The Nineteenth Byte - 2021-01-20

« first day (3644 days earlier) ← previous day next day → last day (1192 days later) »

2:32 AM

@UnrelatedString hi integer

@pxeger the fun this about it is the chaining part. IT takes more time for the challenge to get hard (and in turn, more interesting)

2:46 AM

For some reason, Javascript thought it'd be a good idea to have two replace functions for strings: one that only makes a single replacement and one that replaces all occurrences. It doesn't help that other languages only have a single replace function that usually acts as the latter.

Redwolf Programs

3:07 AM

@Lyxal Don't ever use replaceAll. It uses some weird iterator thing internally and only works on a few newer versions of browsers. Just use the g flag to make it a global regex

Example?

(of how it should be)

Redwolf Programs

"this is text and it is more text".replace(/is/g, "is not")

Just do /regex/g instead of /regex/

I was never using regex

Redwolf Programs

Oh, you're using .replace with strings?

yes

e.g. url.replace("+", "%2B")

Redwolf Programs

3:10 AM

If it's a fixed string for the first parameter I'd just use a regex, although you can do replaceAll if you don't care about browser compatibility (which none of us do :p)

Note that JS does have a function to encode and decode URIs

@RedwolfPrograms it doesn't do +

nor does it do ), (, [ or ]

the brackets screw up cgcc links

and the + screws up the permalink decoding

Redwolf Programs

user image

wait I'm using encodeURI

there's a different URI encoder?

@RedwolfPrograms still doesn't do brackets

Redwolf Programs

It does for me

and brackets need to be done

Redwolf Programs

3:13 AM

The only thing it doesn't do for me is parenthesis

I think those are valid in URIs

I need it to do parenthesis

@RedwolfPrograms but they muck up [links](example.com?d=()

see

that's what i mean

Redwolf Programs

Maybe just do "".replace(/[^a-z...]/g, c => "%" + somefunctiontoturnthecharacterintoacode)

I'll go check the RFC for the correct values for that

@RedwolfPrograms that also mangles =s

Redwolf Programs

You want those unaffected?

yep

I'm refining the permalink buttons on my vyxal interpreter

Okay it's good now

But why can't I get the clipboard writing stuff working?

Redwolf Programs

3:18 AM

You don't base64 encode the data?

@RedwolfPrograms have you ever tried base64 encoding unicode?

Redwolf Programs

I had no idea that didn't work...what's the point of base64 in JS lol

@RedwolfPrograms base64 doesn't play nice with charpoints > 128

and unfortunately, Vyxal has a lot of unicode

it's a golfing language after all

a golfing language that recently hit it's first major release: Vyxal is now v1.0.0

Redwolf Programs

For the clipboard thing, what part of the code are you running it in? There's lots of security stuff involved with the clipboard.

In the head part

on button click

Redwolf Programs

3:20 AM

Hm, it should allow that if it's by user interaction.

when either share or postify is clicked, it should copy to clipboard

Using navigator.clipboard.writeText(...)

Redwolf Programs

Using inline JS to call the functions is probably messing it up

ah

it says the clipboard part is undefined

funny thing is that it was working once upon a time

Redwolf Programs

Apparently it doesn't want to dotheshare

nor dothetio

well it outputs it to the box but not copy it

the html if you want to see it

note that it's a jinja template, so things like {{value}} get replaced at render time

because I'm using a flask app *dabs aggresively*

Redwolf Programs

3:25 AM

Did y'all use a chicken randomizer for the indentation?

no

it's called two collaborators

who probably use different indentations

Why not base85 (or base-whatever) to convert each Unicode codepoint into ASCII domain, then base64?

Redwolf Programs

~~I just use the indent feature in google docs~~

I just love the look of my JS in crisp times new roman

(or just convert the source code to SBCS and base64 that)

(MDN to the rescue)

> console.log(ok.codePointAt(0).toString(16)); // 61: occupies < 1 byte

Apparently the character a occupies zero bytes

@RedwolfPrograms okay the indentation is fixed

and I'm just gonna stick with url encoding for the permalink

the website is just a front for the real elegant stuff

Redwolf Programs

3:58 AM

Doing a school project on Spectre and Meltdown, this should be fun

I've got a really cool analogy for how the CPU cache works and how Spectre manipulates it

4:32 AM

@RedwolfPrograms Uh... a school project? At this time of year? At this time of day? In this part of the internet? Localised entirely within your computer?

Redwolf Programs

I think the answer to most of those questions is yes but Iḿ not sure tbh

@RedwolfPrograms may i see it?

Redwolf Programs

I haven't done anything on it yet, it's 23:00 so I'm just doing some research

@RedwolfPrograms and you just missed the joke

That's supposed to be a steamed hams reference

Redwolf Programs

I had a feeling I was missing some sort of joke :p

Sorry for ruining it

4:38 AM

all's good

you got it except for the "its 23:00 part"

Redwolf Programs

4:58 AM

Just proposed a String#reverse method in the TC39 (ECMAScript) official forum: es.discourse.group/t/string-reverse-method/620

> given that it's a fairly common operation for both data types

[citation needed]

Redwolf Programs

I mean I've reversed a string more times than an array, and TC39 apparently determined that it's important enough for arrays to have a reverse method

Plus it wouldn't be much of a pitch if I was like hey, what if we could reverse strings...nobody would want it but it would be cool in theory

7 hours later…

12:11 PM

Try it online!

Husk people

Is there any way to make this infer?

2 hours later…

New Sandboxed Posts

2:35 PM

0

A: Sandbox for Proposed Challenges

Write a Brainfuck compiler + decompiler! Many Brainfuck compilers will convert to a different language instead of compiling directly. Your job is to write not only a compiler in language X which compiles Brainfuck to language X, but also a decompiler which takes code compiled with said compiler a...

2 hours later…

4:25 PM

@Razetime Not a Husk person, but what are you trying to do?

caird coinheringaahing

@NewMetaPosts Fun facts: As a site, we got one Steward badge (for VLQ) and four Marshall badges (2093 flags)

4:51 PM

@user It starts with 3, then gets the hypotenuse of that i.e. sqrt(3^3 + 4^2)

then it does that with the next value

and so on infinitely

5:27 PM

@Razetime Where do the ^3 and 4^2 come from?

2 hours later…

7:07 PM

Oh, I see, are you trying to answer this question?

7:26 PM

@Razetime I have no idea how yours works and how to make it work (I can't read Husk), but here's an alternate solution that works for at least 5 elements. Good luck!

Here's a 19 byte version, actually, but it uses that ugly lambda :(.

7:47 PM

Looks like this is the shortest you can get, I tried making it pointfree, but it just got messy.

Actually, there's an 18 byter too!

Okay, I'm going to stop talking to myself now.

8:09 PM

Actually, never mind, that answer's invalid because it times out for the sixth element of the sequence. Sorry, I can't help you.

Redwolf Programs

8:37 PM

Any feedback for Bot Factory KoTH (2.0) (any way I can make it shorter while still making sense)?

9:06 PM

@Razetime Btw, your current code infers and is pretty fast, I got the first 10 elements in about 22 seconds; the only problem is that the last few elements are off. Maybe there's a rounding error somewhere?

@RedwolfPrograms What was wrong with 1.0? It looks fine to me.

caird coinheringaahing

Jan 16 at 20:09, by Redwolf Programs

The problem with just counting the characters collected is that a bot can drop and pick up a character over and over again for near infinite points

Redwolf Programs

I had intended to use the bots' scores as the winning criterion, but ended up using the unmodified number of characters picked up instead

Oh, I see

Redwolf Programs

I'm thinking about adding a way for the owner bots to communicate with their workers, but I'm not sure how to do it without it being overpowered

New Sandboxed Posts

0

A: Sandbox for Proposed Challenges

Print a 3D shape write a program/function to print this cube (or whatever it is) in different sizes: ¶L¶L¶L¶L¶L¶L¶L¶L¶L¶L¶L //¶L¶L¶L¶L¶L¶L¶L¶L¶L¶L¶L ////¶L¶L¶L¶L¶L¶L¶L¶L¶L¶L¶L //////¶L¶L¶L¶L¶L¶L¶L¶L¶L¶L¶L ////////¶L¶L¶L¶L¶L¶L¶L¶L¶L¶L¶L //////////...

9:21 PM

So what incentive is there for bots to drop characters? So their owners/workers can pick them up?

Redwolf Programs

Either that or maybe as a distraction or bait

It's not a major aspect of the challenge, but it's something a lot of people had wanted in my first KoTH (about collecting gold coins)

New Sandboxed Posts

9:44 PM

0

A: Sandbox for Proposed Challenges

Is it a vampire number? Repost and improvement of this challenge from 2011 A vampire number is a positive integer \$v\$ with an even number of digits that can be split into 2 smaller integers \$x, y\$ consisting of the digits of \$v\$ such that \$v = xy\$. For example: $$1260 = 21 \times 60$$ so ...

9:57 PM

Does anybody here know how could I get the Husk room unfrozen?

Redwolf Programs

You can ping a moderator (HyperNeutrino, Jo King, or Wheat Wizard) and ask for it to be unfrozen

I think that's the only way, not sure

@RedwolfPrograms Thanks!

@HyperNeutrino could you please unfreeze the Husk room when you have some time?

10:45 PM

G'morning y'all

@Leo You're one of the Husk people, right? Could you help Razetime out on this?

Unrelated String

@Lyxal is it not normal to agree on indentation on a project by project basis

like, i'

m a 4 space guy, but if i'm working on something 2 or 8 space i'm using 2 or 8 space

Redwolf Programs

8 spaces? Weak, I use 32 :p

@UnrelatedString not when it's a collaboration between two cg users

Unrelated String

@RedwolfPrograms smh using a power of 2

11 spaces for life

10:56 PM

I use fibonacci spacing

3

Redwolf Programs

I indent my program really far and deindent my nested stuff

void foo()
{
 foo { /* 1 space */
   bar { /* 3 spaces (1 + 2) */
      baz { /* 6 spaces (1 + 2 + 3) */
           qux { /* 11 spaces (1 + 2 + 3 + 5) */
                   foobar { /* 19 spaces (1 + 2 + 3 + 5 + 8) */
           }
      }
   }
 }
}

Redwolf Programs

That actually looks good tbh

agreed

Y'all losers with your lame indents

I use 69 and 420 spaces

10:58 PM

original meme haha funny

Unrelated String

to get the best mileage out of fibonacci spacing you should separate the regions of space corresponding to the last two terms with semicolons

It goes 69 spaces, then 420 spaces, then 42069 spaces, then 69420, then 4206969, 6942069, 6969420 etc

flat is better than nested amirite

Redwolf Programs

@Lyxal Add that to the OEIS

would it get the 'nice' tag?

10:59 PM

I don't remember my account details

@rak1507 isn't that Fibonacci-2 spacing?

wdym?

I mean, I know you want to avoid 0, 1, 1 indentation, but you could have used Fibonacci-1 spacing 0, 1, 2, 4, 7, 12, 20

ah right yeah ik what you mean

Redwolf Programs

But that's not as instantly recognizable as 1-2-3-5-8

11:10 PM

@RedwolfPrograms is this a valid output format for your most recent challenge?

Redwolf Programs

I'd say yes

@user I'm not sure what's the algorithm they tried to implement there... It seems to be working, but it becomes extremely slow after a dozen elements or so.

@Razetime we can discuss about this and ḟ in the Husk chat as soon as it gets unfrozen :)

11:41 PM

@Leo The slowness is to be expected - mine and other people's algorithms timed out after just 5 elements. The issue seems to be some kind of floating point error, the last couple terms seem to be off, maybe because of the square rooting/unsquarerooting.

11:55 PM

@RedwolfPrograms sure but 0, 1, 3, 6, 11, 19 isn't recognisable either

Redwolf Programs

Oh never mind, I see what you meant. I thought you were saying those'd be the amounts to indent (rather than the sum)

« first day (3644 days earlier) ← previous day next day → last day (1192 days later) »

Transcript for

Jan '2120

The Nineteenth Byte

The Nineteenth Byte: General discussion for codegolf.stackexc...

join this room
about this room

00:00

06:00

12:00

18:00

all times are UTC

site design / logo © 2024 Stack Exchange Inc; legal

mobile