The Nineteenth Byte

feersum

21:03

Wow, 7 minutes.

I doubt the tmp lang was already on there as I only used it a couple times and it's distinctly half-finished.

@TuxCopter To emulate variable bindings, you have to pass them as an argument to an anonymous function, since that's the only way to access a previously calculated value.

It's a pure functional language.

And if you need to access the nth character of a string, you don't.

confusedandamused

Can anyone explain why you would use HttpClient vs HttpWebRequest vs WebRequest in c#? I'm trying to figure out the main differences

Anonymous

@confusedandamused HttpClient is a client class that sends HttpWebRequests. WebRequest is a superclass of HttpWebRequest - there's also FileWebRequest (for file:// URIs) and FtpWebRequest (for FTP requests).

AdmBorkBork

Holy cow. That answer-chaining polyglot now has C as well.

feersum

If Dennis sees the CPU usage generated by the compiler for Templates Considered Harmful, he's going to regret adding it.

DJMcMayhem

@AdmBorkBork C and C++?

AdmBorkBork

21:17

Yep

back-to-back, even

feersum

There are a lot of ways to do it.

It's even possible to polyglot different C compilers without using the preprocessor.

DJMcMayhem

I feel really bad about adding V so early on. The interpreter is horrendously slow and it's screwing everyone over

Anonymous

@DJMcMayhem Well make it faster! :P

DJMcMayhem

I can't. :/

Maybe there are some settings I could play with tough

feersum

You should totally report this issue to Vim developers.

The age of robots is coming and text editors need to be ready for them.

DJMcMayhem

21:22

It's actually neovim

AdmBorkBork

@orlp Your hypercube challenge hurts my brain.

TuxCopter

printf("%d\n", ([](int args[]){return (args[(1 - 1)] + args[(1 - 1)]);})({1}));

Why isn't this working

Riker

majik

confusedandamused

@Mego Thanks mego! Just looking into web services now :)

Anonymous

Glad to help

Anonymous

21:35

@feersum For TMP, would it be possible to add a method of embedding Sts in Sts such that the St contents are inserted at that point? For example, in Hello World, instead of using 32, I want to be able to use SP.

orlp

@AdmBorkBork why?

feersum

@Mego I guess SP could be defined to some magic number outside the range of char, which is then substituted with 32 somehow.

AdmBorkBork

@orlp I try to visualize it in my head.

DJMcMayhem

Is anyone else having issues with TIO? Everything is timing out for me

feersum

What am I saying. It could be defined to 32.

Maybe it's Templates Considered Harmful absorbing all the computrons.

Anonymous

21:38

@feersum That's not exactly what I meant. For another example, St<'f','o',St<'o','b','a'>,'r'> should evaluate to "foobar"

feersum

@Mego No, you can't have a template argument that can be both a type and an integral value.

AdmBorkBork

Does anyone else read "Automate your first grade counting exercise" as "Automate your first (grade counting) exercise" rather than "Automate your (first grade) counting exercise" or is it just me?

DJMcMayhem

It's not just you, I thought the same thing

Anonymous

@feersum Hmm. Yeah, that's a problem. It could be resolved by using std::string instead of char, but then you lose the nice ability to use ordinals.

DJMcMayhem

This user has 75 posts in 16 days. o_O

feersum

21:43

@Mego std::string can in no way at all be used in template metaprogramming. It's a class that allocates heap data.

Anonymous

Oh, right. Gah. C++ makes everything difficult.

orlp

hey

I know quite a bit about C++ and template metaprogramming

what are you trying to solve?

feersum

Nothing really, just discussing.

Riker

@DJMcMayhem is it bad I assumed it was kenorb

AdmBorkBork

@orlp I think they're discussing Templates Considered Harmful?

Pavel

21:47

Is it just me or did TIO get a bunch of new esolangs in the past few hours?

orlp

@Pavel ask @Dennis

Dennis

I don't think two qualifies as a bunch.

Pavel

Ok, just a bunch I hadn't noticed before.

Mistah Figgins

Like?

orlp

22:10

am I really that weird for refusing to post OEIS links, formulas, algorithms, etc for my challenges?

everytime I do it I get a bunch of commenters asking for those resources

I intend for my challenges to actually be a challenge

xnor

it's not a challenge to plug the sequence into oeis yourself

it's just tedious

Pavel

@orlp Lots of people come here for the challenge of golfing their code, not the challenge of figuring out what their code should do.

Dennis

@orlp There's no need to spoon-feed people everything they're going to need to solve a challenge, but if there's a formula on OEIS that pretty much solves the problem, you might as well include a link.

ais523

@Poke I'm an admin but I have no control over the hardware, all my powers work via the site interface so they don't work when the site's down

(this is much the same concept as Dennis not being able to do anything about Stack Exchange going down, when it goes down)

Pavel

They don't need Dennis for that, they get Jon Skeet to do it.

Dennis

22:21

Do you happen to know what is going on?

xnor

@orlp Maybe we need a meta post to discuss this, since I'm also often unsure how much help to give.

orlp

@xnor I don't really agree

xnor

I think some people would have more fun figuring stuff out, but i worry the less mathematically inclined people will feel exlcluded.

orlp

I think challenge authors should be free to decide in how much/little help they want to give

xnor

I'm looking only for personal guidance, not a site policy.

Dennis

22:22

Plus, once the first good answer has been posted, the cat's out of the bag anyway.

ais523

let me check the logs of Esolang's chat channel

> <int-e> The VM hosting the wiki seems to be down... fizzie might know more, but perhaps it needs divine intervention from Gregor :/

orlp

@Dennis sometimes

ais523

> <fizzie> I don't really have any visibility to the hosting provider, so yes, I think a Gregorvention is required. Or maybe it just comes back up naturally.

orlp

but oftentimes I see that it takes a bit longer for the first answer to pop up

but quite often

during that time people have found independent methods

xnor

I think the hypercube one isn't going to have a nice formula, which is good, but I'd complain a bit about Dennis's alternating power-of-two fibonacci one.

ais523

22:23

wow, Gregor actually turned up too, he's hardly ever there

orlp

so when other people finish their answers there are multiple methods

ais523

so it must be quite a major issue

orlp

@xnor just so I'm not mistaken about this

it's possible to evaluate the integral so that you're 100% certain that the digits you've acquire are correct, right?

I'm not trying to make an impossible challenge here

ais523

anyway, it seems that Gregor decided that he doesn't have permissions to fix the issue either

xnor

@orlp yes, absolutely

ais523

22:24

so he's gone and contacted the owner of the physical machine to ask what's going on

orlp

(that was my goal with the challenge)

(to force people to come up with numerically correct integration for this integral)

there's a bunch of interesting identities on that mathworld page

xnor

i think this challenge is fine because there's no nice formula, and so it looks like you have to do something complex that can be optimized

ais523

@orlp it's always possible to be certain that the first n digits of an answer to that sort of problem are correct, unless the digit after is 0 or 9

Pavel

someone solve this so I can get internet points

xnor

@orlp i think the requirement on certainty will likely move answers though to iterating over all pairs of point in a finite mesh, unless intergration allows a nice formula

Pavel

22:28

One day a Novice came to the Master.
 "Master," he said, "How is it that I may become a Writer of Programs?".
 The Master looked solemnly at the Novice.
 "Have you in your possession a Compiler of Source Code?" the Master asked.
 "No," replied the Novice. The Master sent the Novice on a quest to the Store of Software.
 Many hours later the Novice returned.
 "Master," he said, "How is it that I may become a Writer of Programs?".
 The Master looked solemnly at the Novice.
 "Have you in your possession a Compiler of Source Code?" the Master asked.

orlp

@xnor That would be fine too

DJMcMayhem

@Pavel a link would suffice

That takes up my entire screen

Pavel

It collapses large block though

Mistah Figgins

Mobile?

ATaco

Yeah but many brow- That

Pavel

22:31

And... just as I tried to delete it it timed out.

Sorry about that.

DJMcMayhem

Dec 24 '16 at 17:23, by DJMcMayhem

I think for me, a good rule of thumb is that if I'm on chat, I'm probably on mobile.

:P

Pavel

I thought it collapsed it on mobile too.

Mistah Figgins

Yeah, but small screens

Cyoce

Ruby... why isn't .. an operator with a method? You're making my life hard

DJMcMayhem

@Pavel It does, I just think large text blocks/oneboxes are kinda unnecessary

Pavel

22:35

@Cyoce Use python

DJMcMayhem

Although it's definitely worse on mobile because there's no scrollbar

Cyoce

@Pavel Eww. And Python doesn't have __range__ either iirc?

orlp

@Cyoce what's wrong with Python?

Pavel

It has range(n,m)

Or [n..m]

Cyoce

@Pavel Ruby has n..m, it just can't be overridden by the class of n like other operators can

Pavel

22:38

Oh, yeah, I don't think python has that.

Cyoce

@orlp its syntax is too verbose for an interpreted, dynamic-typed language, among other things...

orlp

@Cyoce huh

Pavel

It's not all that verbose. Besides, have you seen perl? That stuff's messy.

orlp

the only unnecessarily verbose part I can think of in Python is self. in classes

and that's precisely due to being such a dynamic language

Pavel

Also global vars.

orlp

22:41

those are intentionally verbose

because mutable globals are a code smell

Pavel

A method can't say foobar=5 if foobar is global, it has to say global foobar;foobar=5

orlp

I virtually never use global

Cyoce

@orlp __add__ and the like are annoying, half its operators are English words, and (unrelated) it doesn't have a built-in regex type without an ugly import and sticking re and r"..." into everything

Riker

you don't need the r

r is raw string

it's helpful for regex but not needed iirc

Conor O'Brien

you don't need it but it sure is helpful

orlp

22:43

@Cyoce __add__ is again due to the dynamic nature of Python

Cyoce

It's even uglier to use it without r

Conor O'Brien

still verbose either way >_>

Dennis

@Pavel Of course it can say foobar=5, it just won't modify the global variable. That is what's supposed to happen.

Pavel

I haven't even heard of r"".

orlp

the fact that its operators are English

Riker

22:43

@Cyoce what would oyu suggest doing instead?

orlp

actually reduces verbosity

Riker

raw backslashes?

orlp

what is faster to type, and or &&?

Conor O'Brien

equally fast imo

Cyoce

@orlp I don't have a problem with operator overloading, but __add__ is unnecessarily verbose. Ruby just lets you do def + (other)

Riker

22:44

yeah, that would be nice

Cyoce

@Riker /regex/.match

orlp

I'll concede that Python's regex library sucks though

I don't think it should be a builtin type

Dennis

That it does.

Pavel

def __add__(self, other):

orlp

but the library could've been so much better

like whenever I use it I end up looking things up

because it's... weird

Dennis

22:45

waffles

Pavel

I think Java's regex support is actually exactly what anyone could have wanted.

Riker

waffles are gut

~~I might have read something though~~

Mistah Figgins

We all ~~know~~ have no idea what happened here

Pavel

I don't, I've been trying to install Acrobat Reader and it keeps erroring on me.

Apparently it relies on a package that's uninstallable.

Riker

that's your problem

adobe

@MistahFiggins --- is the markdown for strikethrough on SE chat

Pavel

22:48

What should I use for PDFs?

Riker

---strike---

Mistah Figgins

Gotcha

Riker

@Pavel don't

Dennis

The last usable version of Acrobat Reader was 9 or something. It got worse with every subsequent release.

Conor O'Brien

@DJMcMayhem >_> a warning about the content of the link would have been appreciated. it's rather gruesome

Pavel

22:48

I have to open PDF's somehow.

orlp

Sumatra PDF

Pavel

Thanks

Also, according to the wine blog, it says it supports office 2013 now, but the installer tells me my Windows is too outdated >_>

Riker

i.imgur.com/Xs5SDLd.gifv

Mistah Figgins

@Riker Must be really fun for the people on the outsides...

Riker

did you watch til the end?

Mistah Figgins

22:55

Yea, was unexpected

by definition

DJMcMayhem

@ConorO'Brien ._. sorry

Given an array of non-negative integers, determine if exactly one of the elements is greater than 0.

Mistah Figgins

isn't that "determine if the length is 1"?

DJMcMayhem

Sorry, I meant non-negative

Mistah Figgins

Ah

DJMcMayhem

Silly brain

Riker

23:02

distinct ints?

DJMcMayhem

Not necessarily

Riker

ok

Mistah Figgins

compare the sum to the max

if you have short built ins

xnor

@orlp Is there any bound on the input? For a large input, the output is a large float which cannot have 5 digits of accuracy because of float precision.

DJMcMayhem

Crap, you're right

Mistah Figgins

23:04

maybe check if the # of positives is a certain number, given as input

xnor

@orlp one would need to use some infinite precision datatype or somehow generate digits

DJMcMayhem

How about check if every element is either zero or equal to the number of non-zero elements?

orlp

@xnor if your answer fails due to floating point overflow it's forgiven

Riker

@DJMcMayhem hm

orlp

(as a rule of thumb for my challenges, if you solve a discrete problem with floating point numbers and your answer fails because of it, it's not allowed, but if the problem itself is continuous in nature and calls for floating point it's allowed)

e.g. if I ask you for the floored square root of an integer then int(n**0.5) is wrong, but if I ask you for the average of two real numbers (a + b)/2 is fine even though a+b might overflow

Pavel

23:08

I once tried to check if something was a power of two with (int)Math.log(a,2)==Math.log(a,2) (or something like that, I don't remember how log works in java)

Riker

@DJMcMayhem long solution, still golfing: def f(x):y=filter(lambda n:n!=0,x);print len(y)==y[0]

orlp

@Pavel n&(n-1)

or n&~-n

Pavel

Which would give 0 if it's a power of 2, correct?

orlp

@Pavel yep

although it also gives 0 for 0

DJMcMayhem

@Riker lambda n:n

Riker

23:10

oh true

DJMcMayhem

No, even better: int

orlp

@DJMcMayhem huh

Riker

@DJMcMayhem true

DJMcMayhem

lambda x:len(filter(int,x))==max(x)

orlp

@DJMcMayhem what exactly is the challenge now

you posed two

Christopher Peart

23:11

I wanna do smth

Pavel

n->{int a=0;for(int i:n)a+=(i==0)?0:1;return a==1;}

Riker

lang?

js?

Pavel

Java

DJMcMayhem

@orlp the second

@Riker I outgolfed you by 18 bytes. :P

Riker

ye

DJMcMayhem

23:13

Wait mine is invalid

And so is yours

Riker

why

orlp

@DJMcMayhem is not correct

Riker

@DJMcMayhem what's wrong with it

wat

kunaki.com/sales.asp?PID=PX009XQC1L

orlp

lambda l:set(l)<={0,sum(map(bool,l))}

Riker

23:15

damnit orlp

42 bytes

DJMcMayhem

@Riker [0, 2, 1] should be false

Riker

only 5 longer htan orlp

@DJMcMayhem ah I see

wat

CMC: ->s,t,u{s.gsub t,u}

orlp

@wat what is gsub?

what language is this?

wat

in your favorite language

orlp

23:17

that's no way to ask a challenge

wat

this is ruby

Pavel

n->{n.removeAll(0);return n.length()==1;}

I think that's how removeAll works.

orlp

@Pavel [2, 2] is supposed to be true

:35159784 no, because of the potential zeroes

Maltysen

oh i see

Pavel

Eh? What's the challenge?

wat

23:18

irb(main):001:0> ->s,t,u{s.gsub t,u}["Hello world", " ", "watwatwat"]
=> "Hellowatwatwatworld"

orlp

CMC (by DjMcMayhem): Check if every element is either zero or equal to the number of non-zero elements

Pavel

Ok

orlp

eh nope

sec, fixing my answer

[0, 0, 2, 0, 2] is false

Pavel

n->{n.removeAll(0);for(int j:n)if(j==n.length())return 1<0;return 1>0;}

orlp

lambda l:set(l)=={sum(map(bool,l))}

@DJMcMayhem the CMC can be simplified actually

Check if every element is ~~either zero or~~ equal to the number of non-zero elements

DJMcMayhem

23:20

Oh yeah

Wait no

0 =/= 2

orlp

consider [0, 0, 0, 0]

every element is equal to the number of non-zero elements

Pavel

Consider [0,1]

DJMcMayhem

Do you mean "Check if every non zero element is equal to the number of non-zero elements?"

orlp

oh wait my earlier answer wasn't wrong

zz I'm confusing myself

lambda l:set(l)<={0,sum(map(bool,l))}

Pavel

23:35

f[x_]:=(x=x~DeleteCases~_0;AllTrue[#==Length@x&/@x,#==0<1&])

That feels waayyy too long.

orlp

@feersum I'm a bit confused, why can an arbitrarily small value change the value of the first 5 decimal digits if we round first?

feersum

@orlp .999985000000001 or .999984999999999 would have different values after rounding to 5 places.

Christopher Peart

23:50

My chem teacher is insane

wat

@ChristopherPeart What did he/she/they/whatever do?

orlp

@feersum right

Pavel

Guys, my desktop borked. It's showing a bunch of random colors, and changes in response to me moving windows around. Any ideas?

wat

@Pavel install arch linux

Pavel

Why would I do that?

wat

23:55

buy my disc

Riker

because it's cool

Pavel

You keep saying that, but I've yet to hear one reason.

wat

It

It has a very large potential for customization (you build out your system the way you want it to be) and also there's a huge software repository called the AUR

Pavel

@wat I can create a bootable USB myself, thanks.

wat

@Pavel D:

Pavel

23:57

I mean, really, how many people have bought that?

Ok, uninstalling the NVidia driver and grabbing an open-source one fixed the background issue, but it didn't remove NVidia settings

Christopher Peart

@wat he had us make a explosive on our desks no goggles then we went to the parking lot and made loud noise by blowing up the explosive my ears were ringing. Then we played catch with fire balls made of tee shirt and lighter fluid

wat

...........

<--

Pavel

@ChristopherPeart You have an awesome teacher.

Transcript for