The Nineteenth Byte

Destructible Lemon

12:00 AM

I think I might try sorting the pixels by different things and seeing what happens

maybe I should make a room for my images

@ATaco would you be interested in seeing more better ones of these?

ATaco

Probably.

totallyhuman

Hmm I need to get familiar with PIL

Destructible Lemon

ok brb making a new room

my all colour images

I will post some all colour images I generated. if anyone else...

Downgoat

Question: in the language of math how would I say 'n is in the domain of function t'

Wheat Wizard

Thats how

totallyhuman

12:35 AM

There's a fancy symbol

musicman523

@totallyhuman I actually could not find it, would you care to enlighten us?

totallyhuman

This one: ∈ I think

musicman523

Well that's element of

I think Downgoat was asking for element of the domain of

I could not find a symbol for domain of

totallyhuman

f(x) = √(x) x ∈ [0, ∞)?

musicman523

12:50 AM

Yes, if you know the domain is [0, ∞) ahead of time. I think what Downgoat was looking for was a symbol that represents the domain of an arbitrary function

Riker

@musicman523 isn't it R?

and lol I thought you were phi

musicman523

@Riker ℝ is the Unicode symbol that represents the set of all real numbers

Downgoat asked for a symbol that represents "domain of T" where T is an arbitrary function

Riker

ye I thought it was also for domains of functions

musicman523

I don't think so?

Helka Homba

anyone for codenames shortly? chat.stackexchange.com/rooms/59120/codenames

musicman523

1:02 AM

That wouldn't make any sense

New Sandboxed Posts

0

A: Sandbox for Proposed Challenges

Find the Second Zero code-golf number Challenge Given an integer in 32-bit two's complement format, return the index of the second zero digit in the binary representation, where an index of 0 represents the least significant bit, and an index of 31 represents the most significant bit. If there...

ATaco

1:41 AM

I wonder which golfing langs have Arbitrary Precision Integer support.

musicman523

I would imagine any that are implemented in python

Which AFAIK is many

ATaco

Poor Japt.

Downgoat

Fun fact: this graph took 2 hours to make >_<

Wheat Wizard

tikz?

Downgoat

yeah

Wheat Wizard

1:50 AM

I've wasted so many hours in tikz

Downgoat

The horizontal arrows on the right are a pain in the ass to do in tikz so just hardcoded some (x,y) offsets for them

ATaco

And I don't understand it at all.

Downgoat

though discovering tikz actually has docs was a huge step

Wheat Wizard

You should try tikz golf. Its pretty crazy

Downgoat

oh my god what that is an actual thing? O_o

13

Q: Tips for golfing in TikZ

TikZ is a Latex package used for drawing precise images in Latex documents. It has a massive specification and a boatload of features that may be useful for graphical-output style questions. What tips do people have for golfing in TikZ? As always, tips should be specific to to TikZ (e.g. "Remo...

holy shit

ATaco

1:52 AM

You'd be amazed what's TC.

Downgoat

I'm not too surprised TikZ is TC because for loops and ifs

Wheat Wizard

@Downgoat codegolf.stackexchange.com/a/110054/56656

You can even take input

Downgoat

though I applaud anyone brave enough to golf in that monstrosity.

ATaco

Best I can do is MathJS for "Things that shouldn't be TC but are"

Downgoat

@WheatWizard how long did that take you to write O_o

Wheat Wizard

1:54 AM

Probably an hour total. I come back to it every so often

Downgoat

CMC: golf this

Wheat Wizard

I would but I'm too busy right now, sorry

Downgoat

@WheatWizard BRB bountying

ATaco

#RemoveWhitespace

Wheat Wizard

@Downgoat Thanks.

Anonymous

2:13 AM

Yet another request for Sandbox feedback

Newly Featured Questions

43

Q: Attack, Decay, Sustain, Release

Sound synthesizers use envelope generators to control how certain parameters of the sound (such as overall loudness) change with time. In many synthesizers an envelope is defined by four parameters, as represented in the following figure from Wikipedia: Attack time (A): time taken for the envel...

code-golf number graphical-output music

ATaco

Fun fact, a % 256 == a & 255, but & is usually much faster.

Same applies for any 2^n

musicman523

Many times compilers will make this optimization if the modulus (?) is a constant power of 2

ATaco

(Although % will return a negative number when passed a negative n, & will not)

Sadly, Lua doesn't do that.

musicman523

Is Lua a compiled or interpreted language?

ATaco

2:19 AM

JIT Compiled.

(Well, some implementations are JIT compiled)

musicman523

Remind me, what's the difference between that and interpreting?

I know Java is compiled to bytecode and then (JIT-compiled or interpreted?) to native machine code

ATaco

JIT, Or, Just in Time Compiling, Does have a compile step prior to actually running things. This offers a great increase in speed when looping, and a minor increase inspeed generally.

JVM is JIT compiled to Machine code, depending on what it's running.

musicman523

Okay that makes sense

How old is Lua as a language?

Looks like it got started in '93

ATaco

1993, 24 years old.

musicman523

I guess it could only rewrite x % (power of 2) to x & (1 less than that power of 2) if it knew for certain x would be nonnegative

ATaco

2:24 AM

Although, you could just rewrite it as sign(x)*(x & 2^n)

Anonymous

@musicman523 There are 3 major methods of running code. Interpreted code is parsed and run by an interpreter. Ahead-of-time compiled code is translated into another language, and the resulting code is run. Just-in-time compiled code is sort of a hybrid of the two - the code is read and parsed like interpreted code, but as code paths are executed, they are compiled into a different language and run. Those compiled bits are saved for later executions so the compilation step only has to be run once.

musicman523

multiplication is usually as slow as div/mod though

Thank you Mego!

Anonymous

The big difference between AOT compilation and JIT compilation is, if there are code paths that are never executed, JIT compilation doesn't have to pay the cost of compiling them

musicman523

The downside is that JIT compilation can't make whole-program optimizations

ATaco

And it has to compile at runtime.

Anonymous

2:26 AM

@musicman523 Not without an extra full-program parsing step

musicman523

@Mego which I would think negates the other benefits of using JIT compilation

Anonymous

Not usually

ATaco

JIT is usually better than just interpreting.

musicman523

Wait, parsing step? Just parsing?

ATaco

And it offers more flexibility than AOT.

Anonymous

2:27 AM

@musicman523 If the JIT compiler reads and parses the whole program first, it can make whole-program optimizations before running the code, but it's not usually done because the benefits aren't usually enough to offset the cost.

musicman523

I've never written a JIT compiler or studied it in school which is why I don't know anything about it

I've written and studied interpreters and AOT compilers though

Phoenix

JIT is complicated.

ATaco

JIT's not incredibly popular.

Phoenix

@ATaco Actually, Java 9 has an option for AOT compilation.

musicman523

Oh that's cool

Anonymous

2:29 AM

JIT compilers compile statement-by-statement rather than whole programs, so it's usually slower than running a compiled program, but can make that up by not compiling everything, especially if there's a lot of dead code.

Phoenix

JIT is basically only used in JVM and .NET languages.

Anonymous

JIT is hard to do right, but it can be very beneficial

Phoenix

So basically only Java and C#

musicman523

And Scala, Clojure, Lua, Kotlin, and any other JVM language (there are a lot)

Anonymous

Smalltalk is pretty famous for being one of the first JIT-compiled languages

Phoenix

2:31 AM

Does JIT necessarily require bundling a massive VM runtime?

Nathan Merrill

if you JIT compile down to ASM, how do you send that to the CPU?

Phoenix

@musicman523 There are a lot, and there are a lot of .NET languages, but not many use them.

Nathan Merrill

like, you can't modify the currently running program

musicman523

@Phoenix Scala and Clojure are pretty popular (Scala especially is very popular right now), and Kotlin is growing

Phoenix

I suppose.

@NathanMerrill You can modify the currently tunning program.

Nathan Merrill

2:32 AM

popularity of languages largely depends on who you hang out with

musicman523

And Java and C# are still very big in industry

Phoenix

You can see this especially in the debugger.

musicman523

Yeah, you can modify/add/remove instructions in memory at runtime

Phoenix

Visual Studio even has an option to attach a class to a currently running .NET application and let it do things.

Nathan Merrill

@Phoenix I'm assuming that requires another level of abstraction

can you modify stuff at runtime, with say, C++?

Anonymous

2:34 AM

@NathanMerrill You can exec anything

Phoenix

Also, JVM and .NET both support reflection.

Anonymous

@NathanMerrill To some degree, but it's very very hard

Nathan Merrill

right, but the JVM and .NET are huge abstractions over assembly.

Anonymous

You can load libraries at runtime with dlopen and whatever the equivalent is in the MS ecosystem

musicman523

@NathanMerrill You would almost need to know ahead of time what the assembly around you would be, and modify it in memory

Nathan Merrill

2:35 AM

@Mego is exec an assembly command?

Phoenix

As a JVM or .NET program is being run, the runtime is aware of what all the functions are called and stuff.

In fact, you can decompile .NET or JVM code back into C# or Java.

Anonymous

@NathanMerrill exec is a system call. It replaces the currently-running program with a new program.

Phoenix

It might look ugly, but the structure will all be there.

Nathan Merrill

@Mego does it "return back" when the program finishes like a function?

Phoenix

It does not.

I suppose you could make it return back with some magic.

musicman523

2:36 AM

Oh you mean the exec syscall

Not like python exec

Anonymous

@Phoenix There are some debugging settings you can use in the .NET C# compiler to preserve symbol names, so it doesn't even have to be ugly

musicman523

Yeah no, the syscall exec literally overwrites the current address space with the new program (source: have implemented it)

Anonymous

@NathanMerrill Not by default, but like Phoenix said - with some magic, you can make it work. It's almost certainly not how most JIT compilers do it, but it's definitely possible to run arbitrary machine code at runtime.

Phoenix

@Mego The symbol names are preserved anyway.

Reflection has to work somehow.

Nathan Merrill

@Mego that's cool

Anonymous

2:38 AM

@Phoenix I meant symbol names inside of local scopes that wouldn't be touched with reflection, like local variables inside of a function.

Phoenix

Ah

musicman523

@Mego You cannot make the syscall exec return back to the original program in the normal way because it's no longer in memory

You could do one thing, though

Phoenix

You could store the orginial program somewhere else

musicman523

Even if you could, say, load the program back into memory, change the entry point (the "main") to where you left off, and jump there

You wouldn't have any of your variables or things allocated in memory that were there before

The only way to do it would be to write the entire contents of memory to disk and then restore them

And then go back to where you left off in the original program

But it's not done. A better way would be to fork a new process running the same code. The parent process waits for the child process to exit, and the child process execs the new program

fork, wait, exec, and exit all being syscalls as well

Aaaaaaaaaaaaaaand I've killed the room. :)

ATaco

😉

Anonymous

2:46 AM

@musicman523 Continuations are neat and do all things

musicman523

@Mego Continuations are SO neat

I implemented lazy evaluation with continuations once, it was so much fun

Downgoat

3:03 AM

TikZ question: if I have these styles:

    ast/.style={ shape=rectangle, fill=white, draw=black },
    shrink/.style={ ast, below=-7mm, label=left:#1 }

does anyone know why TikZ would not apply any of the styles as defined by shrink

Wheat Wizard

Nope, styles are out of my depth.

Downgoat

do you know how you would declare a TikZ macro or it's equivilent?

I optimally want equivilant of C: #define shrink(name) node[ast, below=-7mm, label=left:name]

Wheat Wizard

I use \def

\def\shrink[#1]{node[ast, below=-7mm, label=left:#1]}

musicman523

Hey C++ experts, when you manually call the destructor for an instance of a class does it free the memory or not?

Wheat Wizard

Depends on how you have defined your destructor

musicman523

3:12 AM

implicit

Try it online!

Wheat Wizard

Oh no clue then. I usually define my own.

Just Valgrind it

musicman523

good call

Weird result.....2 allocs, 1 free

But I only ever call new once...

Wheat Wizard

I have an issue where my computer always allocs a static amount of memory for every program.

So it looks like I'm always leaking memory in Valgrind

musicman523

That's so weird

What operating system are you running?

Wheat Wizard

Arch Linux

I actually nuked it. So I'm running mac rn which does not have the problem

musicman523

3:23 AM

Strange. I wonder if it's something with the OS or the implementation of malloc

Wheat Wizard

It doesn't happen on every arch linux install

I don't know anyone else that has the problem

musicman523

it really seems more like that is a "feature" than a bug with a particular install

Helka Homba

come play stuff @goat and sundry chat.stackexchange.com/rooms/59120/codenames

musicman523

that is really weird

Wheat Wizard

I had a CS course in Data Structures and every lab I had to convince the TAs that my computer just leaks memory no matter what.

They would take off points for memory leaks.

Downgoat

3:27 AM

@musicman523 wel it depends on how you call your fields but you should use std::unique_ptr and std::shared_ptr not new

New Sandboxed Posts

3:40 AM

0

A: Sandbox for Proposed Challenges

Is there a total ordering? code-golf string decision-problem TL;DR: Given a set of strings, determine whether the characters expose a total ordering based on location in the strings. In this challenge, strings are used as a predicate to determine the order in which characters should appear i...

musicman523

3:58 AM

@Downgoat I learned old C++, not new C++, so I've never used any pointers but raw ones

@WheatWizard I guess the only way around that would have been to set up a VM and see if that also leaked, and write/run your programs in that

Wheat Wizard

I was able to convince them so I didn't lose points

musicman523

That works

Wheat Wizard

Once I even was able to convince them that memory I was not supposed to leak was part of the standard leak

musicman523

you monster! :P

For our data structures course you would submit your source code to a web server which would then compile and execute your code in a sandboxed environment with prewritten test cases

(And also throw everyone's code into MOSS to check for plagiarism)

Wheat Wizard

We had that too, but not for labs

musicman523

4:04 AM

What for, then?

Wheat Wizard

For homeworks

musicman523

Ah. Our class had 8 "assignments" which constituted labs and homeworks

When I took it, only 1 was a written assignment. The professor who taught it last semester did some that were just programming and some that were half and half

Wheat Wizard

We had homework and labs every week. Ended up being no joke 50hrs of class work a week.

musicman523

That's rough. Ours was first assignment = 1 week, and 2 weeks for all the other assignments

Wheat Wizard

The were using Data structures as a strainer course for CS majors. Trying to get rid of as many people as possible

musicman523

4:08 AM

Ours also ends up thinning the waters quite a bit

Not because they are trying to get rid of people, but simply because the people who fail are seriously not ready for junior-level courses

Last semester I believe the failure/withdrawal rate was 38%

Wheat Wizard

Thats pretty steep

musicman523

No wait, I think the professor said that was if he didn't curve the class

which he absolutely did

Wheat Wizard

I don't know what ours was but a few people I know dropped CS because of it.

musicman523

Yeah I know a few people who have dropped CS because of that course and the one before it, but I know many more have simply because they aren't in the next course

Wheat Wizard

What do you mean?

musicman523

4:12 AM

Because of the limitations on when courses are offered after data structures, there are few that make sense for CS majors to take, and I was in most of them and knew people in all of them

And the numbers definitely dropped significantly, even if some people are taking fewer CS courses at a time

Wheat Wizard

Ah I see.

musicman523

Our Intro to CS courses regularly handle 600 students and are growing (though many students are from other departments). We graduate about 120-150 per year

Wheat Wizard

We have about the same

CS 1 is required for all students and there is only one section so it is a bit of a shitshow.

Actually CS majors don't have to take CS 1 ironically

musicman523

Ah. My school offers CS 1 for non-majors and CS 1 & 2 for majors. In the fall CS 1 is larger, and in the spring CS 2 is larger; during these semesters, they each have 2 sections of just under 300 students each (taught by the same professor)

Although this fall they have 2 professors teaching CS1 so I'm not sure how that's going to work

Not all students are required to take a CS course though, so most don't unless they are EE, Comp. E, or an applied math major (or a CS major of course)

New Main Posts

-1

Q: text to braille conversion

like this i want code for text to braille in mat-lab code please any one help me This program works on user input. So for example if a user inputs A, the braille combination is displayed. If a user inputs a word the braille combination of that word is displayed. The same goes for sentences. text...

image-processing matlab

Wheat Wizard

4:19 AM

Anyone in our school of science and engineering is required to take CS 1 for sure. I don't know about other people because there are not a lot of them and I don't know any of them.

musicman523

The school of engineering offers their own course in Matlab because the CS department doesn't teach one. They also started offering their own Intro to Programming in C/C++ because they thought (rightfully) that our Intro to CS courses don't do a good job

Wheat Wizard

That makes more sense. My uni just teaches everyone how to write simple python scripts and expects people to use it. I feel like it might be more practical for engineers to do matlab and such.

musicman523

Our Intro to CS 1 is really Intro to Object-Oriented Programming, and our Intro to CS 2 is really Exceptions and Data Structures

Which I think are both crummy topics to focus on for the first year

But that's changing soon

Downgoat

idk if I've already asked but how would you represent this in an equation: 'x is in domain of function f'

I am like 68% sure there is operator for

musicman523

You did ask, but we didn't come up with a symbol for "domain of"

Destructible Lemon

4:29 AM

I'm making an extra big image now :O full size with all the colours

Wheat Wizard

$f:X\rightarrow Y;x \in X$

Destructible Lemon

it will take about an hour maybe?

Wheat Wizard

I don't think there is a domain of symbol

New Main Posts

0

Q: Computers will never taste a refreshing cookie

Inspired by github.com/JackToaster/Reassuring-Parable-Generator, in turn inspired by xkcd.com/1263. The possible words are derived from the reassuring.cfg of that repository. Having a look at the reassuring.cfg (use the 12th commit) is suggested to see the grammar the output matches (The output ...

code-golf string kolmogorov-complexity grammars

Downgoat

@WheatWizard what is Y in this context?

musicman523

4:33 AM

$x \in A\ where\ f:A\to B$ could work as well

Still don't have MathJax working :(

Wheat Wizard

@Downgoat Y is the codomain

musicman523

Did you mean domain or codomain?

I think Downgoat asked for domain

Wheat Wizard

Y is the codomain X is the domain

Leaky Nun

@musicman523 Y is the codomain, X is the domain

Downgoat

codomain = range right

musicman523

4:34 AM

oh yes right thank

brain

Leaky Nun

@Downgoat no, that's for surjective functions

musicman523

@Downgoat range is a subset of codomain

ATaco

@musicman523 A-ta.co is back up.

Wheat Wizard

ninja'd

Downgoat

musicman523

4:36 AM

@ATaco Thank you!

Downgoat

does this make any sense

Leaky Nun

@Downgoat yes

Downgoat

:D

Leaky Nun

$n$ is mapped to $t(n)$

Wheat Wizard

kindof, It doesn't mean much

Downgoat

4:36 AM

I basically have:

Wheat Wizard

It means literally nothing

musicman523

Yeah, to me it means "mapping of n to t(n)" or in other words "t"

Downgoat

I want to say for all n in V(T), t(n) is defined

V = all nodes in graph T

musicman523

Vertices

Gotcha

Leaky Nun

@Downgoat i wouldn't say this

Destructible Lemon

4:37 AM

generating a big image!!

the anticipation...

Wheat Wizard

@LeakyNun Yeah I realized

musicman523

@DestructibleLemon what mean you this, mysterious lemon?

Destructible Lemon

my all colour images

I will post some all colour images I generated. if anyone else...

I'm making the max size one

Leaky Nun

@Downgoat just say $V(T)$ is a subset of the domain of $t$

musicman523

Oh that's cool!

Downgoat

4:39 AM

@LeakyNun is there a syntax for domain of. or should I just do like:

Leaky Nun

not sure

what exactly do you need it for?

musicman523

I can ask my mom in the morning

Downgoat

@LeakyNun I am writing a proof

Leaky Nun

@Downgoat just show me the whole context

Wheat Wizard

Just write it in english

often english is easier to understand than symbols

musicman523

4:40 AM

Are you required to use symbols?

Downgoat

wait shit should not have (n)

Leaky Nun

@Downgoat not enough context. show me where you define $t$.

Downgoat

it's defined as t(N) = (C, L, Xi)

Leaky Nun

and what do those symbols mean?

Downgoat

(C, L, Xi) are random variables defined elsewhere

musicman523

4:42 AM

Oh god the memory haunts me

I had to write out an algorithm in pseudocode

That involved N_{ijk} and N_{ij} and X_i and all this nonsense

the horror

the horror

Destructible Lemon

also I think I might have slightly messed up my older images :/

Leaky Nun

3

A: How to denote domain and range

Notations vary. You should pick something straightforward, and state it clearly up front. One common choice would be $$\operatorname{dom} f\\ \operatorname{ran} f$$ another reasonable choice might be $$\mathscr{D}(f)\\\mathscr{R}(f)$$ Principia Mathematica used:

musicman523

oh no D:

Leaky Nun

@Downgoat just use $\operatorname{dom}(t)$

musicman523

I agree, I think "dom" is the most clear

Downgoat

4:44 AM

:O can I use the fancy D :P

musicman523

Although the fancy D symbol is nice

Downgoat

>_< shit that sound inapproriate :P

ok so I ended up doing:

can I do like V(T) -> dom(t) as shorthand

wait = doesn't this mean V(T) = dom(t)

Wheat Wizard

Why not V(T) subset of dom(t)?

Downgoat

because I am stupid :P

Wheat Wizard

Thats the definition of subset :p

ATaco

4:49 AM

I like $\mathscr{D}(f)$

Downgoat

I like it too but apparently I need some weird fonts install and LaTeX editor throws a lot of BORK ALERT 🚨 BORK ALERT 🚨 BORK ALERT

Destructible Lemon

5:34 AM

ooo, the image should be done soon

Downgoat

5:49 AM

omg I just installed a Mac binary from latest source and it is apparently only for PowerPC macs

Destructible Lemon

:/ that is annoying

Downgoat

like PowerPC is from like last generation

Destructible Lemon

:/ is it borking computer?

ok I really should have put a loading bar on this image generator so I could tell how far through it is

Downgoat

great :| the makefile is so old it uses ancient flags for build >_<

ASCII-only

6:02 AM

@Downgoat gcc --ancient? :P

H Walters

gcc -x antikythera

Downgoat

Now I understand why we don't have flying cars yet— humanity is still using TeX

Phoenix

Object Oriented HTML

> There are a lot of IDEs for object-oriented languages, and a lot for HTML, but no IDE or IDE plugin for object-oriented HTML.
We should change that.

Downgoat

@Phoenix this is called React

Phoenix

<html extends="file2.oohtml">

Downgoat

6:11 AM

react:

Phoenix

@Downgoat Isn't React a JS library?

Downgoat

react: class MyHTMLElement extends MyOtherHTMLElement { ... }

@Phoenix it blurs the line between JS/CSS/HTML

really they are all one of the same

and then you can do:

<html>
    <MyHTMLElement> ... </MyHTMLElement>
</html>

React is really awesome— except for when it doesn't work then it's a piece of crap

mınxomaτ

10:14 AM

@PhiNotPi Nice.

TheLethalCoder

10:43 AM

@Cowsquack Is anything before the ^ in Carrot a string literal? i.e. would the Hello part of the following be considered a string literal? Hello^*1

Adám

@Cowsquack I keep reading your username as Cow-squack.

Erik the Outgolfer

11:01 AM

oh wait cows quack is only in 5 rooms atm...o_o

Cows quack

@TheLethalCoder Right now, I would say yes. But later it will be different depending on the stack-mode.

right now, it always prepends the literal formed by the carets to the stack as a string, but later it might add a float, or push an element to an array, etc.

TheLethalCoder

@Cowsquack Thought it might be at the moment...

Cows quack

if you want, I could give you a link of planned features that acts as an incomplete specsheet

TheLethalCoder

Sure, I'm enjoying playing with Carrot. I'd love to see where you're going with it

Okx

@Phoenix how dare they. IDs are supposed to identify HTML, not classify it.

Cows quack

11:13 AM

@TheLethalCoder Here

Okx

such blasphemy puu.sh/wItfo/78e296a581.png

TheLethalCoder

@Cowsquack Thanks! Regards to your TODOs on the caret in array-mode would they not just append and prepend an item to the array for the caret and down-caret respectively?

Cows quack

that is a possibility, but I also thought about just performing the down/up caret on the last element of the array instead. Since I am not really sure, I added the TODOs there. What do you think is better?

TheLethalCoder

Will you be adding the ability to access an array by index? (I haven't read through the entire doc yet) If so I'd say appending and prepending, if not on the last element.

Anonymous

11:32 AM

Last chance for feedback before this goes up on main in a few hours

totallyhuman

11:49 AM

CMC: given a number $n$, output $i^n$ (imaginary unit)

Adám

@totallyhuman APL: 0J1∘*

Anonymous

@totallyhuman Actually, 2 bytes: ïⁿ

totallyhuman

@totallyhuman Python, 33 bytes: lambda n:['1','i','-1','-i'][n%4]

Anonymous

Oh I misread

totallyhuman

@Mego actually has a builtin for $i$?

Anonymous

11:54 AM

Yep

totallyhuman

also would this be too trivial of a challenge for main?

Anonymous

If it's just integer powers, probably. With real/complex powers, probably not, but it's probably a dupe.

totallyhuman

oh

man i cannot think of a good challenge

18

Q: Find i^n, given n

The Challenge In as few characters as possible, find the value of i^n, given n, a positive integer greater than 0. This should be outputted as a String. For those that don't know, i is defined such that i^2=-1. So: i^1=i i^2=-1 i^3=-i i^4=1 This then repeats.. Rules If your language supp...

code-golf math arithmetic complex-numbers

Anonymous

@totallyhuman You can shorten that to lambda n:[1,1j,-1,-1j][n%4]

totallyhuman

or to lambda n:'1i--'[n%4::-2]

Anonymous

12:02 PM

Or lambda n:1j**n, or 1j.__pow__

Bruce Forte

Hi, what is the general consensus on taking multiple arguments as a list?

And what if you need to take 4 args, eg. m,n,a,b can you mix this to say accept m,n as single args and a,b as a list?

I had a look over at meta but could only find that it's acceptable to take a single arg as a singleton list :S

TheLethalCoder

I believe if your language supports multiple arguments you should take them in as such but I'm not 100% on that. TBH probably best asking the challenge OP if taking input as a list would be fine instead of individual arguments.

Leaky Nun

@BruceForte you can

Bruce Forte

Thanks @LeakyNun! Saves me 6 bytes :D

I saw someone other doing it on the same challenge and nobody complained, but that's good to know

Anonymous

12:21 PM

IIRC it's acceptable to mix-and-match input methods unless the challenge specifically says otherwise

Adám

12:39 PM

@totallyhuman Thank you. I beat the currently accepted answer.

totallyhuman

Nice

Leaky Nun

> outer
function (X, Y, FUN = "*", ...)
{
    if (is.array(X)) {
        dX <- dim(X)
        nx <- dimnames(X)
        no.nx <- is.null(nx)
    }
    else {
        dX <- length(X)
        no.nx <- is.null(names(X))
        if (!no.nx)
            nx <- list(names(X))
    }
    if (is.array(Y)) {
        dY <- dim(Y)
        ny <- dimnames(Y)
        no.ny <- is.null(ny)
    }
    else {
        dY <- length(Y)
        no.ny <- is.null(names(Y))
        if (!no.ny)
            ny <- list(names(Y))

CMC: guess the language (no cheating!)