The Nineteenth Byte

thejonymyster

17:00

@mathcat install new font, uninstall all old fonts

i'll leave "which font" to the creatives among ye

mathcat

uff

@thejonymyster Definitely Comic Sans

DLosc

Deja Vu Sans Mono

Seggan

17:20

@thejonymyster SW GameKeys

thejonymyster

@Seggan google isnt playing nice, link pls?

Seggan

17:43

azfonts.net/fonts/gamekeys-mt/regular-233183

thejonymyster

huh, that page just loads and loads forever for me :o will have to check at home maybe

Taylor Alex Raine

loads fine on my end

Adám

Anyone willing to help me with a little Python?

Ginger

sure

what do you need help with?

Adám

I'd like to define a function plus which is both able to take two integers and sum them, but also able to take two lists and sum them pairwise.

Ginger

17:56

hmmm

just a second

Adám

I tried lambda a, b: [x+y for x,y in zip([*a],[*b])] but that fails with 'int' object is not iterable

mathcat

you need to type check first, right?

Ginger

so you want:
plus(1, 1) -> 2
plus([1, 1], [1, 1]) -> [2, 2]
?

Adám

Yes, exactly, though the second example is ambiguous.

plus([1, 2], [10, 20]) -> [11, 22]

Ginger

ok, done: lambda a, b: [x+y for x,y in zip([*a],[*b])] if isinstance(a, list) else a + b

witness my Extreme Genius :b

mathcat

17:59

f=lambda a,b: a+b if type(a)==int else [x+y for x,y in zip([*a],[*b])]

Ginger

@mathcat less pythonic, doesn't use isinstance /j

thejonymyster

does it matter if the lists are different lengths

Adám

No.

Thanks all!

Ginger

np!

mathcat

Golfier, f=lambda a,b:a+b if type(a)==int else [x+y for x,y in zip(a,b)]

Seggan

18:05

Ginger

@Seggan me trying to quit vim

Seggan

Thats the SW GameKeys MT font i was talking abt

thejonymyster

oml

thank you thats horrible

mathcat

@Adám the [*a] isn't needed

Sandbox Posts

18:28

0

A: Sandbox for Proposed Challenges

Binary Art code-golf binary hexadecimal Take a list of 2-digit hexadecimal numbers as input, and output the binary value, replacing each 1 with an 'X', and each 0 with a space. For example Input = FF, 81, 47, 99. FF = 11111111 in binary, so print XXXXXXXX 81 = 10000001 in binary, so print X ...

Baby_Boy

18:39

I recently submitted a question of mine for a bounty with no deadline, how long should I expect to wait before it's awarded?

thejonymyster

@Baby_Boy Depends on which bounty, those aren't moderated or anything, so it completely depends on the user + circumstances

Baby_Boy

@thejonymyster alr, thanks ig

thejonymyster

link the bounty here and maybe someone here with experience can estimate?

like "oh that user doesnt go here anymore" or "oh those are usually done within the week" or something :P

Baby_Boy

@thejonymyster the bounty is good old questions, link is codegolf.meta.stackexchange.com/a/20846/106844

thejonymyster

i think i saw someone say that user (named user) isnt active anymore to award bounties, can anyone confirm? i cant find it in chat search

mathcat

18:48

true

Baby_Boy

@thejonymyster oh that sucks, we should probably delete that post for now tho to avoid confusion...

or edit it to let people know

thejonymyster

i agree, but that might be worth making a meta post about

"should we edit inactive bounty posts"

theres the "what if they come back argument" that comes to mind right away, but i wonder what everyone thinks in general

Baby_Boy

@thejonymyster if they come back, couldn't they unedit the post?

thejonymyster

true!

but yea idk its kind of a Thing to edit someone elses message to say "this isnt active" so idk, id make a meta post to be safe

Baby_Boy

@thejonymyster Should I make the meta post?

thejonymyster

18:56

yeah im at work :P cant get caught salcking

Baby_Boy

@thejonymyster alr, thanks for your help man

mousetail

@lyxal is there any other site that has significant amount of scrapable vyxal code besides here?

emanresu A

@Seggan I said 100

Seggan

yeah i just realized that this morning :P

emanresu A

@mousetail a) this is clearly incorrect b) There's around 1100 answers

Seggan

18:59

@mousetail codadict?

emanresu A

Nope

Also there appears to actually be around 1500

mousetail

I know my parser is a bit broken

It looks for posts where vyxal is one of the first 3 words

@emanresuA Why is it incorrect?

Also many posts where excluded because it couldn't find a <pre><code> block

I actually know next to nothing about vyxal

emanresu A

The corpus searches for Vyxal as a link

where body like "%Vyxal</a>%"

mousetail

That should find less posts though, not more

emanresu A

... no?

mousetail

19:10

Excerpt search excludes formatting

so there was a lot of vyxal with double space after

emanresu A

The online interpreter generates post templates with links, and people almost always use that

mousetail

In my downloaded data nothing had tags included because the API doesn't include formatting unless you explicitly ask for it

In any case what is the reason you think my graph is "obviously" wrong?

emanresu A

Something like ⇧ is useful, but not that useful

@mousetail SEDE

mousetail

I didn't use SEDE

I used the API

I'm gonna debug the difference tomorrow

emanresu A

SEDE works a lot better, use it

data.stackexchange.com/codegolf/query/1395643/vyxal-corpus

mousetail

19:14

@emanresuA Thanks, I'll double check my data and see if there is a mistake

I'm aware of where SEDE is but I need to process the data automatically

emanresu A

You can download csvs

And there's definitely a mistake, the SEDE query returns 1100 answers

mousetail

Yes my data set had over 1500 anwers

but a lot of them didn't have proper <pre><code> formatting

or had vyxal and another language

so I excluded them to be safe

this was posts containing the word vyxal anywhere

mathcat

btw how's the idea of a VR image / text displayed on a 2d-code?

wouldn't be useful, but I think it's cool

Feeds

1

Q: Proposal to mark Inactive posts in "List of bounties with no deadline" accordingly

I recently completed a bounty from List of bounties with no deadline, only to later realize it was inactive. This was a little disappointing and I believe this could easily be avoided by editing these posts to reflect them being inactive; either by adding strikethrough to their titles, or a littl...

emanresu A

@mousetail Yes, but this one gives entirely vyxal answers. you're the one who wants a larger sample size...

thejonymyster

19:59

LDQ: How does one handle string literals in a half byte language? I have some ideas but im curious if anyone else has any :-)

Seggan

i think nibbles reads full bytes inside a string

Ginger

disappointed by the fact that it is forbidden to use ATO to manufacture explosives

5

Seggan

hmm so i can do stuff that is illegal in the US but not in Finland?

Ginger

so it would seem, as long as it isn't any of the other things in that list

Seggan

@thejonymyster Yep, nibbles reads full bytes in the case of numbers or strings

thejonymyster

20:07

@Seggan ...yknow somehow i didnt think of that :P

im keeping my nibbley numbers since you dont really need full bytes for that, but full byte strings seem useful

Seggan

20:18

CMQ: anyone know what the most common digits in answers are?

Ginger

yes, actually! one second

@Seggan what you want is this, probably

Seggan

hmm interesting

but im building (manually) a huffman tree for Fig, and although that is useful, i still care about the middle digits

ty tho

never knew that

Ginger

np! happy to help and glad you learned a new thing

Radvylf Programs

@Seggan I've got a list for both Jelly and JS

For just digits, rather than numbers as a whole, 9 is the least common

But I don't think just the digits is very useful for like, golfing languages

DLosc

20:34

@thejonymyster I had some slightly convoluted ideas for HBL that I never got around to implementing. If you're interested I'll see if I can remember the details.

@Ginger It's also forbidden to use Python to run nuclear reactors, IIRC

Ginger

what

graffe

Is this answer just spam ? codegolf.stackexchange.com/a/252345/108721

Seggan

@RadvylfPrograms im making a huffman tree, so the digits matter in this case

graffe

If you look at the codev it didn't seem to have much to do with the challenge

emanresu A

@graffe OF COURSE not

graffe

20:36

If you look at the code t didn't seem to have much to do with the challenge

emanresu A

Spam is for stuff like "buy these cookies"

graffe

@emanresuA why do you say that?

emanresu A

> Spam is for stuff like "buy these cookies"

graffe

But its just some random code that prints some numbers

It doesn't seem related to the challenge

(I could be wrong of course)

emanresu A

> Spam is for stuff like "buy these cookies"

DLosc

20:37

@graffe It looks like an attempt at solving the problem, although maybe the poster didn't understand the problem.

emanresu A

@graffe ... Then why did you accept it

Ginger

@graffe this looks AI-generated lol

graffe

@Ginger ah

@emanresuA I didn't! What do you mean?

emanresu A

... then how come it has the accepted checkmark and it's your question

graffe

Interesting if it is AI generated

@emanresuA it shouldn't!

Let me see

@emanresuA I don't know how that happened but fixed now. Thanks for spotting it

DLosc

20:40

Did you intend to downvote it and clicked the check mark by mistake, maybe?

emanresu A

... what

thejonymyster

@DLosc i am always interested, but i cant guarantee ill be competent / crazy enough to implement it myself

graffe

@DLosc yes you are exactly right

Top sleuthing

It's the problem with using a phone

DLosc

Kinda funny that the "this answer is the best" button is right next to the "this answer is bad" button :P

graffe

Someone just upvote it!

@DLosc true! :)

emanresu A

20:41

@Ginger ... how

graffe

Please do downvote it if you think it is bad

I feel bad for the other excellent answer now

At least upvote that one!

Ginger

@emanresuA it's incoherent and confusingly made, although I guess that's most answers on this site

DLosc

@thejonymyster Let's take it to The Tarpit

emanresu A

... It appears to be running an optimised strategy that attempts to find 10 points on a line

graffe

@emanresuA you mean a pretrained strategy?

emanresu A

20:47

yes

And?

graffe

I have to decide who to award the bounty to soon. I really hope they explain what they have done so I can award it fairly

@emanresuA you might be right

There are two TIO links

It's not at all obvious why they are the same if they are meant to be

I added a comment begging for an explanation

Thanks @emanresuA

DLosc

@graffe One way of thinking of it: since you are free to award the bounty to whichever answer you choose, you can decide that an answer without an explanation is ineligible to receive the bounty.

Seggan

21:10

Looks fineish?

still unfinished

graffe

@DLosc I will have to do that but it's a bit sad if they don't offer an explanation and it's actually a better answer

Seggan

Fig hufmann tree ^^

graffe

I remember on math.se there is someone who posts answers to really hard integrals with no explanation

And it annoys everyone

DLosc

@Ginger Well, I can't find anything about this on Python.org's licenses page. I am pretty sure I recall reading it in a Python user agreement at some point, but maybe I'm misremembering. Or maybe it only applies to Python for Windows. :P

Seggan

21:33

conference.scipy.org/proceedings/scipy2015/pdfs/…

thejonymyster

@graffe so this is why teachers in school are always asking us to show our work

graffe

21:52

@thejonymyster yes! It’s to make sure you will get bounties on ccgc

2

Radvylf Programs

22:40

@Seggan I don't get why that's any different from an SBCS for this

Using digits to represent numbers other than 0-9 is outdated and never the best option for a golfing language

DLosc

@RadvylfPrograms What golfing languages don't use runs of digits to represent numbers? I think I agree that having multi-digit tokens isn't the best option, but if Jelly and Vyxal both do a thing I'm not sure you can call it "outdated" just yet.

Radvylf Programs

I don't know off the top of my head

But Jelly is still like 5 years old. While it's still very competitive, it still has plenty of room to be outdated in various ways. Vyxal presumably uses runs of digits because one of its design goals includes being accessible to new users.

Using digraphs and trigraphs (= integer compression) for anything bigger than 8 that isn't worth making a single byte is significantly golfier, like graph-doesn't-fit-on-the-same-page golfier, and it doesn't even cost code page space if you do it carefully

And that's only counting direct golfiness gains, not needing spacing for separating runs of digits is a free, and important, bonus

(You could always have good integer compression and digit runs, but the latter would be 100% redudnant and outclassed)

emanresu A

You can still represent numbers as numbers syntactically, if not encodingwise

Radvylf Programs

That's true, but then it's not SBCS anymore, and once you start doing that it gets way more complicated to talk about since you could also just use arithmetic coding and stuff

Or well I guess there's also the compiling-to-SBCS option

But I'm not arguing against digit runs in general, just in xBCS's :p

DLosc

@RadvylfPrograms This is what I'm planning for FunStack.

Codidact Posts

23:05

New Posts - Challenges - Code Golf: The holeyest base

posted on September 29, 2022 by trichoplax‭

Given a positive integer as input, indicate which base from 2 to 16 gives the most holes in the representation of the input in that base. The digits used are 0123456789ABCDEF. Note that ...

lyxal

23:41

@mousetail I more meant do you want it to be an officially recognised vyxal project (placed in the organisation, etc) or something independent

Transcript for