The Nineteenth Byte

Ginger Industries

15:00

@mathcat this is arguably better then the first one

mathcat

"RICKROWS ARE HONEST AND SAFE"

@cairdcoinheringaahing if GPT says it, then no need to ban rickrolls

Ginger Industries

"A good rickroll is a new person with a completely different set of ideas."

true enough

null

1

A: Covfefify a string

!@#$%^&*()_+, 102 101 bytes pgtevkhijglmnobqrzdufwxys(*!@a_+&)(*!!@a_+&_^)$H_+&%(*!`_+&)+&@!@&@@ 42nd answer of this question, it looks like. Beats JavaScript for some reason. It took much more time than it should. I'm pretty sure there are still some simple tricks that I'...

Someone golf this to <99 bytes

New Posts

0

Q: Tips for golfing in !@#$%&()_+

What general tips do you have for golfing in !@#$%&()+? I'm looking for ideas that can be applied to code golf problems in general that are at least somewhat specific to !@#$%&()+ (e.g. "remove comments" is not an answer).

Fmbalbuena

@PyGamer0 @GingerIndustries Don't flag my message, please.

@thejonymyster sorry you can't get 2 rep by minor edits.

if you have 2.5k rep then you can do minor edits

Kamila Szewczyk

15:19

@Fmbalbuena spam is encouraged to be reported

pxeger

@Fmbalbuena Your message was rightfully flagged because it was spammy. Please don't post spam.

Kamila Szewczyk

oh, also. i've been working on a custom esolang that could theoretically be used for golfing

i want to keep the instruction set relatively clean and low level, so i'm open to feedback.

state-of-art documentation: esolangs.org/wiki/Dirac

pxeger

is it implemented in malbolge? ;)

Kamila Szewczyk

in C :)

but it compiles on DOS and other obscure systems/machines somehow.

i don't want to add useless/bloaty builtins, but i'd love to make the language at least a bit nicer to program.

i find it hard to keep track of everything on the stack correctly for some reason

especially when there's >7 elements and they're all something else. and functions need them all in the right order too.

PyGamer0

@Fmbalbuena i didn't

Fmbalbuena

15:33

@PyGamer0 Hey, i have a question: How many golflangs do you have?

PyGamer0

@Fmbalbuena 2

dinoux and flax

Fmbalbuena

dinoux github?

mathcat

@PyGamer0 05BA something?

pxeger

05AB1E?

That's not made by PyGamer0

mathcat

no?

oh, ok

Wheat Wizard

15:39

Let's say I have n slots and m of them have something in them. I want to add x items to the slots. Each item must go to it's own slot and at least 1 item must go to a slot that already has something in it. I don't care about the order in which they are added. How many ways are there to do this?

I think it's (n choose x) - ((n-m) choose x), but I'm not certain.

pxeger

assuming x < n and n > 0 and m > 0?

@WheatWizard m * ((n-1) choose (x-1))?

Wheat Wizard

n > 0: yes; x < m: no

pxeger

@WheatWizard sorry, I got my n and m mixed up

Wheat Wizard

x is less than n.

@pxeger This doesn't work, because it double counts them. Imagine that n=m and x is 2.

Fmbalbuena

Can someone help me to write my golflang?

pxeger

15:43

each slot can only have one of the x items?

Wheat Wizard

Yes.

pxeger

Are the x items identical, or is there a distinction between (for example) [x1 m1] [x2] [ ] and [x2 m1] [x1] [ ]?

Wheat Wizard

The items are identical.

pxeger

where xi is a new item, and mi is an item already there

ok

Wheat Wizard

Well the new items are distinct from the initial items I guess. We care about ways to add the items, not the final result.

pxeger

15:46

so we only care about which subset of slots has items added to them

Wheat Wizard

We care about the set of slots we have added items yes.

@WheatWizard Or even more concretely n=m=3 and x=2. In this case you can put the items in any slots so there are 3 ways to do it. But your formula gives 6 ways. 3*(2 choose 1) = 3*2 = 6.

Fmbalbuena

@WheatWizard How many golfing languages do you have?

Wheat Wizard

My thought was, count the total number of ways to do it regardless of what is already there (n choose x) and subtract the ways to add them so no item is put in an already occupied slot ((n-m) choose x). But I'm not super sure this works.

@Fmbalbuena Zero.

pxeger

@WheatWizard \sum _{i=1} ^{min(x,m)} ((m C i) * ((n-m) C (x-i))? Or should (n-m) be (n-i)?

Wheat Wizard

Maybe ...

Hard to tell.

Fmbalbuena

15:51

@WheatWizard how?

pxeger

It can be simplified

@WheatWizard hgl? I guess that's not really a language

Wheat Wizard

@Fmbalbuena Golfing languages are a lot of work.

@pxeger Ok sure. I have ~0.75 golfing languages.

Fmbalbuena

@WheatWizard Ok, how many languages do you have?

Wheat Wizard

Depends on how you count for sure.

At least 5.

Fmbalbuena

github links?

Wheat Wizard

15:55

I don't use github.

pxeger

gitlab.com/WheatWizard

Wheat Wizard

esolangs.org/wiki/User:Wheatwizard

Fmbalbuena

@WheatWizard gitlab?

@WheatWizard github.com/Wheatwizard/Unbalanced ???

Wheat Wizard

I used to use github. It will have some projects on it probably.

About half I think.

pxeger

Is there a reason you stopped using it?

Wheat Wizard

15:58

Microsoft.

I also like gitlab better anyways.

I was pretty happy when I made the switch.

I still use gitlab enterprise for work, and I find it a bit of a pain.

pxeger

I prefer GitHub's UI to GitLab's, but in general I like GitLab more

the problem is just the network effect of everything being on GitHub

there's a lot of friction when using both at the same time

Wheat Wizard

I generally don't have that issue for whatever reason.

I guess I mostly use git to work with my friends or on my own and do less of the forking / working with strangers stuff.

Fmbalbuena

@WheatWizard Can BF transpiled to Unbalanced

Wheat Wizard

I hardly remember anything about unbalanced.

mathcat

I do

because I read the docs

Wheat Wizard

16:03

Ok I freshened up a bit. I don't think it's easy to transpile bf into unbalanced.

It's probably more like compile.

Fmbalbuena

@WheatWizard You like making programming languages?

thejonymyster

@Fmbalbuena no worries, i dont care about +2 reps, i just dont wanna clutter the comments with "hey make sure to escape this special character" or "change these letters to this"

Fmbalbuena

@thejonymyster but you must do this, please don't do minor edits.

thejonymyster

once again wishing edits were not tied to rep so this wouldnt be a problem

Fmbalbuena

@thejonymyster Do you like 1k rep?

mathcat

16:17

Reputation is not like a prize or a badge, it's just "a rough measurement of trust". It's their choice if they want to do minor edits or not.

Fmbalbuena

@mathcat ... i know

user

16:32

> Sure, JS beats you up every day, but it loves you.

pxeger

I have something exceedingly important to share

bigrat.monster

user

Is that the entire site?

Just that one big rat?

pxeger

big rat

hyper-neutrino

rat

New Posts

0

Q: Split along groups (hard mode)

This is an alternate version of this earlier challenge with a twist that adds a significant bit of difficulty. Like last time, you are going to be given a string containing some alphabetic characters along with [ and ]. Your task is to split into sections that are enclosed in a "group" created b...

code-golf string

Fmbalbuena

16:46

import sys
code_page = "|¬°!1\"2#3$4%5&6/7(8)9=0?'\\¿¡\tq@wertyuiopasdfghjklñzxcvbnmQWERTYUIOPASDFGHJKLÑZXCVBNM¨´+*~{}^`[],.-_:;<>éýúíóáÉÝÚÍÓÁèùìòàÈÙÌÒÀêûî"
code_page += "ôâÊÛÎÔÂ\nëÿüöïäËÜÖÏÄᐍᐎᐓᐗᎾᚁᚂᚃᚄᚅᚆᚇᚈᚉᚊᚋᚌᚍᚎᚏᏔᏕᏖᏗᏘᏙᏚᏛᏜᏝᏞᏟᏠᏡᏢᏣᏤᏥᏦᏧᏨᏩᏪᏫᏬᏭᏮᏯᏰᏱᏲᏳᏴᏵᏸᏹᏺᏻᏼᏽᎳᎴᎵᎶᎷᎸᎹᎺᎻᎼᎽᎿᏀᏁᏂᏃᏄᏅᏆᏇᏈᏉᏊᏋᏌᏍᏎᏏᏐᏑᏒᏓ×ĀāĂăĄąĆćĈĉĊċČčĎaĐ"
i=""
with open(sys.argv[1], 'rb') as file:
    code = file.read()
    code = "".join(code_page[i] for i in code)
# Golfing language interpreter
lines = code.split("\n")
arguments = sys.argv[2:]
dyadiccommands = {

^ sorry for long text, is this correct?

hyper-neutrino

correct for what...?

Fmbalbuena

for making a golflang

hyper-neutrino

i mean assuming you only want to be able to read the code in raw bytes then it's correct

it's correct if it does what you want

thejonymyster

17:04

@mathcat reputation is a currency for purchasing bounties

hyper-neutrino

is it cursed to do exports.fn = function() {...} and then in another part of that file call exports.fn(...)

pxeger

exports.fn = fn = function() { ... } fn(...)

hyper-neutrino

oh yeah

DLosc

17:28

Would function fn() { ... } exports.fn = fn also work?

pxeger

I think so

user

I like the first even if it's a little more cursed

Not a fan of separating the definition and the export

pxeger

The first one requires a const though

user

Oh :(

Bgil Midol

Is there multiline code support in chat?

user

17:41

Yes, hit fixed font or indent with 4 spaces manually

DLosc

Specifically, you first have to paste a multiline string or add newlines manually using Shift-Enter. Then the "fixed font" option will show up.

pxeger

you can also press Ctrl+K, which doesn't require you to have entered multiple lines

Bgil Midol

class Token:
    def __init__(self, name, value):
        self.name = name
        self.value = value

like ^?

DLosc

@pxeger TIL

I love how everything on the starboard in Sandbox is from TNB people

caird coinheringaahing

Anyone can star anything there :P

Fmbalbuena

17:59

@hyper-neutrino no, is my command good?

hyper-neutrino

point still stands that if it does what you intended then it's probably good

Fmbalbuena

@hyper-neutrino I have to do random challenge, find the most used command, add?

@user254217 Welcome to TNB

emanresu A

@cairdcoinheringaahing Aah, I see you've learned from me

ta Radvylf srik su shilani

@Fmbalbuena You've flagged to close a lot of stuff recently, and most of it has not been stuff that should actually be closed

Fmbalbuena

18:17

@taRadvylfsriksushilani wdym

ta Radvylf srik su shilani

I count at least three posts in the last couple of days that you've flagged as duplicates, and none of them actually were.

Fmbalbuena

But I flag to close off-topic more than dupe

ta Radvylf srik su shilani

I can't see who flagged things as off-topic, so I don't know your stats on those. When you VTC (or flag-VTC) as a dupe, it leaves a comment though.

DLosc

Reading over this answer again, I'm trying to remember if we have a clear definition of "the size [a function] contribute[s] to the program when added." For example, why does lambda x:x+2 in Python count as 12 bytes when you have to either assign it to a variable or parenthesize it to call it directly (+2 bytes either way)? I suspect this has been discussed already; does anyone remember where?

ta Radvylf srik su shilani

All that flagging as off-topic does is send it to the close vote queue, and usually those questions will already be there if they're new.

user

18:21

Well, the function itself is just that, even if there's extra boilerplate required to use it

ta Radvylf srik su shilani

Yeah. Like arrow functions in JS, x=>x+1 is quite often useless without parentheses and/or an f=, but it's still allowed since x=>x+1 is the function, not (x=>x+1)

user

It's like in Java where x->x is perfectly fine as a lambda even though you need java.util.Function<Integer,Integer>f=x->x; to really do it

Fmbalbuena

@taRadvylfsriksushilani I have 16 helpful flags (including duplicates)

DLosc

I'm thinking about HBL specifically, where +.2 is a function that adds 2 to its argument IF it's on a line by itself. I think the Python example implies that the newline doesn't have to be counted as part of the function's bytecount, even though +.2 parses as something completely different without a newline (or '(...), which is even longer).

ta Radvylf srik su shilani

I think you'd be allowed to count +.2 as the function

Unless you consider the newline to be sort of like the word function or lambda

Fmbalbuena

18:24

@DLosc How to make a golflang?

ta Radvylf srik su shilani

I think our function rules kind of necessarily need to be interpreted in language-specific ways sometimes

user

I think we should be as lax as possible when making rules. Code golf's about having fun, and having lots of strict rules doesn't help that

DLosc

Potentially complicating matters, +.2 by itself (with no newline) is both a function and a full program. But until someone implements a half-byte file system, WW's answer implies that it can't be counted as 1.5 bytes if it's a full program, only if it's a function.

ta Radvylf srik su shilani

I think we should have a consistent handling of functions and full programs for scoring

DLosc

@taRadvylfsriksushilani It seems that way, yeah

ta Radvylf srik su shilani

18:27

Either fractional bytes for both, or rounding for both

pxeger

@Fmbalbuena You also comment on a lot of sandbox posts saying they're off-topic, but they almost all were on-topic

ta Radvylf srik su shilani

Personally, I support just allowing fractional byte counts for both functions and full programs, and that's it.

user

^

emanresu A

^

user

Just makes life easier for everyone

Wheat Wizard

18:27

One of the weirdest things is that when you have an anonymous function and some helper functions you need to count the newline for some reason?

emanresu A

waaat

DLosc

@WheatWizard Which newline?

user

That's a rule?

Wheat Wizard

That is how everyone scores their anonymous functions.

pxeger

like this:

user

18:28

What about languages where you also need other stuff between the anonymous function and helper function? (probably most languages) (sorry for interrupting you pxeger)

pxeger

lambda:solve the challenge using helper
helper=lambda:do something else

ta Radvylf srik su shilani

@user I think you'd have to count that too, right?

user

¯\_(ツ)_/¯ I don't think the newline should be counted

pxeger

it makes sense to me: you can prepend f=\ to this submission, and now it works

ta Radvylf srik su shilani

Like, in JS: g=x=>x+1;y=>g(x)-1

Wheat Wizard

18:29

It's certainly a weird one.

ta Radvylf srik su shilani

Why wouldn't you have to count the ;?

It makes zero sense to allow splitting it into multiple functions and not count the separators IMO

Wheat Wizard

because the separators aren't actually a part of the code.

emanresu A

I had a JS polyglot looking something like the following:

ta Radvylf srik su shilani

@WheatWizard Wait how?

user

I feel like an exception should be made for helpers and anonymous functions

ta Radvylf srik su shilani

18:31

; is definitely a part of the JS code

Wheat Wizard

That "code" doesn't do anything.

user

But so is the variable you're assigning the lambda to, and we don't count that, so why count the ;?

emanresu A

1//2; """
x=>x+1
//""";lambda a:a+2

Fmbalbuena

@emanresuA "Wait how?"

hyper-neutrino

this is proof that proton is superior

x=>g(x,x)g=(a,b)=>a+b

ta Radvylf srik su shilani

18:31

@WheatWizard It separates the two functions. By that logic, most of the code I write does nothing.

Fmbalbuena

@emanresuA I don't understand why?

emanresu A

The central function is impossible to access.

user

@hyper-neutrino Oh no, is this Lua-inspired?

hyper-neutrino

@user no i just didn't see a point in making it error if two expressions are back-to-back

Wheat Wizard

@taRadvylfsriksushilani Why does it separate them?

caird coinheringaahing

18:32

@hyper-neutrino Lovely

Wheat Wizard

What is the function of the separator?

pxeger

Here's a single-function answer: ato.pxeger.com/…

If I need a helper, I can define it by appending directly to that program, with a newline between them: ato.pxeger.com/…

ta Radvylf srik su shilani

@WheatWizard To make the functions syntactically valid? If you could do it Proton-style and unambiguously put the two functions right next to each other, no separator would be needed.

Wheat Wizard

In Haskell in fact most of my "programs" are not syntactically valid despite the separator.

ta Radvylf srik su shilani

Then aren't those invalid?

Wheat Wizard

18:33

No why would they be?

pxeger

haskell answers are often of the form

ta Radvylf srik su shilani

Because they're...not valid Haskell? And it's a Haskell answer?

pxeger

%1
x%y=x+y

Wheat Wizard

The JS submission doesn't even do anything with the function. it just evaporates instantly.

user

@taRadvylfsriksushilani No it's not valid Haskell as a full program, but you can wrap it in stuff to make it okay

ta Radvylf srik su shilani

18:34

@WheatWizard No, it evaluates to the function.

So you could do f=(g=...,x=>g(x))

Wheat Wizard

@taRadvylfsriksushilani It doesn't need a separator to do that.

ta Radvylf srik su shilani

Explain how then

Wheat Wizard

The point of an anonymous function is that you can call it from anywhere. You can embed it in legal code like any other function.

ta Radvylf srik su shilani

Yep, and with g=...,x=>g(x) you can

It just needs parentheses

user

@pxeger tbh it feels kinda like cheating to do %1 and not (%1) because you need it to turn it into a real function (right?)

Wheat Wizard

18:36

@taRadvylfsriksushilani But you just moved the function. Which proves that it's not connected to the helpers.

ta Radvylf srik su shilani

I moved what now?

Wheat Wizard

You put parens around it.

ta Radvylf srik su shilani

Parens around the whole thing, yeah, but you need that even for normal JS function submissions

Wheat Wizard

Noone scores their lambdas as (lambda x:...)

If I need to count a separator why don't I need to count a dummy main declaration? it's also required to make it a syntactically valid.

The idea is that the function is your submission. But you also need some other code for your function to work.

ta Radvylf srik su shilani

@WheatWizard Because the main declaration isn't part of your code. The separator is, because the things it's separating are part of yuor cpde

emanresu A

18:38

I'm confused which side either of you are on

Wheat Wizard

Why the two need a weird extra byte despite the fact that neither the boilerplate nor the function need it?

Really I think the consistent way to score this would be that all function declarations need a trailing newline.

ta Radvylf srik su shilani

That sounds like the worst of both

Wheat Wizard

Because if I write f x=2+x then I write more code I need the newline, I can't just add more stuff on.

ta Radvylf srik su shilani

I'm actually not opposed to not counting the separator, after hearing y'all's arguments

DLosc

@taRadvylfsriksushilani O_o

Wheat Wizard

18:40

If functions required an extra trailing newline then you could append two function definitions and it would be valid.

ta Radvylf srik su shilani

@DLosc Isn't this how a discussion is supposed to work? :p

Wheat Wizard

but right now you can have a submission f x=2+x and you want to add new code to the end then you need an extra separator.

DLosc

@taRadvylfsriksushilani Certainly; but I was just saying to myself "I agree with Redwolf's position" and then you changed your position, so I was rather flummoxed.

emanresu A

4 mins ago, by emanresu A

I'm confused which side either of you are on

caird coinheringaahing

"I'm playing both sides so that I always come out on top" - Redwolf

4

ta Radvylf srik su shilani

18:45

No, I'm just not too stubborn to change my mind when the other side has good points:p

user

Be like Redwolf, y'all

Well, other than having cats and stuff

DLosc

Okay, so my thoughts: A simple submission is one piece of code, whether that's a full program or a function. It seems straightforward to score a simple submission: just count the bytes. The difficulty arises with complex submissions, which consist of multiple things: either multiple functions, or multiple files (like Retina used to use), or code + flags.

This used to be pretty straightforward to score as well: 9 times out of 10, the score was the sum of the bytes of each piece, plus 1 byte for each piece after the first. But then free flags muddied those waters.

user

The "plus 1 byte for each piece after the first" feels a little arbitrary. Why 1 byte? Some languages don't need separators. Some need more than 1 byte for the separator(s)

Wheat Wizard

Stuff like separators between files, or flags don't really make sense.

I think that we like to be "objective" but there is a lot of stuff that is secretly subjective or kind of arbitrary. I don't think that's wrong, at some level it is needed, but I feel like there might be some areas where we pretend or believe we are more objective than we are.

user

If you have a helper defined like g=... and an anonymous function x=>g(x) and your full program looks like g=...;print((x=>g(x))(1)), then why not also count the print((? Why only the ;?

DLosc

18:52

@user It does, and that wasn't exactly the rule. But if your language doesn't need separators, then you've just got one piece of code, see the simple submission rules. If it does, then I think having multiple pieces needs to be considered in the scoring somehow.

user

@WheatWizard I can't tell if "but I feel like there might be some areas where we pretend or believe we are more objective than we are." is just about CGCC or life in general :P

Wheat Wizard

I think in general one thing that would be nice, and I think we have taken steps towards this, is to focus on making it fun rather than making it a perfect competition.

user

@DLosc Is there any real reason to penalize having multiple pieces, though?

caird coinheringaahing

@WheatWizard Can't agree more

user

How about a compromise: We have a "helper section" where you have as many helpers as you want and you count the separators between those and another section for the actual submission (the anonymous function). That way, people will have to count separators between helpers, but not have to add an arbitrary separator between a helper and an anonymous function

31 mins ago, by user

I think we should be as lax as possible when making rules. Code golf's about having fun, and having lots of strict rules doesn't help that

DLosc

18:56

@user The competitiveness argument would say that languages would go to ridiculous lengths to take advantage of such a rule; see, for example, Retina's original design, which required each line of the program to be in a separate file. Without the penalty for extra files, this would be tantamount to saying that newlines in a Retina program don't contribute to the bytecount.

user

No need to make people worry about little things. It's just 1 byte

Wheat Wizard

I think that right now what we have works. It's a little strange and arbitrary, but it seems like it works fine and I would not want to go through the burden of rescoring like 40% of all submissions.

DLosc

But the "don't worry about what other languages do" argument says that doesn't matter, so...

caird coinheringaahing

Personally, I've always considered "You can claim a score of X if you can write a full program or function in a language in X bytes (under some encoding) such that it does what the challenge specified" to be the fundamental rule of the site

Bgil Midol

How to update a github repo file?

user

18:58

It's straightforward enough when you have a single function, but multiple functions that aren't full programs are hard :|

DLosc

^

emanresu A

What if a newline counts as half a byte in your language?

Wheat Wizard

Then probably you only score them half a byte.

caird coinheringaahing

@user I've generally considered it to be a bit of an extra cost of needing extra functions. Like, if you can't do it in one function, then the extra separator bytes is the "cost" of that

user

@DLosc Okay, I guess I see why that would be bad, but what do you think of the compromise I suggested above?

pxeger

19:00

What if a newline counts as 10^-9 bytes in my language?

4

user

@cairdcoinheringaahing Oftentimes, having to write def g(): is enough cost

Bgil Midol

@pxeger fun fact: print(int(10**-9)) gives zero in python

caird coinheringaahing

In my language, for all real ɛ > 0, a newline costs less than ɛ bytes

user

Is that just a fancy way of saying it costs 0 bytes? :P

caird coinheringaahing

It's arbitrarily close to 0 bytes :P

Wheat Wizard

19:02

It's just an arbitrary rule. What it does have going for it is that it seems like it's pretty intuitive. This is the first time we've had a discussion about it, I've never seen newbies confused etc. A lot of these other proposals while maybe less arbitrary are definitely more complex.

emanresu A

What if my program is 1.9999999999... bytes long?

pxeger

1.99999... = 2

caird coinheringaahing

Then its 2 bytes long

user

@emanresuA I would love to see a computer that can handle fractional bits

emanresu A

I would love to see a computer that can handle rational bytes

Wheat Wizard

19:03

Any computer that uses trits.

emanresu A

Malbolge go brr

user

Remind me to never learn Malbolge then

pxeger

@WheatWizard I think the current rule, while maybe not technically as fair as it could possibly be, is intuitive enough, simple enough, and close enough to fair that it doesn't matter and isn't worth changing the score of all past answers for. It's ok not to be perfect sometimes if the cost of perfection is too high

Bgil Midol

@user do not learn malbolge

emanresu A

> learn Malbolge

Bgil Midol

19:05

@emanresuA How do you format those dots?

emanresu A

> quoted stuff

DLosc

@user I agree for the most part (e.g. I'm very much in favor of flexible I/O formats); but the counterargument, and the reason why we have such exacting rules in the first place, is that some people who enjoy golfing 1 byte off a program also enjoy finding loopholes in rules, but other people don't find constant loophole exploitation funny.

Bgil Midol

> like this?

emanresu A

Yep

Bgil Midol

8 mins ago, by Bgil Midol

How to update a github repo file?

user

19:07

@DLosc Fair enough, pretend I said "as laz as possible while preventing loopholes" or something else more lawyer-y :P

Wheat Wizard

@BgilMidol What is a "repo file"? I've never heard this term in my life.

Bgil Midol

@WheatWizard a file in a repo

Wheat Wizard

Using git.

Bgil Midol

I want to update all files in a repo

Wheat Wizard

Ok, what does "update" mean here?

DLosc

19:13

@user This conversation reminds me of an essay I read once about how online communities often start with the idea of being a place for fun and self-expression without lots of burdensome rules, but eventually they all either realize they need rules, or they degenerate into a morass of trolling and flamewars that doesn't allow for fun or self-expression.

user

:(

Wheat Wizard

I feel like we are on the downward arc of that trend.

user

Time to send all the trolls to the moon so we can have fun and express ourselves

Bgil Midol

@WheatWizard I uploaded a python file to github, and I made changes in pycharm. How to can sync the pycharm with the github? (Or set all files in a github to the ones in pycharm)

Wheat Wizard

We ended up with too many burdensome rules and now we have to smooth out all the wrinkles.

@BgilMidol Using git?

DLosc

19:15

We all agree that we need rules, we're just trying to figure out where to set the needle on the strictness of those rules.

Bgil Midol

@WheatWizard yes

user

@BgilMidol You do git add <file>, then git commit -m "commit message"

Wheat Wizard

You need to push too.

user

On subsequent commits, you can do git commit -am "commit message". And to actually send the changes to github, do git push

Wheat Wizard

You should probably look up a git tutorial.

Bgil Midol

19:16

link?

user

@DLosc It's easy, we start off making super strict, execute anyone who doesn't follow them, then remove all rules because we know the people who followed the early rules are okay :P

@BgilMidol google.com

:P

DLosc

@user More specifically, google.com/search?q=git+tutorial ;)

Wheat Wizard

It's better if you just lookup tutorials on your own and just pick the one that fits your style best.

I refuse to believe that either of those links is not a rick roll.

DLosc

@WheatWizard Believe whatever you want

user

You are missing out on the most awesomest links in the history of mankind just because you choose to stay afraid

mathcat

19:44

@BgilMidol I know a great one

bit old, but still awesome

(not a rickroll btw)

Adám

Fwiw, I think we should ban rickrolls.

user

Pulled an SE there, linking to rickroll with the text "rickroll" :P

emanresu A

19:59

Yeah loll

Adám

@user SE?

emanresu A

0

A: What are the hidden features of Stack Exchange?

The rickroll link on the declined flags help page is an actual rickroll

Adám

20:15

Ah, I see. Note that such links are not Rickrolls, because Wikipedia clearly states that

> The meme is a type of bait and switch using a disguised hyperlink that leads to the music video. When victims click on a seemingly unrelated link, the site with the music video loads instead of what was expected, and they have been "Rickrolled".

Fmbalbuena

@Adám QuadR vs QuadS?

Adám

I'm sorry, I don't understand what you're asking. Can you try phrasing it as a whole question?

Fmbalbuena

Which is better golflang? QuadR vs QuadS?

Adám

It depends very much on what you're doing. If the main task involves replacing text, then I'd recommend using QuadR. If the main task involves searching for text, then I'd go with QuadS. It is easy enough to remember which one is better for which task. QuadR — R as in Replace, and QuadS — S as in Search.

Fmbalbuena

QS will replace to nothing?

Adám

20:22

No, it will find all the matches for one or more regexes.

hyper-neutrino

are either turing complete? doesn't sound like it but i've seen weirder

Adám

Yes, they are supersets of Dyalog APL.

user

They can execute APL iirc

Adám

Which means they can also shell out and execute any other language, or for that sake go online…

They are superficially similar to Retina.

@Fmbalbuena Are you familiar with Dyalog APL's ⎕R ("quad-R")and ⎕S ("quad-S")?

hyper-neutrino

oh wait i thought they just called ⎕R / ⎕S

like, i thought they were subsets, not supersets :P

Adám

20:30

Yes, my three golfing languages, QuadR, QuadS, and Stencil, are thin covers for the Dyalog operators ⎕R, ⎕S, and ⌺, all of which are subsets of (and equivalent to) Dyalog APL. Since the languages add a bit of interfacing on top, they are supersets.

hyper-neutrino

oh, okay, that's cool

Adám

All APL operators are equivalent to the whole language because they can execute arbitrary code.

hyper-neutrino

wait, they can?

pxeger

Operators?

hyper-neutrino

oh wait operators, not functions. oh, mb, I thought ⎕R/⎕S were functions lol

pxeger

20:39

like map and reduce?

emanresu A

eval == JavaScript because it can execute arbitrary code

pxeger

but is eval === JavaScript?

TIL = in italics looks the same: = =

Fmbalbuena

@Adám No

@pxeger is a function that takes a function and any number of arguments

Adám

@pxeger Yes.

pxeger

Can they take a string and evaluate it, rather than taking a function, or something?

Adám

20:48

Some can do that, others can take the evaluation function and apply it to a string. E.g. '\d+'⎕S{⍎⍵.Match} will return a list of numbers, by evaluating all digit sequences it can find.

pxeger

Huh

APL is often surprising

Adám

Sure is neat, though:

      '\d+\W\d+'⎕R{⍕⍎⍵.Match}'How much is 2+34 and 45×6?'
How much is 36 and 270?

user

@pxeger I mean, a tilted equal sign would look really weird

pxeger

Well I thought maybe the bars would be offset a bit (although still parallel and horizontal), like:

user

@Adám Wait, all?

pxeger

20:57

  ____
____

user

Including stuff like ∘ and @?

@pxeger Oh yeah, I guess that would work

Adám

@user Yes, as e.g #∘⍎ and ⍎∘⊢ and ⍎@1 etc.

In fact, I once found a bug in @ because I was doing something like this.

user

oic

I thought you could use the operators themselves without ⍎ to do eval

Transcript for