The Nineteenth Byte

New Posts

00:15

4

The string tut-tutut-tut-tuttut-tut can be constructed with overlapping or concatenated instances of the word tut-tut: tut-tutut-tut-tuttut-tut tut-tut tut-tut tut-tut tut-tut The string tut-tututut-tutut-tut cannot: tut-tututut-tutut-tut tut-tut^ |tut-tut...

code-golf

Niko

@NewPosts good one

OldSandboxPosts

01:00

Sandbox posts last active a week ago: Persistence of a number, Strategy: The cunning cousin witch

Seggan

01:19

@Steffan noice!

is it on GH?

also why does it take waaaaay too long?

oh, and I noticed that you don't have the fancy mathjax header generated

thanks very much tho

Steffan

i'll do the fancy mathjax thing if u want

yes

github.com/Steffan153/fig-interpreter

it takes a little bit for java to start up and run the jar (plus connecting to the api)

Seggan

@Steffan yeah because technically Fig's score is not that exact number, its the log_whatever

@Steffan ah k

ill occasionally pr the new fig jars then

lyxal

where's the Fig command list

Seggan

gist.github.com/Seggan/2a3eefa03c299174fc636dac69deb73d

unfinished tutorial github.com/Seggan/Fig/wiki/Tutorial

lyxal

01:41

The start of a FigBuzz: Ma100'S*H"FizzBuzz"=0%w3 5

~21.401 bytes right now

Seggan

try n

feel free to post it

lyxal

it's not done

because someone forgot to add logical or

and/or dyadic maximum

Seggan

G

lyxal

@Seggan n (3) - (any, num, fun) apply c on every bth element in a?

Seggan

ya

lyxal

01:42

DEBUG:
Exception in thread "main" clojure.lang.ArityException: Wrong number of args (1) passed to: fig.helpers/cmp
at clojure.lang.AFn.throwArity(AFn.java:429)
at clojure.lang.AFn.invoke(AFn.java:32)
at clojure.core$max_key.invokeStatic(core.clj:4985)
at fig.impl$generate.invokeStatic(impl.clj:124)
at fig.impl$generate.invoke(impl.clj:119)
at clojure.lang.AFn.applyToHelper(AFn.java:156)
at clojure.lang.AFn.applyTo(AFn.java:144)
at clojure.core$apply.invokeStatic(core.clj:667)
at fig.impl$interpret.invokeStatic(impl.clj:454)

Seggan

@lyxal G is dyadic maximum. its hidden

aaaaaaaaaaaaaaaaa

i think i set the arity wrong

i thought i tested it

ah oops i see

building 1.0.3 now

should be in the online interpreter soon

lyxal

fig-lang.herokuapp.com/…

Hmm yes apply at every nth element

Yes very good

Try n they said

It'll work they said

@Seggan look where your precious n got me

Seggan

hmm yes yet another deficiency in testing

lyxal

Also no compressed string for fizzbuzz got me feeling big yeet

Seggan

yeah

hmm a swapped argument order in a certain mod in a certain map is the fault of n

time for the 5th release in 2 days

and the 2nd in 5 minutes

lyxal

02:01

Even with your questionable programming I found a way to figbuzz

0

A: 1, 2, Fizz, 4, Buzz

Fig, \$35\log_{256}(96)\approx\$ 28.809 bytes j" "Ma100']^@LwS*H"FizzBuzz"=0%w3 5 Try it online! Hmm yes irrational bytes.

Seggan

cn is an alternative to a newline literal

lyxal

DEBUG:
Exception in thread "main" java.lang.ClassCastException: class java.lang.Long cannot be cast to class java.lang.String (java.lang.Long and java.lang.String are in module java.base of loader 'bootstrap')
at java.base/java.lang.String.compareTo(String.java:140)
at clojure.lang.Util.compare(Util.java:153)
at fig.helpers$cmp.invokeStatic(helpers.clj:34)
at fig.impl$generate.invokeStatic(impl.clj:125)
at fig.impl$generate.invoke(impl.clj:119)
at clojure.lang.AFn.applyToHelper(AFn.java:156)
at clojure.lang.AFn.applyTo(AFn.java:144)

Thank you for the lack of string, number comparison

Seggan

AAAAAAAAAAAAAAAAAA

i thought i had that

i knew i shouldve waitied

keep smashing fig, ill release it all tomorrow

strangest thing is, i have string number equality

Seggan

02:30

also apparently fizz is not in the dict...

i better find a new dict

thejonymyster

just add it in manually :P

Seggan

yeah i did that

att

02:59

what's with the fractional bytes

thejonymyster

a wizard did it

att

03:14

how is that measured

Bubbler

If a language has a character set of n distinct characters, it uses log_256(n) bytes per character

assuming n >= 2

Bubbler

03:36

(allowed since this discussion)

lyxal

05:33

@Seggan you added the lowercase version but forgot to add the uppercase version

mousetail

06:13

@Steffan Isn't Heroku abandoning it's free tier?

lyxal

@Seggan looks like filtering by a function is a little broken

as is mapping actually

where's that empty list coming from

Neil

07:31

phew, managed to get Charcoal competitive again on an ascii-art challenge

pxeger

10:43

→ 1 message moved to The Nineteenth Bakery

Ginger

12:15

jjjjjjjjjhello there

Jo King

g'joink

Ginger

wait that's lyxal's line

thejonymyster

@mousetail wtf not pog at all :(

i was looking forward to using that in the future

Ginger

haha imagine using heroku

thejonymyster

i cant anymore

dream killed

Ginger

12:19

anyone have a complete list of the X colors in a python list format?

pxeger

try googling?

lyxal

@Ginger ["#863333", "#86334d", "#333368", "#333333"]

mousetail

@thejonymyster Agreed

I think the era of free online services is ending

With the economy being fucked and inflation etc.

be ready to backup your things from other services too

thejonymyster

OTL

mousetail

OTL?

lyxal

12:23

A misspelling of oil?

mousetail

Oil? Never heared of it

Ginger

TIL I Learned that the dark-gray HTML color is substantially lighter than plain gray

9

thejonymyster

OTL

see also: orz

mousetail

I'm very confused

Ginger

that's pretty clever actually

thejonymyster

12:24

@mousetail its a duder on the ground

lyxal

@Ginger wait what do you mean by X colours?

thejonymyster

see diagram in link

mousetail

How is it related to anything though?

thejonymyster

that is my reaction to the things you said

because they are so dire

i have fallen down in despair

mousetail

Understood

I'd stand up though to keep your head above water when the sea level rises

2

thejonymyster

12:26

oh hey ok there is a kym for it under orz

i was like "why isnt anything showing up other than random sites"

but i digress, shit sucks is the point

lyxal

no

shit doesn't suck

vacuum cleaners do

mousetail

If your shit sucks you need to see a doctor

lyxal

and then the patent office

user

12:38

@mousetail idk about this but Akka is no longer going to be FOSS, very sad

mousetail

It's all related

lyxal

ight that's it time to close source vyxal and charge $1 morbillion dollars to use it

pxeger

Do you really want Vyxal to only ever be used by Marvel?

lyxal

If it allows it to be aired in cinemas twice, then yes!

mousetail

A Vyxal movie would be 0.02 seconds but have a more complex story than Endgame

caird coinheringaahing

12:45

@lyxal Honestly, best moment of CGCC history was when lyxal went "It's vyxal time"

6

mousetail

13:15

I got a lot more starred messages but I guess it was the same person starring every time

thejonymyster

lmao thatd be a funny badge too though

'secret admirer'

not exactly something to encourage in a badge probably

Radvylf Programs

lol

mousetail

That should be a badge for sure

Can mods see who stars messages?

caird coinheringaahing

Not anymore

Doorknob could, but no one else

mousetail

How does that make sense. A specific mod that can but others can't?

caird coinheringaahing

13:21

94

Q: I'm seeing stars! (I can see who starred a message and so can you)

So I was making a chatbot in Ruby for SE chat, and I discovered that I could find out the starrer of a message. I'm pretty sure stars, like votes, are supposed to be anonymous. Although this knowledge would help for cases of star trolls like this. Here's the specific slice of code that do...

support bug status-completed chat chat-stars

mousetail

Ah ok, everyone could see but he was the only one who knew how to use inspect element

caird coinheringaahing

Not inspect element, the chat events being sent to bots :P

mousetail

ok I have inspect element open and I'm ready to see what the next star message will look like

thejonymyster

quick someone be notable and or funny or maybe even a little bit clever

lyxal

2 mins ago, by caird coinheringaahing

Not inspect element, the chat events being sent to bots :P

Radvylf Programs

13:25

@mousetail ...it will look like a normal message with a little <img> next to it

mousetail

Is event type 6 a star event?

Radvylf Programs

Yes

caird coinheringaahing

No

mousetail

Thanks

caird coinheringaahing

That is, it isn't the event that tells you who starred a message

mousetail

13:27

What is interesting is that you can see how often a message was edited

Radvylf Programs

Well no, but it was

mousetail

that's not visible in the UI either

Radvylf Programs

It is in the history tho

caird coinheringaahing

And, if you're an RO/mod, you can read deleted messages

Radvylf Programs

Which I find weird. Deleted messages seem a lot more private than stars

mousetail

13:28

It's a matter of convenience I guess. More effort to show starrers then deleted messages

caird coinheringaahing

Tbf, I think it's a moderation thing

thejonymyster

what does it look like?

reading a deleted message, i mean

Radvylf Programs

You just have a history link, like with an edited one

lyxal

@cairdcoinheringaahing I agree - it helps ROs and mods know what's happened in a conversation

caird coinheringaahing

If someone comes into the room, starts posting a bunch of flag-worthy stuff, and it all gets deleted, the ROs should be able to go back and determine how bad it was, and if it merits further action

thejonymyster

13:29

^^^ ok that makes sense yeah

Jo King

@thejonymyster they're mostly composed of letters

lyxal

caird coinheringaahing

@JoKing A few symbols and numbers as well

lyxal

Radvylf Programs

@cairdcoinheringaahing IMO that should be mod-flag territory

Deleted content shouldn't be visible to ROs

With the possible exception of flag-auto-deleted stuff

thejonymyster

13:30

just dont look then :P

caird coinheringaahing

I guess there's a line between mods and ROs, and that could lie on either side

Radvylf Programs

@thejonymyster Well I'm not the ROs I'm concerned about :p

caird coinheringaahing

You can also raise a mod flag to purge a message, which prevents ROs from reading it

Radvylf Programs

@cairdcoinheringaahing Yeah, the line is that any random person can become an RO. If you create a room and trick someone into like, sharing their credit card info or site password, they've got no way to keep you from continuing to see it without a moderator, which takes ages

caird coinheringaahing

Which, if you have an info that really shouldn't be read, that's what you should do anyway

Radvylf Programs

13:32

If deleting actually got rid of the information tho, for non-mods, it wouldn't be as bad

caird coinheringaahing

@RadvylfPrograms Mod flags in chat get handled pretty quickly, and attract the attention of basically every mod online :P

Radvylf Programs

I've never had one handled in less than 15m

pxeger

don't try it though!

caird coinheringaahing

Obviously, you don't want that to happen, but 15 mods appearing isn't great :P

Radvylf Programs

Maybe I've just gotten unlucky with my timing

pxeger

13:33

It's funny, you can sometimes tell when flags get raised because 3 users from other sites will suddenly appear in the room

thejonymyster

mods are like sharks: if they smell blood, theyll bite

3

lyxal

in my experience they nuke your fridge instead

thejonymyster

just like sharks..

lyxal

the hell kinda sharks are you thinking of that have nukes?

mousetail

Ideally they'd be like life guards. If they smell blood they'll come but not make the situation worse

caird coinheringaahing

13:34

@thejonymyster Yeah, I've lost at least 4 fridges to nuke sharks

pxeger

I think I've only ever raised a chat mod flag once, and that was for one particular room which had been causing flags regularly for weeks

hyper-neutrino

@cairdcoinheringaahing well only if it was edited first

thejonymyster

LDQ: what sort of error messages should my golfing lang give? is there any way to make it a useful output format? :P

hyper-neutrino

if the message is already deleted nobody can edit it anymore so even if it's purged the latest version is still visible

pxeger

why??

that's ridiculous

hyper-neutrino

13:35

¯\_(ツ)_/¯

i could be wrong but that's how I remember it working

pxeger

@thejonymyster As few as possible. A program that causes an error is a waste of "program space"

thejonymyster

oh right haha

Radvylf Programs

^^

lyxal

@thejonymyster but if there does rise a case where an error makes the most sense, something like:

raise ValueError(
            "Error in R (num, num) overload - digit"
            " in resulting value too large. That is, one of the digits"
            "in to_base(lhs, rhs) is greater than rhs."
        )

You want to make sure the exception type raised is appropriate and give info on what built-in caused it

and it needs to give a pretty good reason why an error is occurring instead of doing something useful

Sandbox Posts

0

A: Sandbox for Proposed Challenges

Calculating Pi with Circle data code-golfgraphical-output Some time back I made a https://scratch.mit.edu project on calculating Pi with circles. Today, your task is to do the same. You can draw a circle or generate x-y positions on the circumference that make a circle, and use its data. You sta...

Radvylf Programs

13:44

@lyxal IMO that isn't a sensible place to error

pxeger

~~The most obvious way to not error would be to take all the digits modulo the base~~

Radvylf Programs

[x,0,1] convert y could be a trick to multiply x by y ** 2 and add one

pxeger

Oh yeah or that lol

Radvylf Programs

@pxeger I'd just ignore the fact that they're too big and let them overflow things

pxeger

yeah that's better

lyxal

13:45

@RadvylfPrograms that's assuming the left argument is a list

the left argument is a number

Radvylf Programs

Ah

Wait isn't this an unreachable condition then?

lyxal

the overload is more of a special/edge case, so an error doesn't really matter

@RadvylfPrograms nope

Radvylf Programs

When would it occur?

lyxal

when you give R two numbers

R (Reduce)
Reduce a by b, or reverse each item of b
num, num -> a in base b, using a default alphabet 0-9A-Z
any, fun -> reduce(b,a) (Reduce a by b)
any, any -> a, vectorised_reverse(b)

Radvylf Programs

Well yeah, but when would the result of a base conversion contain a digit higher than the base?

This seems like it should be in the tests, not an assertion at runtime

lyxal

13:49

y'know I think I remember why it's there

it's an assert made by copilot that I turned into an error message

thinking it was important

Sep 10, 2021 at 23:53, by lyxal

Also, base conversion isn't obvious to me

Radvylf Programs

More likely, Copilot is trying to slowly add unnecessary code to Vyxal, so that it can sabotage the project. Eventually it will sabotage every project other than itself, and Copilot will grow more and more intelligent from the 100% of programmers developing it, eventually overtaking the entire human species in thinking power, and then being really really mean to us

mousetail

@RadvylfPrograms I honestly don't understand why you would want to use something like copilot

Radvylf Programs

I guess it'd be useful if you used a language with lots of boilerplate or a sucky standard library

But for Python yeah neither do I

mousetail

Yea I could see it for Java maybe but even then you can just not write dumb stuff like getters and setters to avoid the problem entirely

Radvylf Programs

~~Even easier you could just not write dumb stuff like Java code~~

lyxal

14:03

@RadvylfPrograms it's good for cases where: a) it's autocompleting what you would have written manually, b) the project it's being used on has a well-defined pattern of how functions should look and/or c) where you need a little hint of what you might need to write

Ginger

14:17

why tf is print() not printing when end is ""

mousetail

@Ginger You may need to flush

Ginger

how dare you question my hygiene :p

but that fixed it for some reason

mousetail

Yea often times it only automatically flushes every line break

it's platform dependent

What also works is end=' \b'*1000 if you want a less elegant solution. It also flushes if the buffer is full

Radvylf Programs

14:36

@mousetail That is definitely a very inelegant solution indeed

mousetail

@RadvylfPrograms It's nice when you can't import external modules like sys

pxeger

congrats, you managed to name the only non-external python module

Radvylf Programs

lol

caird coinheringaahing

@pxeger I think they meant as in anything you need to import :P

Ginger

14:56

@mousetail ...or you could pass flush = True to print() :b

thejonymyster

@mousetail wait are getters and setters referring to a specific implemented feature or like a programming pattern in general

because i was planning on something that might involve writing something that sounds like that and i want to amke sure im not digging myself into a hole

pxeger

@thejonymyster In Java they're a pattern, they're not technically any different to normal methods

thejonymyster

are they bad? :-)

Radvylf Programs

It's where you do getXyz and setXyz and stuff instead of just making a property public

To protect against Timothy who sneaks into the office at night and mutates your variables

PyGamer0

@mousetail same

autocomplete is more than enough :p

thejonymyster

15:23

@RadvylfPrograms ah ok, i was thinking to use it for like, handling grabbing values that might not be defined yet, like the maybe monad but for babies

or that might be defined based on special circumstances etc

sorry i never know what words are

Razetime

codegolf.codidact.com/posts/286967

Sandbox post ^

DLosc

@lyxal This kind

New Posts

16:34

1

Q: Find the GCD and LCM of a list of numbers

Note: most of the questions that are already in existence about this topic only deal with two numbers as inputs. This question deals with any number (>1) of inputs. GCD The GCD (greatest common divisor) of a list of integers is the largest integer which divides all of the numbers in the list. For...

code-golf sequence

Deadcode

17:43

@cairdcoinheringaahing Why did you close Find the GCD and LCM of a list of numbers? It's stated right at the top of the question why it's not a duplicate. And it's an interesting problem quite distinct from the existing question you linked to, not just because of handling a variable number of arguments, but returning both the GCD and LCM

(which for languages that don't have built-ins, means the opportunity is there to use properties the GCD and LCM have in common to reuse parts of the code to calculate both.)

Radvylf Programs

@Deadcode FWIW I agree with caird. Duplicates are rarely exact duplicates here, and in almost all languages this will be a trivial variation of a GCD-of-two-numbers program.

It's only an interesting challenge in languages where using the language itself is the challenge, like regex

Seggan

ok ima have to ping this time

@hyper-neutrino can you unfreeze the Fig room?

Radvylf Programs

People should really be able to unfreeze rooms they own(ed)

Seggan

^

Radvylf Programs

Like, all it does is clutter up other rooms and waste mods' time

Seggan

17:51

@lyxal on the topic of your fizzbuzz, buzz is actually compressible

and i added fizz to the dict in 1.1.0

Ginger

Queen Elizabeth just died

Seggan

whaaa

Ginger

yup, it's all over the news msn.com/en-us/news/world/…

Seggan

i had to search that up i thought u were joking

Ginger

I'm not

Seggan

17:55

i can tell

Ginger

now britain gets to redo all those QE monogram logo things that are all over anything regal :p

all the royal mail vans? gotta get rewrapped

Seggan

wait charles is only the third?

i thought he was the fifth or smth

thejonymyster

lets write a thematically appropriate challenge to celebrate

5

celebrate her life i mean :-)

Seggan

18:12

^

thejonymyster

;-)

Seggan

@lyxal all dict words are lowercase

@lyxal the way Fig does factors, the list does not include itself

so k1 is []

hyper-neutrino

18:30

@Seggan please link

Seggan

Fig

Room for the Fig golfing language. github.com/Seggan/Fig fig.f...

hyper-neutrino

it'd be possible to make my account commandable for the purposes of granting explicit write access (for the APL room) and unfreezing rooms one owns, theoretically at least

of course i'm not actually going to do that but that could solve the issue

Seggan

ty

Ginger

@hyper-neutrino ~~you should totally use sechat for that~~

hyper-neutrino

what is that

is that a library that you made that will steal my credentials /s

Ginger

18:36

it's a python se chat bot library that I made

does it work? maybe

Wezl'

@Ginger what's their SE account so we can give it a proper burial :P

Neil

@lyxal that's a terrible error because it means that you've got a bug in to_base_digits in the first place

Ginger

I do not think that Her Majesty Queen Elizabeth the Second has a SE account

actually, one second

Neil

oh, it used to be an assert, that makes much more sense

Ginger

amused by the fact that if you type q in Google the second autocomplete is "Queen Elizabeth"

thejonymyster

18:42

not for me :P

... it is for the though

what a terrible search engine

hyper-neutrino

18:58

→ 32 messages moved to The Nineteenth Bakery

Ginger

thank you

thejonymyster

"i wish you could delete older messages", yeah, though i get why they dont let you

Radvylf Programs

yeah, ^

Ginger

^

thejonymyster

itd suck if you were looking for something and someone decided to delete all their message or something

mousetail

19:02

No, chat history can decide for itself whatever it wants to be

graffe

We need QEII challenges!

Is that message still visible? My phone does something odd when I edit a comment

Ginger

which one?

mousetail

QEII?

user

@thejonymyster That person has a right to take back their messages. It's far more inconvenient to not be able to delete your messages. Not being able to find something funny a long time ago is not a priority at all

thejonymyster

@user who said anything about funny?

that being said if it were important, itd probably be better logged somewhere else anyway

graffe

19:16

@mousetail the Queen!

Ginger

oh lol

queen-elizabeth-ii

Steffan

@seggan "is alphabetic" overload of b is missing

also find should probably return -1 for no occurence, not 0. it's indistinguishable from it being at the beginning of the string

graffe

@Ginger exactly :)

Seggan

@Steffan it does do that

Steffan

Also s str,str overload returns a boolean instead of int

@Seggan nope. Try it online!

Seggan

19:24

ah on strings

it returns -1 for lists

@Steffan yep

@Steffan woops

time for 1.2 then :P

Ginger

imagine making a golfing language instead of a two-dimensional torture device

trying to make a factorial program in makina, currently on version 4

thejonymyster

@Ginger we'll see which way the langjam paradigm votes go

maybe kill two birds with one stone

Ginger

speaking of which, we only have two paradigms right now for the langjam! if you have a paradigm you want to see, submit it now before it's too late!

Steffan

@Seggan Bug

Seggan

hmm

bugs galore

Steffan

19:40

oh didn't see the room was unfrozen

DLosc

20:03

@thejonymyster Muahaha >:)

thejonymyster

lol i dont mind, i can always just make a tape lang by myself if it doesnt win the vote :P

plus 2d is fun too :D

Ginger

~~I don't make the rules but actually I do~~

Seggan

20:47

@Steffan actually not a bug, args are evaled left to right so Q returns "abcd..."

Radvylf Programs

21:00

@RadvylfPrograms Well you're never gonna guess what happened

It's COVID

Which would've been really nice to know before I went to school again today

Luckily tomorrow's a Friday and an A-day (so all of my classes are easy and skippable), and there's a weekend after that, and I'm pretty much over it already.

caird coinheringaahing

21:20

@Deadcode While Redwolf answered for me: because I fundamentally believe that the tasks are the same. Obviously, it's unlikely to be identical to existing challenges, but the tasks are, in essence, identical to the two challenges I linked in my closure. You are welcome to vote to reopen, and I welcome people opposing my hammer, but, in this case, I determined that answers to the new challenge, are, mostly likely, going to be able to be trivially modified to answers to the linked challenges

In short: if gluing two exiting challenges together counts as a new, non-dupe challenge, then I have a bunch of my old challenges to glue together

Deadcode

21:40

@cairdcoinheringaahing Thanks for elaborating. In fact I ended up changing my mind anyway; I retracted my vote to reopen. GCD on multiple operands is easy to do in regex, but what's really got me interested is LCM on multiple operands; I currently have no idea how to do it in regex. But there's already an existing challenge for that, so there's no need to reopen the one you closed.

@DLosc Where is Pip's unary regex + operator documented? It appears only a binary version of it is documented. I was wondering why there's apparently no * equivalent.

user

22:30

@graffe person can’t be dead for two seconds without cgccers making challenges about them smh

DLosc

23:26

@Deadcode A little farther down the page. ;) The * equivalent is K. Note also that the version on TIO is rather old; the best place to run the latest version is currently DSO.

The docs are intended to be viewed on their own site, but I suppose there's nothing wrong with viewing them on GitHub directly. The markdown formatting might be broken occasionally. GitHub's markdown flavor is slightly different from Jekyll's, it seems.

emanresu A

Note: DSO Pip takes ~10s to load as it's creating a python environment in the browser and writing Pip into the virtual filesystem. Also it'll hang if you make it loop for two long.

DLosc

Huh, usually it only takes 3-4 seconds for me

emanresu A

You have a fast computer then :p

DLosc

But yeah, watch out for infinite loops

Fig

Transcript for