The Nineteenth Byte

Lembik

9:14 AM

hi all

are code-golf assembly questions acceptable?

oh do we have too few people who know assembly?

Anonymous

@Lembik Why wouldn't they be?

Qwerp-Derp

SQL is bad when something like SQL injection is real

Is MongoDB relatively better?

Lembik

@Mego sometimes people don't like questions that restrict the language

Helka Homba

Is there a quick way to make a dict like {'a':138, 'b':278} from string "a 138\nb 278" in python?

Lembik

there have only ever been 14 assembly questions! codegolf.stackexchange.com/questions/tagged/assembly

Qwerp-Derp

9:19 AM

I need a good name for a really small lifeform other than cell

Like something a virus "generates" but it may or may not generate that in reality

Lembik

virlet?

Qwerp-Derp

I need a simple name

Lembik

that was quite simple :) Do you mean you want to use a real word?

DLosc

@HelkaHomba Not exactly (AFAIK). By "quick" do you mean "golfy," or "built-in," or...?

Qwerp-Derp

@Lembik Yeah, preferably

If not I'll have to make do with organism

Helka Homba

9:21 AM

@DLosc I guess I mean without for i in range(0, 2, len(inp)): stuff

golfy or built-in

@Qwerp-Derp microorganism

DLosc

My initial stab:

d={}
for entry in inp.split('\n'):
 k,v=entry.split()
 d[k]=int(v)

Helka Homba

oh, not bad

DLosc

Not fully golfed. Might be shorter as a dictionary comprehension.

Qwerp-Derp

@HelkaHomba Too long

Helka Homba

@Qwerp-Derp microbe

Qwerp-Derp

9:22 AM

@HelkaHomba Ooh that's a good one

Helka Homba

or nanobe :P

betseg

This guy wrote a C compiler that only outputs mov instructions, and it works. github.com/xoreaxeaxeax/movfuscator

Helka Homba

yoctobe

Qwerp-Derp

@betseg I've seen that a lot of times already

betseg

@Qwerp-Derp I haven't ;_;

Qwerp-Derp

9:24 AM

@betseg But it is cool

You didn't let me finish :(

betseg

Aww ;_;

wizzwizz4

@HelkaHomba dict(x.split()for x in"1 138\nb 278".split("\n"))

Qwerp-Derp

Is it efficient or inefficient? I'm curious

betseg

Obviously inefficient

Qwerp-Derp

But mov is fast isn't it?

It's inefficient of course, but not that inefficient I think

betseg

9:27 AM

Don't underestimate new processors :p

Lembik

@betseg how do if statements work?

betseg

No idea.

mınxomaτ

@Lembik Watch the talk.

REcon 2015 - The movfuscator (Christopher Domas)

►

It's all explained from the ground up.

Lembik

49:14!

like is too short :)

DLosc

@wizzwizz4 Ah, nice. Only one caveat: this will make the values strings, whereas the original request had values as ints. Dunno if this makes a difference to what @HelkaHomba is doing with it.

Lembik

9:29 AM

I wonder how they produced the video

actually if is explained at 8:50

Helka Homba

@xnor this is what you look like in my new language, Kinetic Logic

.>.>
(-(-
.{}.
..{}
[--]
>+].
+)..

Qwerp-Derp

@HelkaHomba What's the lang about? The name seems underwhelming compared to other names like Stackylogic :P

We need a slightly punny name

wizzwizz4

@DLosc Oh. That's easy enough to fix: dict((x,int(y))for x,y in (x.split()for x in"1 138\nb 278".split("\n")))

DLosc

... At which point, however, using an actual for loop is shorter. ;)

Helka Homba

@Qwerp-Derp It's a language meant to be (relatively) easy to physically execute in real life with balls as bits that are directed down tubes and pathways

Qwerp-Derp

9:34 AM

@HelkaHomba Marbelous 2.0?

Uhh Grabity? The "bit" doesn't really stand out though

Helka Homba

@Qwerp-Derp Not really. Physically implementing some marbelous commands looks tough or prohibitive

Qwerp-Derp

Also @HelkaHomba do you have a repository?

Helka Homba

Sep 7 '16 at 21:00, by Helka Homba

I would still love to build an entire K'NEX computer the size of a gymnasium someday

@Qwerp-Derp not yet

DLosc

@wizzwizz4 {x:int(y)for x,y in(e.split()for e in string.split("\n"))}

(May not work in older versions of Python, can't recall)

wizzwizz4

@DLosc Of course you can do that! :-/

Qwerp-Derp

9:37 AM

Domino computers are pretty cool, I saw Matt Parker's domino computer but I'm not sure if that's optimal

I also think I might have a way to put memory in dominoes? I'm not sure

DLosc

That's the exact same length as

d={}
for e in string.split("\n"):k,v=e.split();d[k]=int(v)

Which one is shorter therefore depends on if you just need the dictionary once in an expression, or if you need it stored in a variable. Also, whether this is inside an indented block.

Lembik

so my challenge would be simply to divide two complex numbers in assembly

code-golf

wizzwizz4

@DLosc That code hangs on my machine if the variables aren't defined.

Helka Homba

Guys, I've got it good enough ;)

DLosc

@HelkaHomba We inferred an implicit "CMC:" at the beginning of your request. ;)

@wizzwizz4 Which variables? string?

Helka Homba

9:47 AM

CMC: Given 3 distinct 1-digit numbers (0-9) output the middle one. e.g. 4 2 9 -> 4

as in second largest

DLosc

Given as a string? List? Whatever's most convenient?

Qwerp-Derp

Uhh what about a language that's based off of an infinite list of integers just going 1, 2, 3, 4... and you have to modify it to get what you want?

Helka Homba

@DLosc whatev convenient

DLosc

Fairly boring answer in Pip, 6 bytes: (SNa1) (where a is a 3-character string of the digits)

In pseudocode: sortNumeric(arg)[1]

Now for an answer in Acc!!

Helka Homba

yeah, not too hard

Kritixi Lithos

9:53 AM

V, 5 bytes. (trailing newline in output) Try it online!

Anonymous

@HelkaHomba Actually, 2 bytes: S║

Qwerp-Derp

Maybe reduce commands weaved in there?

mınxomaτ

TIL: blog.wolfram.com/2017/01/31/…

wizzwizz4

@DLosc I had no variables defined

DLosc

Yeah, and my code assumes you've got the string in string. Modify accordingly.

Fatalize

9:55 AM

@HelkaHomba Brachylog, 3 bytes: o↺h

Anonymous

21 bytes in Python 3: lambda*x:sorted(x)[1]

Kritixi Lithos

@KritixiLithos Wait, that is wrong

Dada

Perl, 23 bytes $_=(sort{$a-$b}@F)[1] (+ -ap)

Fatalize

Loads of 3 bytes answers in fact, o↻t, obh, okt, o∋₁

Matthew Roh

My nerd score is 69

Which means

Ummmmm

idk

Kritixi Lithos

10:06 AM

Mine is >70

Probably would have increased since the last time I took the test

Matthew Roh

But, 69?

69????

~~FUUUUUUUUUU~~

Anonymous

@MatthewRoh That's not really necessary.

Matthew Roh

explodes into pieces

Qwerp-Derp

Nerd test what

DLosc

@HelkaHomba Acc!!, first attempt, 133 bytes:

3600*N+60*N+N
Count i while i-2 {
Count j while j-2 {
_+(_/60^j%60-_/60/60^j%60)*60^j*59*(_/60^j%60/(_/60^j/60%60))
}
}
Write _/60%60

(Inputs three characters, runs two rounds of bubble sort, outputs the middle one)

Actually, unless the huge expression can be golfed further, I think that might be the shortest it gets. Subtracting out 48 to get the actual digits won't help much because all the 60s would only become 10s, not saving any bytes.

Matthew Roh

10:20 AM

Acoording to nerdtest 2.0 I am a nerd god

Qwerp-Derp

wtf i searched up nerd test and this came up:

Which of the following is not a real programming language?
BASIC
C++
PASCAL
Assembly
FORTRAN
COBOL

WTF all of these are programming languages

Sp3000

Assembly's technically a type of language, so considering that nothing else makes sense if I had to choose I'd go with that

Anonymous

There are two correct answers, from two different viewpoints

Anonymous

One is Assembly, for the reason Sp3000 has stated

wizzwizz4

{x:int(y)for x,y in map(str.split,"1 138\nb 278".split("\n"))}

Qwerp-Derp

10:28 AM

I'm assuming the other is COBOL?

Anonymous

An answer without a reason is much the same as a doorframe without a door

Qwerp-Derp

Actually no I don't know, just a random guess

Sp3000

Oh, I think I see what Mego's getting at. Similar reason, isn't it...

Qwerp-Derp

@Mego What's the other reason?

Anonymous

@Sp3000 I see what you think you see, and no, because that reason is technically incorrect (the worst kind of incorrect).

Anonymous

10:31 AM

@Qwerp-Derp The other answer is COBOL, and its reason is that COBOL is an instrument of torture, not a programming language.

Qwerp-Derp

@Mego Dammit Mego why do you have to do this :(

Anonymous

Sp thought I meant BASIC, because there are many types of BASIC, but there is actually a programming language named BASIC.

DLosc

@Sp3000 Cue someone here creating an esolang and naming it Assembly

Sp3000

I was going to say "Oh I think I see why what I think you thought I thought was technically incorrect"... but yeah, had to look up BASIC because I thought it was a language but wanted to double check and got myself confused doing so

DLosc

@DLosc Golfed 7 bytes:

Count i while i-3 {
_+N
Count j while j-2 {
_+_/60^j%60/(_/60^j/60%60)*(_/60^j%60-_/60/60^j%60)*60^j*59
}
}
Write _/60%60

Anonymous

10:36 AM

@Sp3000 It's good that you interrupted the recursion before it become incomprehensible.

DLosc

Whoops, nvm, that gives a zero division error apparently.

Qwerp-Derp

10:59 AM

What about a language that worked on reduce functions?

Anonymous

11:10 AM

I can't fathom what made you think that was appropriate. Please consider your messages before posting them, not after.

2

Matthew Roh

Okay

I think PPCG is the place where most esolang has been born, do you think so too?

ATaco

I don't have evidence to suggest otherwise.

Esolangs are, by nature, obscure. If many where created on some strange far away forum, We'd have no reason to know about it, and such, we couldn't determine if it were more or less than our current creation rate.

Anonymous

@MatthewRoh We're up against the esolangs wiki and Chris Pressey. That's some steep competition.

Matthew Roh

But acoording to the esolangs wiki, we did make a lot

ATaco

A thousand brainfuck Derivatives is cheating.

Matthew Roh

11:19 AM

True, just ignore the non-clever-and-just-a-copy-of-bf ones

And that includes 420

Which is a direct copy of chicken

ATaco

A Brainfuck derivative is to be considered a valid abstraction, and thus, valid Esolang, if the difference between itself and either brainfuck or another derivative is more than simply additional or modified command set.

Matthew Roh

Are electronic circuits valid as a language

ATaco

Yes, but their generally non-competative.

It's hard to score

Anonymous

People have used Verilog and VHDL in the past, though I'm not sure if there was ever a consensus on how to score them

Matthew Roh

But if we can, it'd most likely be a matter of parts

Like logic gates

ATaco

11:23 AM

Hard to determine if you validate random chips.

Matthew Roh

Hmm

ATaco

import custom_chip doesn't work on Hardware.

Matthew Roh

So we should restrict the parts

Anonymous

NAND gate golfing would be cool - solve the task using as few NAND gates as possible

ATaco

Nah, Restricting the parts kills a chunk of the fun.

Matthew Roh

11:26 AM

I interest in EE but EE + CS = X-Treme fun

So I dont really go pure EE

Erik the Outgolfer

I'd like NAND golfing, but I'm not sure how I'd score it.

ATaco

Wait... If there exists a binary file which takes a number input on STDIN and pushes to STDOUT whether or not it's prime, assuming it's older than the challange, would that be a 0 byte solution to the Primacy Challange?

Anonymous

@EriktheOutgolfer Number of NAND gates used would be the most obvious

Matthew Roh

The amounts of nand gates?

Umm, three posts simultaneously 😮

Anonymous

Here's an interesting idea for a EE hardware challenge: given a voltage difference N between two inputs, output a 5V difference if N is prime, and a 0V difference if not.

mınxomaτ

11:28 AM

In 2017 I shouldn't be startled by emoji suddenly being rendered everywhere, but I still am.

Anonymous

@ATaco Yes, but said binary file would also have to satisfy our definition of a programming language

Anonymous

@mınxomaτ Fuuuuutuuuuure...

ATaco

That challenge in particular allows any langauge.

It specifies "Basically our method of TC testing"

Anonymous

@ATaco Yes, any programming language

Matthew Roh

I like this answer of mine, most likely one of the first approaches of golfing with cellular automatons

ATaco

11:30 AM

Mego that is a rude abuse of my words.

Anonymous

@ATaco 1) They aren't your words. 2) That's neither rude nor an abuse.

ATaco

I feel personally rude'd at.

Anonymous

I'm sorry you feel that way.

Matthew Roh

@Mego *Any turing-complete object/non-object things

Erik the Outgolfer

I don't find anything rude there. It's just a clarification.

ATaco

11:32 AM

I hope it's obvious I'm joking.

Matthew Roh

@Mego So your saying, if the input voltage is prime, output 5v, if not, output 0v?

Anonymous

@MatthewRoh Yep. I was being very specific, since a lot of people don't realize that voltage numbers are meaningless without a reference.

Anonymous

@ATaco It's really not.

ATaco

My Apologies.

Matthew Roh

There should be some form of calculating the input voltage

But I cant find a way without getting crazy

So I'd rather say its near impossible

Well We could use some form of a microcontroller but that ruins the whole fun

ATaco

11:39 AM

Let's just hard-code the first 500 values and leave it at that.

Matthew Roh

We can use binary input but detecting primes with that? No.

Anonymous

Alright, how about this: you get 3 inputs. One is ground, one is +1V, and one is +NV. Outputs are ground and either +5V if N is prime or +0V if N is not prime.

Matthew Roh

@ATaco 😞

ATaco

The easiest way, surprisingly, to do this is to build a very small basically functioning computer.

Matthew Roh

idk, I dont see a way to really do that >_>

ATaco

11:41 AM

Me neither.

I'm a software guy.

Matthew Roh

I'm a soft/hardware guy but understanding some math is sometimes hard as hell

ATaco

I can't think of a way to do any kind of reliable math on a single line.

Matthew Roh

You eventually will learn things of those sorts

TuxCopter

Hello

Anonymous

With transistors, you can make op-amps. With op-amps, you can do voltage addition and subtraction. From there, you can implement division, and use trial division for the primality test.

Anonymous

11:47 AM

For even greater simplification: output is either +1V or +0V

Matthew Roh

op-amps..

:|

LegionMammal978

12:58 PM

Kinda reminds me of those chromosome charts

(For anyone wondering, it's the graph of the trajectory of a certain function)

betseg

6

Erik the Outgolfer

@betseg Where do they sell them? Also, what is the source of this image?

betseg

@EriktheOutgolfer found it on reddit ;_;

Erik the Outgolfer

1:15 PM

I don't know if the image is completely true though, it might be a prank, because that is normally STDERR output.

Matthew Roh

1:30 PM

So what should be output to STDERR, was been output to STDFCOOKIE?

betseg

enum { stdin, stdout, stderr, stdfcookie } ;

wizzwizz4

@betseg Where on Reddit?

betseg

@wizzwizz4 reddit.com/r/linuxmasterrace/comments/5szg3j/…

Matthew Roh

1:48 PM

Lel

We need a language where every output directs to stdfcookie

Fatalize

Question for native English speakers: are "necessary", "sufficient", "necessary and sufficient" standard ways of qualifying conditions in math?

Anonymous

@Fatalize Yep

Fatalize

Thanks

I really need to find a good translation dictionary for math stuff

Anonymous

I'm not sure that one exists outside of wading through Wikipedia and MathWorld articles.

Fatalize

2:05 PM

There's this or this but they mostly contain specific words and not phrases

AdmBorkBork

2:36 PM

Yes, but this is English. Where "sufficient" means different stuff depending upon the context.

For example, "sufficient" in statistics doesn't mean exactly the same thing that "sufficient" does in logic.

Fatalize

which is my problem

AdmBorkBork

"Sufficient" in statistics is more like the result. You've gotten as much information from this sample as you can, so this statistic is sufficient.

In logic, "necessary and sufficient" is an equivalent of "if and only if"

Fatalize

the second meaning is what I want

I just didn't know if "nécessaire et suffisant" tarnslated to English would be idiomatic

but seems like it is

AdmBorkBork

Yeah, sounds like it works for what you're wanting.

New Meta Posts

0

Q: Can the “How to Answer” pop-up disregard the association bonus?

We now have a custom “How to Answer” pop-up (yay!). However, this pop-up isn't shown to users who come here with the association bonus from other SE sites. In general, this behaviour makes sense: the pop-up is the same virtually everywhere and it teaches people how the Q&A platform in general wor...

Anonymous

2:52 PM

@NewMetaPosts Woohoo!

Anonymous

2

A: Let's customise the "How to Answer" pop-up

This was recently changed to the text proposed by coredump. New users who enter the Your Answer box will see:

7

Downgoat

@LegionMammal978 gel electrophoresis things?

TuxCopter

@Downgoat Nope. Both are [esperɛ].

Downgoat

Ok cool thanks :D

now french teacher won't be able to tell if I get wrong or right :D

@Fatalize do note sufficient generally has a "close enough" connotation so I'd be careful using it in math contexts

Fatalize

But Mego told me that a "sufficient condition" was idiomatic in Math...

(and with a precise meaning)

Anonymous

3:03 PM

@Downgoat The meaning of "sufficient" in math is different from its meaning in common speaking usage, in that it has a well-defined meaning.

Anonymous

"P is a sufficient condition for Q" means "P -> Q"

Downgoat

@Mego well depends on usage context too "X plus Y is sufficiently close to Z" is eeehh

Fatalize

^^ I'm talking about that usage

and "P is necessary for Q" means "Q -> P"

wizzwizz4

@Downgoat HTTPS works, but the actual site doesn't. See the "ES6's Unique Symbols" link in this article. You might have to script a patch.

Better still, add a redirect; cool URIs don't change.

Anonymous

@Downgoat That's abs(X + Y - Z) < ε. It's an entirely different usage, and an entirely different meaning conveyed. There's no confusion.

wizzwizz4

3:08 PM

Your new URIs seem easier to not change, but the old ones must be kept.

Lembik

any math geniuses in? I am trying to work out the asymptotic formula for {n+k-1 \choose k} when k grows with n

flawr

@Lembik What relation do you have between k and n?

Lembik

n = k^c for some constant c >=1

Anonymous

@Lembik Not a math genius, but I play one on TV. Glad to try to help.

Anonymous

So it can be expressed as C(k**c+k-1,k) for some constant c >= 1

Lembik

3:13 PM

yes

thanks!

flawr

Do you agree with this so far?

Anonymous

Which gives (k**c+k-1)!/k!(k**c-1)!

Lembik

yes!

I like this route already

Anonymous

I might be an idiot, lemme double-check my work

Lembik

really? No lower order terms?

wizzwizz4

3:14 PM

@Mego That seems overtly complex (i.e. hard to compute). What function is that for?

Downgoat

@wizzwizz4 that is weird it's loading from Wikipedia

Anonymous

Well, the numerator dominates the growth of the expression - (a+k)! is much larger than a!k! (for a = k**c-1)

Lembik

that is true

wizzwizz4

@Downgoat ... Really? On my machine it just sent me to /index.php/date/datey/date/Article-Name

It should've sent me to /Article-Name.

Downgoat

@Mego I didn't mean it strictly mathematically, merely in a mathematical context "too "X plus Y is sufficiently close to Z to draw conclusion...."

flawr

3:18 PM

@Lembik Do you agree with this: latex.codecogs.com/…

Downgoat

@wizzwizz4 ah shit

Thanks for the note

Anonymous

@Downgoat But we're talking about strictly-mathematical definitions here, so your concern is tangential

Lembik

@flawr let me thinkg

Anonymous

@flawr That's the simplification I was trying to make in my head but a lack of coffee was preventing me

Lembik

@Mego is it correct? :)

flawr

3:20 PM

Anonymous

@Lembik It looks correct to me

wizzwizz4

@Downgoat I also think there might be a mistake in the git thing; it's using HEAD~ in the code, but the text implies that it should be HEAD, and says that the code does what HEAD does. I'm not a git wizard, so this might be a false positive.

Anonymous

So it's equivalent to k**(ck) - 1

Anonymous

As flawr already posted :P

flawr

I'd add some parenthesis in your pseudo code=)

Lembik

3:23 PM

I am still working on checking this :)

flawr

@Mego Now I agree even less unless ** has a lower precedence than -

Lembik

I have to go for 30 mins.Thanks so much for this. back soon

New Main Posts

0

Q: Dynamic programming - decreasing number

I am given a number n.How many operations will it take me to get 1? I can 1. divide it by 3. 2. divide it by 2. 3. Subtract 1 Examples: Input Output 1 0 5 3 10 3

math

wizzwizz4

@flawr There's no reason that it should...

Anonymous

@flawr I'm incapable of putting parentheses in the right place without coffee.

Anonymous

3:26 PM

This is why I don't try LISP in the morning

flawr

@Mego blame-coffee

@Mego Then haskell might be a thing for you :D

Anonymous

Nonono, coffee is not the problem. Coffee is the solution.

flawr

Hm I just wanted to ping @Lynn, but it seem she hasn't been here recently.

Anonymous

Man, RO powers don't let me eleven my own messages. What even is the point?

Pavel

dammit I can not find a powerline font where the triangle aligns with the rectabgle bit properly

mınxomaτ

3:29 PM

@Pavel Wait, I had the same problem, lemme look quickly.

flawr

@Mego oh congrats btw! @mınxomaτ @DJMcMayhem too!

mınxomaτ

DjaVu Sans Mono for Powerline works for me.

But I remember I use something else on linux...

Pavel

I'll try that. At this point I'm ready for anything but courier new

Anonymous

@flawr Thanks :)

flawr

@Mego Contrary to popular belief room owner doesn't mean that you own the room but that the room owns you.

Pavel

3:31 PM

@mınxomaτ Now the triangle is below the rectangle.

mınxomaτ

I'm at 10pt btw.

Anonymous

@flawr From what I've seen the other room owners have to deal with, I'm strongly inclined to agree.

Anonymous

I tried closing the tab but the close button is gone

mınxomaτ

@Pavel Try github.com/powerline/fonts/tree/master/Cousine

wizzwizz4

@mınxomaτ Am I having deja vu, or was that link posted this morning?

mınxomaτ

3:37 PM

Timezones are a thing. I have no idea what you mean by "this morning".

wizzwizz4

@mınxomaτ I'm stupid. ~7 hours ago.

mınxomaτ

Dunno

Anonymous

@wizzwizz4 Nope

mınxomaτ

Searching for "Cousine" would have probably been enough. That's very specific.

Anonymous

Still nope

wizzwizz4

3:41 PM

@Mego This is what I saw

Kritixi Lithos

@Pavel which font are you currently using?

betseg

Have I just implemented OOP in C?

flawr

@betseg What is oop about this?

PS: stackoverflow.com/questions/351733/…

betseg

ok im over-exaggerating, but i haven't.... and i saw your link

flawr

"over-exaggerating" is somehow self descriptive XD

wizzwizz4

3:54 PM

@betseg Oh. I've just worked out what your code does; that's instance methods, not class methods.

In C, people usually write functions that take a struct as the first parameter (a bit like Python's self) and the compiler will complain if you use other data types as input.

mınxomaτ

I once added OOP to a non-OOP language: github.com/turbo/OOPEAu3/wiki

AdmBorkBork

> Javascript is a pestilence in the present days web and you open up for all bad you could imagine if you enable it feather-headed. And, sorry, but I had enough bad experiences with Microsoft in the past to not trust them running code on my system(s). -- some guy on a mailing list, complaining about Microsoft's websites

wizzwizz4

@AdmBorkBork They probably use Windows.

Either that or Mac...

mınxomaτ

macOS. Or OS X. Not Mac. It's not like anyone calls Windows "Surface".

AdmBorkBork

I dunno. From some of his other posts, he seems like a greybeard RMS follower.

I mean, I'm obviously stereotyping, but that's the impression I get.

Transcript for