks/posts/dx-rewrite

@KamilaSzewczyk Reading it, but I like the colour scheme

Is it correct to say that it is (inspired by?) the Dracula theme?

sort of

first paragraph “and then move onto implementing it C or C++∙" → “and then move onto implementing it in C or C++”

The sentence “We check if this token '⍝'=” seems a bit odd to me. I think there is a word missing..?

woo, love the footnote functionality that allows you to jump back to where the footnote is

@RGS i mixed APL code with an english sentence

it's meant to say "if this token is '⍝'"

@KamilaSzewczyk that I understand, but I would still expect something more in there. Something like “We check if this token is '⍝'” or “We check if this token is ='⍝'”

but hey, feel free to ignore wtv I say; I'm just trying to be helpful ○/

“We take find the index in the table”

should be fixed now

it seems like i accidentally messed up the stylesheets and firefox refuses to tell me what's wrong

@KamilaSzewczyk What's messed up?

Tangent; when you write “I also decided to borrow Adam Brudzewsky’s range function and translate the tradfn alternant from the dfns workspace into a more sane and modern version:”, what are you saying you rewrote?

i took Adam's range function nearly verbatim

and I copied+pasted the alternant code and turned it into a dfn.

If you are talking about alt, it is actually a dfn: dfns.dyalog.com/c_alt.htm

the algorithm, variable names, etc.. are still the same

That is why it is called the dfns workspace :P

oh, well.

dfns.dyalog.com/n_alt.htm

this was a tradfn

Cool blog post!

I liked the domain extension through unicode characters. I do have a question; how does 60⌶ tokenise series of unicode characters?

Are they split up by the tokeniser?

@RGS i tried to add APL386 to the pre code blocks, but the font was proportional for some reason...?

@RGS yes

      60⌶'⍭√⍢⍛⍙…⍗φ⍁ΦΦ⌂'
┌─┬─┬─┬─┬─┬─┬─┬─┬─┬─┬─┬─┐
│⍭│√│⍢│⍛│⍙│…│⍗│φ│⍁│Φ│Φ│⌂│
└─┴─┴─┴─┴─┴─┴─┴─┴─┴─┴─┴─┘

@KamilaSzewczyk I wish I could help, but unless you ask me to test specific things, I'm afraid I won't be able to help much.

you can take a peek in the devtools if you feel like it

@KamilaSzewczyk Right, I figured I could've just opened the interpreter instead of bothering you 🙃🙃

yeah, I also have Fira Code; I also took a peak at "all fonts in the page" and APL 386 doesn't show up in there.

And also the stylesheets that are loaded for the page don't have any matches for the search "apl"

I removed it a second ago

because it didn't work

interestingly enough, this problem occurs only with APL386

So any other font works well?

Where are you loading the font from? Is it local or are you using some sort of CDN or wtv those things are called

i put it in my static folder

and i load it using @font-face { ...

that being said it's late for me already and i'll be heading to sleep.

⍤/

Bye, thanks for sharing the blog post ○/