The Nineteenth Byte

12:02 PM

@DJMcMayhem Here is a stack-safe program that detects whether a number is negative; how to detect if a number is less than the other is left to the reader as an exercise.

(I will not be creating a one-stack version of this.)

Anastasiya-Romanova 秀

@LeakyNun But if it's 0, it returns 0 (indicating 0 is a positive number)

Leaky Nun

@Anastasiya-Romanova秀 No, I said this program detects whether a number is negative

0 is not negative, so it returns false

0 is neither negative nor positive.

Fatalize

0 is both positive and negative

Leaky Nun

@Fatalize Is this what is taught in your educational system?

Fatalize

@LeakyNun Yes

Positive means >= 0

Leaky Nun

12:15 PM

@Fatalize That's weird.

Fatalize

Strictly positive is > 0

Leaky Nun

mathworld.wolfram.com/Positive.html

Fatalize

N contains 0

Leaky Nun

> A quantity x>0, which may be written with an explicit plus sign for emphasis, +x.

Fatalize

N* is [1,+inf[

Leaky Nun

12:16 PM

@Fatalize Oh, if we don't make that distinction, then positive just means >0

@Fatalize This is not universal. Some authors use N contains 0 and some authors do not.

Anastasiya-Romanova 秀

@LeakyNun True, but somehow the output is indicating if it's not negative, then it's positive since you only assign it either 0 or 1

Leaky Nun

@Anastasiya-Romanova秀 No, because 0 is neither negative nor positive.

We have two nomenclatures:
1. x<0 is "strictly negative" and x<=0 is "negative"
2. x<0 is "negative" and x<=0 is "non-positive"

Usually we use the second nomenclature in informal discussions.

Anastasiya-Romanova 秀

@LeakyNun Yes, I agree. That's why I said "true", but your output code indicating it's a positive number

Leaky Nun

@Anastasiya-Romanova秀 No, I said that it is not negative, which does not indicate that it is positive.

Fatalize

@LeakyNun Here negative always means =< 0

Leaky Nun

12:18 PM

@Fatalize in France?

Fatalize

yes

Leaky Nun

I see.

Fatalize

French wiki says (translated): Zero is the only number which is real, positive, negative and pure imaginary

Leaky Nun

@Fatalize I just read that also, and the English wiki says "Positive number [is] a number that is greater than 0"

Anastasiya-Romanova 秀

@LeakyNun Yes, but for someone who has no proper math knowledge would interpret either one of them

Leaky Nun

12:21 PM

@Anastasiya-Romanova秀 Not my responsibility to entertain people with no proper math knowledge

> Some authors begin the natural numbers with 0, corresponding to the non-negative integers 0, 1, 2, 3, …, whereas others start with 1, corresponding to the positive integers 1, 2, 3, ….

From fr.wiki:

> La tradition francophone inclut dans les entiers naturels le nombre zéro, ce qui n'est pas toujours le cas dans la tradition anglophone 2.

(The French tradition includes 0 in the natural numbers, which is not always the case in the English tradition.)

So TIL thanks to @Fatalize

Fatalize

Your tradition is weird I say!

Leaky Nun

@Fatalize I am not from the England.

Fatalize

@LeakyNun But you follow anglosaxon mathematical traditions

Leaky Nun

> anglosaxon

Adnan

20

A: Is zero positive or negative?

No. $\textbf{} \textbf{} \textbf{} $

4

Leaky Nun

12:24 PM

@Fatalize I follow the Chinese tradition

Fatalize

Hm?

Leaky Nun

From zh.wiki:

> 自然數，可以是指正整數 ( 1 , 2 , 3 , 4 , … ) ，亦可以是非負整數 ( 0 , 1 , 2 , 3 , 4 , … ) 。

(The natural numbers may refer to the positive numbers or the non-negative numbers)

> 正整數，在數學中是指大於0的整數。 (正整數 is integers greater than 0)

Fatalize

TIL chinese people are as wrong as the brits

:p

Leaky Nun

@Fatalize You may argue that the translation of "正整數" should not be "positive integers"

Anastasiya-Romanova 秀

@LeakyNun I respectfully disagree with this statement

Fatalize

12:26 PM

I don't know mandarin, so I can't argue anything

Leaky Nun

@Fatalize I mean, it is just a term that we use.

@Fatalize From my point of view the French people are wrong.

From my point of view the Jedi are evil.

Then you are lost!

@Fatalize lol

You were the chosen one! You were supposed to end the british, not join them!

Leaky Nun

...

@Fatalize You were supposed to do this challenge!

Fatalize

12:32 PM

Was I?

Leaky Nun

Why don't you? It's quite easy to do in Brachylog

Releasing Helium Nuclei

I tried the challenge, didn't understand how to :(

Leaky Nun

@ReleasingHeliumNuclei in which language?

Releasing Helium Nuclei

@LeakyNun pencil and paper :p i always try like that. couldn't get a(4)=1804 etc

I understood after some answers

Leaky Nun

@ReleasingHeliumNuclei ?!

Alright

Fatalize

12:36 PM

@LeakyNun Is it?

Leaky Nun

@Fatalize Not?

Fatalize

It doesn't seem that easy

Leaky Nun

@Fatalize So I do it?

Fatalize

to get a short solution

You don't need my approval to use my language, so if you want to do it go ahead

Leaky Nun

@Fatalize No, I've been doing in too many languages

I just want to give you this chance

Fatalize

12:38 PM

Are you planning to use the formula from you Brainflak answer?

Leaky Nun

I'll just try the approaches and see which one is the shortest

@Fatalize here

@Fatalize How does [1]* returns 2?

Fatalize

It doesn't?

it does

wat

Leaky Nun

@Fatalize It does

Fatalize

how is that possible

and it doesn't

Leaky Nun

...

Fatalize

12:44 PM

I'm pretty sure I know where this comes from

Leaky Nun

where

Fatalize

brachylog.tryitonline.net/#code=WzFdKg&input=SQ&args=Sg

It's a bug where [1] is seen as 1 after a comma

Leaky Nun

@Fatalize Wait, you can use variables in the input....

That could golf off some bytes

and is unfair

How does that go to the byte-count?

Fatalize

Of course you can, it's a variable just like the output

Leaky Nun

I think you should specify how the byte-count is counted

Fatalize

12:47 PM

I just don't use it to output things

Leaky Nun

@MartinEnder Can you explain to me \z vs $?

@Fatalize Then how is the byte-count counted?

Fatalize

What do you mean?

Leaky Nun

@Fatalize Well, that could be unfair if you have variables in the input

Some argue that the Z argument should be counted as well

Fatalize

Well I don't when I answer here so I don't see what the problem is

Leaky Nun

Because some think that it should be counted

Fatalize

12:50 PM

and even then I don't see how that's that unfair considering it's exactly the same in Prolog, since the output value is simply a unified variable

Leaky Nun

Well, I could put [1:Z] in the output argument

Fatalize

You could

Leaky Nun

which is what I mean when I said it could be unfair

So you saved one byte by not having to write t in the code

Martin Ender

@LeakyNun \z doesn't match in front of a trailing \n

Leaky Nun

@MartinEnder And why does $ fail?

Martin Ender

12:51 PM

because it matches in front of a trailing \n...

Fatalize

@LeakyNun No because the name Z in the input is independent from the Z in the first rule

Leaky Nun

But I thought $ is the end of string

Martin Ender

well, most of the time, it is.

Leaky Nun

When is it not?

@Fatalize Why doesn't this work?

Martin Ender

when there's a trailing \n...

most flavours do this, actually

Leaky Nun

12:53 PM

@MartinEnder I never knew this... this is strange

Fatalize

@LeakyNun It does, but the input is not a free variable (though it contains one) so it doesn't report its value

So it's not as abusable as you say it is

Leaky Nun

@Fatalize No?

Business Cat

If there's a trailing \n it won't include it because it matches the end of the line

Martin Ender

@LeakyNun the reason is that in many languages the "readline" function returns a string that includes the linefeed at the end of that line, and $ is then still expected to match after the last non-linefeed character.

Fatalize

@LeakyNun Output is not a free variable either

It will print the input or output only if it's an uppecase letter

Leaky Nun

12:54 PM

@Fatalize I see

Fatalize

basically

Leaky Nun

@MartinEnder I see

Fatalize

@LeakyNun You should trademark that sentence

Martin Ender

@LeakyNun it actually saves another to avoid that issue altogether: retina.tryitonline.net/…

Leaky Nun

@MartinEnder I see that you did not change the input this time xd

Martin Ender

12:59 PM

...

yeh

10/10 input

^

morning

@Downgoat Because it's a compiled language with an interpreter :D

ish

Downgoat

1:32 PM

@Poke morning

@Poke watwatwat

Rohan Jhunjhunwala

@Downgoat can u link me to the chesse factory

Poke

this is dumb schneier.com/blog/archives/2016/08/the_nsa_is_hoar.html

nsa needs to reevaluate how they do things

Rohan Jhunjhunwala

@Poke I agree what they do is moderatley shady

Poke

Read that blog post and it won't change your mind :D

Rohan Jhunjhunwala

My favorite graf

There are probably some overly pedantic word games going on. Last year, the NSA said that it discloses 91 percent of the vulnerabilities it finds. Leaving aside the question of whether that remaining 9 percent represents 1, 10, or 1,000 vulnerabilities, there's the bigger question of what qualifies in the NSA's eyes as a "vulnerability."

Poke

1:42 PM

yeah

also the vulnerabilities that they can't use for anything (discussed later in the blog)

Rohan Jhunjhunwala

The big question is do we need them, I think the answer may (sadly) be yes.

Poke

not to mention the NSA should likely be disclosing 100% of what it finds

Rohan Jhunjhunwala

Eh, if you aren't a criminal you probably have nothing to worry about. There's not much I personally can do about it, save writing in to a congressman who will promptly toss my letter

Poke

They have a function; it's just ill-defined

Rohan Jhunjhunwala

^

Poke

1:45 PM

Their website looks like an advertisement.

4

New Main Posts

1:58 PM

-1

Q: How take input for string of multiple line?

How take input for string of multiple line? either by whole one string, or by list of string for each line, or i have to use file handling for this Djokovic:Murray:2-6,6-7,7-6,6-3,6-1 Murray:Djokovic:6-3,4-6,6-4,6-3 Djokovic:Murray:6-0,7-6,6-7,6-3 Murray:Djokovic:6-4,6-4 Djokovic:Murray:2-6,...

python

TùxCräftîñg

0

A: How take input for string of multiple line?

in python how to read input from the keyboard (standard input) containing the results of several tennis matches. Each match's score is recorded on a separate line

... wat

Fatalize

how juic Federer

Rohan Jhunjhunwala

Flagged as not an answer @TùxCräftîñg

TùxCräftîñg

2:14 PM

@RohanJhunjhunwala in SILOS, is the space needed in x - 5?

Rohan Jhunjhunwala

@TùxCräftîñg yes, due to poor quality interpreter design. Call it a feature... I require quality code in a quality language. :( The next iteration of SILOS will probably have looser whitespace rules

TùxCräftîñg

k

Rohan Jhunjhunwala

@TùxCräftîñg did you flag that answer? I think its not an answer

TùxCräftîñg

yes i flagged it

Rohan Jhunjhunwala

Oh its already deleted

That was quick

TùxCräftîñg

2:16 PM

69

Q: Can you outgolf me? (Cops section)

Cops section The robbers section can be found here. Thanks to FryAmTheEggman, Peter Taylor, Nathan Merrill, xnor, Dennis, Laikoni and Mego for their contributions. Challenge Your task is to write 2 different programs (full programs/functions/etc.) in the same language and the same version (...

code-challenge number sequence cops-and-robbers

^ trying to answer this with SILOS

but OEIS is really slow :/

Releasing Helium Nuclei

any CMCs?

Adnan

@TùxCräftîñg You win if it survives for 7 days :p

TùxCräftîñg

brb

Rohan Jhunjhunwala

@TùxCräftîñg thanks, let me know when you answer is done so I can provide an upvote

@ReleasingHeliumNuclei my HTML parser with regex is not yet done

(The halting problem is an optional bonus)

Releasing Helium Nuclei

@RohanJhunjhunwala ;_;

<div><p><undefined tag>lol is HTML

you cant regex this

TimmyD

2:21 PM

CMC -- Given a month/day combination in any convenient format (e.g., mm-dd, MMM dd, etc.), output what day of the year it is. Assume this is not a leap year. -- Example, today 08-26 output 238.

Rohan Jhunjhunwala

@ReleasingHeliumNuclei you can, here take this guys advice

882

A: RegEx match open tags except XHTML self-contained tags

Don't listen to these guys. You actually can parse context-free grammars with regex if you break the task into smaller pieces. You can generate the correct pattern with a script that does each of these in order: Solve the Halting Problem. Square a circle (simulate the "ruler and compass" metho...

Fatalize

Is there any golfing language based on Haskell?

TimmyD

PowerShell, 29 bytes -- (date "2015-$args").DayOfYear ... built-ins ftw

Sp3000

Python's probably got something under some expensive date/calendar import :P

@Fatalize I don't know the language well, but maybe Burlesque? github.com/FMNSSun/Burlesque

(might be completely wrong, but first Haskell-related golfing lang that came to mind)

TùxCräftîñg

sry

Leaky Nun

2:29 PM

If nobody is going to this challenge in Sesos, then I will do it.

I shall wait for half an hour before I start so as to ensure that nobody is going to do it.

At 15:00 UTC I will post my solution if nobody does it.

TimmyD

Why program in Sesos when you could program in PowerShell?

Rohan Jhunjhunwala

Why program in Sesos when you can use SILOS

TùxCräftîñg

^

Leaky Nun

@RohanJhunjhunwala Because you didn't?

TimmyD

Ahahaha ... we're implementing a new application, and it has the ability to email documents into the application, which then parses the order number in the email and attaches files correctly. And the people responsible for the application integration decided to use 5-digit numbers with no accompaniments for the order number ... postal codes in the US are also 5-digit numbers.

They asked for a rule implemented on Exchange that parses emails containing a 5-digit number and sends a copy to their application. I (barely) refrained from laughing and told them "No."

Geobits

2:41 PM

You should have done it. It would be worth the small amount of work involved to see them hastily ask for it to be undone.

El'endia Starman

@TimmyD So, if you had, they would've gotten a copy of all emails with a zip code?

TùxCräftîñg

even calculating a factorial is hard in SILOS :/

TimmyD

Pretty much. And just think of how many corporate business folks have their full address in their signature. I didn't want to a) swamp the poor service account's mailbox trying to process this, and b) have my boss yell at me even though I was doing what they asked.

Geobits

Well clearly they want you to filter out signatures and other non-order-number five digit numbers >_>

Rohan Jhunjhunwala

@LeakyNun yeah, Sesos is an interesting langugage. It's hard though

@TùxCräftîñg do you wish for me to send an implementation?

TimmyD

2:47 PM

Yeah. I think they're now talking about year-state-order, like 2016-MN-12345 which is quite a bit more unique.

TùxCräftîñg

no

Rohan Jhunjhunwala

@TùxCräftîñg it's a pain mainly since there is no for loop, but FOr loops are all GOTO's at the byte code level

TùxCräftîñg

what is the convention for function arguments and return value?

Geobits

That makes more sense I guess. Using just five digits for an order number seems kinda Y2K-ish to me anyway, unless they really don't get that many orders (or they're strictly temporary, which seems weird).

El'endia Starman

CMC: Dissect an obtuse triangle into five acute triangles. I solved this puzzle two or three years ago and was reminded of it last night.

TimmyD

2:50 PM

@Geobits I think that's the key. That department, individually, may not get that many orders, but we have several thousand people in the company across the US, and I don't think they were thinking outside their little domain.

Rohan Jhunjhunwala

@TùxCräftîñg I use N as an argument and like the first letter of the function as a return value

TùxCräftîñg

k

Geobits

@El'endiaStarman So.. draw four lines from the obtuse angle's point to the opposing side?

El'endia Starman

@Geobits Are you sure that all of those triangles are acute?

TimmyD

@Geobits No, those aren't acute.

Rohan Jhunjhunwala

2:52 PM

Really?

Geobits

Ooooh, I see.

Rohan Jhunjhunwala

rip my geometry skills. I don't even use arithetic in calc class

@TùxCräftîñg I have a short implementation of reading an int from the stdin (cla for TIO) and outputting its factorial. It's not that long... (don't worry I wont send it to you since it seems like you don't want the fun spoiled).

TùxCräftîñg

> the fun

Rohan Jhunjhunwala

@TùxCräftîñg :D

Geobits

A quick googling seems to show that you can't do all acute with just five for some obtuse triangles.

TimmyD

2:56 PM

If you take all the altitudes, you have a guaranteed right triangles, but there's six of 'em.

TùxCräftîñg

wow

SILOS source code really need refactoring

Rohan Jhunjhunwala

@TùxCräftîñg ping me if you need any help, the documentation on GitHub and Esolangs should be accurate, but let me know if there are any issues :D.

Sp3000

You can't have a single segment intersecting the triangle's edge, since that'll split one angle to >= 90. So you'd need at least two segments, sort of like if you joined all the midpoints of sides to form a triangle (V shape touching the edges), but of course that does nothing to the original obtuse angle

Rohan Jhunjhunwala

@TùxCräftîñg just like the code written in SILOS is spaghetti, the interpreter is spaghetti

TùxCräftîñg

documentation? useless. i have the source code.

Rohan Jhunjhunwala

2:59 PM

@TùxCräftîñg haha... ok, let me know if you have any suggestions to greatly improve readability. Feel free to pull request or ping (or raise an issue).

TùxCräftîñg

i will PR soon to make this less spaghetti

Geobits

There's a method for seven acutes by creating a pentagon around an inner node like this:

Rohan Jhunjhunwala

@Geobitscan you prove that formally?

@Geobits I'm sure that works all the time

El'endia Starman

Now I'm wondering if it was supposed to be 7, not 5...

Geobits

Supposedly Gardner has a proof. Looking for it now.

[It is easy to see that seven is minimal. The obtuse angle must be
divided by a line. This line cannot go all the way to the other side,
for then it would form another obtuse triangle, which in turn
would have to be dissected, consequently the pattern for the
large triangle would not be minimal. The line dividing the obtuse
angle must, therefore, terminate at a point inside the triangle.
At this vertex, at least five lines must meet, otherwise the angles
at this vertex would not all be acute. This creates the inner pentagon

Stupid broken markdown crap

TùxCräftîñg

3:06 PM

@RohanJhunjhunwala github.com/rjhunjhunwala/S.I.L.O.S/blob/master/Silos.java#L31 what is prompts?

TimmyD

@Geobits Looks good to me

TùxCräftîñg

oh nvm found

El'endia Starman

@Geobits Yeah, I probably meant 7.

Geobits

I can see where you'd get the 5 from though :P

El'endia Starman

Yeah... :P

Rohan Jhunjhunwala

3:16 PM

prompts is true if its running on a desktop @TùxCräftîñg

Syxer

Ugh - I'm becoming increasingly annoyed by libraries with unhelpful names.

Rohan Jhunjhunwala

@TùxCräftîñg on TIO prompts is false and input is passed through command line arguments

Syxer

No, like, libr

WHAT IS R

TùxCräftîñg

a language?

Syxer

Nah this is a submodule for Android

I somehow doubt that it has native R.

TùxCräftîñg

3:17 PM

wat the heck

Geobits

As in resources? Like R.android.something?

Syxer

Lemme find the line

Geobits

Or R.layout.my_layout

Syxer

Oh wait sorry

It's a command on Android

TimmyD

It's the 18th version. They've already used liba through libq

Syxer

3:19 PM

Install: /home/vtcakavsmoace/android/system/out/target/product/huashan/system/bin/r

It also had a lib install but I can't find it.

Still begs the question: why

:O IT IS THE LANGUAGE

Apparently either AOKP or CM decided it was a great idea

Geobits

Oh, well if you're using non-standard Android I don't know why :P

Syxer

Yeah - the stock ROM for my device isn't available anymore, sooo...

New Sandboxed Posts

0

A: Sandbox for Proposed Challenges

I'm sure this must be a duplicate but couldn't find it. What's My Name? Write a program or function such that when it is ran it outputs the name of the language. Input There will be no input. Output The name of the language that the code is ran in. Specs The name cannot be hardcoded or s...

Syxer

@ConorO'Brien no

Geobits

@Syxer Oh, well if you're just developing for your own device it's no big deal.

Syxer

3:33 PM

@Geobits Yup. :P

To be completely honest, I'm just building a modified kernel, but I keep having to modify kernel source because the version that actually works isn't published yet. ._.

But it looks like this one is actually building now. It's in the packing stage.

Geobits

Ah. I haven't bothered playing with modified kernels yet. Good luck :)

TimmyD

@ConorO'Brien Wow, just stumbled across this answer of yours on the "Happy Birthday" challenge. Really neat GIF!

Syxer

YES

#### make completed successfully (34:36 (mm:ss)) ####

oh man that was a ride

I've spent the last 12 hours on that oml

It would always fail about 30 minutes in and I'd die internally

Nathan Merrill

was ColdGolf chat banned?

I haven't seen him around

El'endia Starman

@NathanMerrill Not permanently.

Rohan Jhunjhunwala

3:43 PM

I have never seen him around

After that whole thing

Nathan Merrill

@El'endiaStarman temporarily?

El'endia Starman

They're not suspended at the moment either, but yes, I did notice that they disappeared.

Nathan Merrill

bah, we really need a third person genderless pronoun

Rohan Jhunjhunwala

@TùxCräftîñg how is coding in SILOS? Going good?

Syxer

@NathanMerrill Ze

Sp3000

3:45 PM

Singular they like El'endia did is what I usually do

Syxer

^ That is definitely more common.

Dennis

@Adnan Done.

Nathan Merrill

I usually use "him"

TimmyD

Singular they ftw

2

Nathan Merrill

@Dennis @Dennis you need to start measuring the time you spend pulling :)

er, woops, double ping

Syxer

3:48 PM

Interestingly high number of new users in this post.

Have there been an increasing number of actually good-posting new users?

Nathan Merrill

@Syxer The title is catchy, the post is easy to read, and the task is simple

Dennis

@NathanMerrill Hopefully, that will soon be a thing from the past.

Nathan Merrill

wait, 2.0 will automatically pull?

Syxer

@Dennis Use webhooks. :P

TimmyD

@Syxer It also hit HNQ

Nathan Merrill

3:49 PM

you're going to put your server on the sandbox?

@Syxer I'm not sure if the rate of new users is increasing. We are definitely growing, but new users are becoming old users

Dennis

@NathanMerrill All interpreters will be sandboxed, so running arbitrary user code won't be an issue. For interpreted languages, it will be easy to update them automatically. I still have to figure out how to do that for compiled ones though...

Rainbolt

@Syxer I don't see any users in the post. Did you mean commenting on the post?

TimmyD

@NathanMerrill I'm 37, I'm not old!

Rainbolt

Oh, my chat was scrolled up so I didn't see like ten messages. You guys are talking about the question lol

Ignore me

Nathan Merrill

@Dennis For compiled languages, you need to have a list of commands, and only pipe with the last one.

Syxer

3:54 PM

@Dennis yay chroot

Nathan Merrill

at least, that's what I do with KoTHs

Syxer

@Geobits By the way, Linux chroot on Android works really well.

Dennis

@NathanMerrill No, the problem with compiled languages relates to permissions. I need to create a context in which the files of an interpreter can be modified persistently (to allow compilation) without giving write access to the rest of the system.

@Syxer Chroot is not meant as a security feature, and isn't useful as one. Since only root can chroot, it's usually less secure to chroot an untrusted application than to simply run is as an unprivileged user.

Nathan Merrill

@Dennis how do interpreted languages write to files? or is that not supporteD?

Syxer

@Dennis No, I mean chroot combined with an switch user inside the chroot to a user restricted to commands placed in the chroot.

Dennis

3:58 PM

@NathanMerrill Not in a persistent manner. They can create temporary files, but everything is wiped when the program finishes.

Transcript for