The Nineteenth Byte

@Mego My Python 3 version can only be executed with "python" from the command line

So when I install Python2, it conflicts

Oh okay let me see that

Anonymous

You can have two virtualenvs living in the same directory, and activate either one of them, which overrides the python executable until you deactivate it

@Mego Well that's nice

I'll try it out

Virtualenvs are vital.

oh

12:09 AM

It's not just the Python executable that gets overridden, but all the libraries too.

I think I have a dozen virtualenvs on my work machine and I use three of them with some regularity.

oh

Anonymous

Also, you can have Python 2 and 3 coexist - just use python2 to invoke 2.7

i have never used a virtualenv

okay

which means you should use them

because i am a bad perisn

12:14 AM

oh

12:29 AM

@El'endiaStarman I've found I'm okay without them, I don't think I've had a package conflict with pip like, ever

12:44 AM

Certainly, you can get by without virtualenvs for a while. But if you start maintaining at least two major projects that don't share libraries or especially if they require different versions of those libraries, then you pretty much need virtualenvs. They also make things a lot easier anyway.

Downgoat

@quartata how are you still sane having made things in python without virtualenvs

Last time I tried to deploy a flask project to ubuntu box thinking I could get away without virtual envs it ended in a new ubuntu box :P

1:05 AM

@Downgoat See there's your problem

Ubuntu

Downgoat

debian and centos are not better either

Anonymous

Debian is great if you're eager to try that new C++11 thing

@Mego or python3 to invoke 3.x lol

Anonymous

1:26 AM

That reminds me - I need to install 3.7.2

1:55 AM

CMC: Given an integer n as input, output the kth prime such that the difference between the kth and (k +/- 1)th prime is at least n

2:06 AM

@Downgoat its really never been a problem

i mean

this is for personal stuff

if im given a pipenv to work with ill do it

but i dont default to it

i do keep pretty careful track of dependencies because im not always guaranteed of having pip to wherever i deploy

(dont ask)

@quartata No I will ask anyway.

Pip comes with Python tho???

some people like sticking to what redhat provides in their base repos

one person in particular

no names

rhel doesn't have pip?

no no i mean

the python packages

Ah

2:10 AM

like pymysql you can just get from yum

i suppose that is one way to solve dependency conflicts right

Would be nice if RH/Fedora packaged more npm packages

Since those actually are fukn impossible to install properly with npm

i converted to yarn a while back

JF convinced me

its just way faster

@El'endiaStarman I'd like to know the one person in the universe who calls it "c-hashtag"

Anonymous

Lunar eclipse will be starting soon for the western hemisphere - about 10 minutes until it begins, and about 2 hours and 45 minutes until totality

2:25 AM

@Mego Oh, sweet, thanks for the reminder! Isn't this the blood one

honestly lunar eclipses are just kinda meh

ooh, moon is gone now, like that doesn't happen every month anyway just because of its phase

This one is supposed to make it look like a "blood moon"

i.imgur.com/flqrRMP.png <-- can someone explain this what's a section

@Quintec C She

@Pavel making fun of how modern web design often uses CSS classes for semantic meaning

that is, you have a perfectly good HTML tag called section

but instead youve got a generic div with the class section

(although section was also added in HTML5 so its kinda weird)

Anonymous

2:32 AM

@Quintec Yep

The project I'm working on has a backend dev (me) and like 5 frontend devs (extremely not me) and all the crap the write is just some arcane bullshit

Anonymous

@quartata tbf section and div are usually identical

would have been funnier with <article>

Occasionally I look over and see shit like <div><div><div>...</div></div></div> with different classes and go back to just feeding them stuff from database

html sucks

which is why we created an entire ecosystem for making desktop apps out of it

2:35 AM

@quartata I feel like the article tag could be useful for like RSS readers and whatnot

Anonymous

@Pavel I'm surprised they're not using JSX and SCSS like the rest of the modern web dev world

Anonymous

Actually, I'm not surprised - we don't use either at my company, either (yet)

@Mego We considered it before realizing it's a moot point since none of us know how to work that into our build systems

Where by build systems I mean there isn't a build system and it's just flask run

Anonymous

Uhh just using flask run is really bad

@Mego The collective total flask experience between us is that one time I made a few commit to Axtell at the very start

2:39 AM

@Mego Please elucidate; that's pretty much the only way I know to run Flask. Or do you mean the problem is using the built-in webserver?

So please enlighten me

We're using Flask because Python was the language the most of us had in common after JS and I veto'd node

Anonymous

@El'endiaStarman The latter, yeah. Werkzeug's WSGI server is not meant to be used in production, because it's not performant. It's only included as a dev tool. A proper WSGI implementation (uWSGI, Gunicorn) should be used, ideally in conjunction with a proxy server (Nginx, Apache)

@Pavel Django is another Python option.

@El'endiaStarman Yeah but that drops net experience to 0

Anonymous

For example, Axtell's web stack is nginx/uWSGI/Flask

2:41 AM

How do you tell flask to use something other than Werkzeug?

@Mego Good to know! I'll be creating a new website soon using Flask where my prior experience was with Django, so I'm glad to know what to google when the time comes. :P

Anonymous

@Pavel You want to use Werkzeug, because that's the part that does the routing. You use another WSGI server instead. This guide was invaluable when I was setting up Axtell.

i hate mod_wsgi

it used to crap out for me all the time in development and i will never know why

I'll be honest we almost went with cgi-bin

luckily theres people out there better at it than me

@Pavel lol

2:47 AM

@Mego Yoink! (Although the webserver I have access to uses CentOS IIRC.)

Anonymous

@El'endiaStarman Shouldn't be too different

Anonymous

digitalocean.com/community/tutorials/…

@Pavel Bash or Python? :P

Anonymous

@Dennis nasm

Python, but I couldn't tell you which is a worse idea

Well, it's not like we're building Axtell we're literally just replacing a wordpress blog because we decided fuck paying for wordpress

So we can probably be ok without best practices

And just do it the easy way

2:54 AM

A plain Python script (without imports) has a start-up time of 50 ms, so it's definitely not the better choice.

Guess it's good thing we did not that

I was caught off guard by an incredible answer to an old challenge:

1

A: An abundance of integers!

Regex (ECMAScript flavour), 1085 855 597 536 511 508 bytes Matching abundant numbers in ECMAScript regex is an entirely different beast than doing so in practically any other regex flavor. The lack of forward/nested backreferences or recursion means that it is impossible to directly count or kee...

2

You could have done worse. Like not deleting one of the three cgroups your script created and bring down a couple of servers.

H.PWiz

I wouldn't describe that user as new

My current greatest mystery is why it takes so long for this one particular link to load in on the main page when there's literally no JS

2:59 AM

@DJMcMayhem why would v be hard though

It's just, you load the page, and a second later the link pops into existance

What link?

Anonymous

@Pavel Sounds like a job for Chrome Dev Tools

It's just a bare <a> with the colors set

@H.PWiz Oh yeah, I only saw that they had 4 answers and not a lot of rep, I didn't pay attention to how old the account was

3:00 AM

@DJMcMayhem every single one of their posts uses regex as language btw

@ASCII-only Oh gosh, where do I start

@Pavel userscript/userstyle possily? or long link

Anonymous

@Pavel Can you share the link to the page?

It's not like an image that you link no it's just a fucking <h2><a href="/blog">Blog!</a></h2>

@Mego Here: https://localhost:5000 :P

It's not working.

3:02 AM

Oh wait, my bad

http://localhost:5000

We don't have ssl yet

@El'endiaStarman re: the webpack one, i'm the opposite, i'd prefer if glitch didn't reload after i stop typing for like... 1 second because my brain blanks out for a second

@Pavel oh hi phase

Anonymous

@Pavel Har har har

@Pavel maybe you have 1. a potato router or 2. a potato computer

@Pavel Any scary CSS that applies to this link and nothing else?

oh hey someone updated it and now it's all the text on the page and not just the link

3:05 AM

Progress!

On the other hand, the pop-in time went down to like a couple frames

@El'endiaStarman also :| why don't they git commit -am

@Mego Thanks!

Anonymous

@El'endiaStarman You're welcome. Good luck!

wait a second, light IDE themes really are bad, but... why do we put up with light SE and light chat :(

Anonymous

3:09 AM

@ASCII-only It was requested in 2011

Newly Featured Questions

26

Q: An abundance of integers!

An Abundant number is any number where the sum of its proper divisors is greater than the original number. For example, the proper divisors of 12 are: 1, 2, 3, 4, 6 And summing these results in 16. Since 16 is larger than 12, 12 is abundant. Note that this does not include "Perfect numbers", e...

@ASCII-only Because there's a huge intersection between people that use SE Chat and people that know how to apply their own CSS to pages

SE: we'll be done in 6-8 ~~years~~ decades

@Pavel i guess. but it's really annoying when you're on another computer and your dark theme is gone

plus, there are people that can apply their own css, sure. but how many people really have enough time to make it look decent?

remembers previous TNB description oh wait

this one is me with Charcoal's parser. also the fill algorithm before i looked up a more efficient one (if you're wondering how bad the fill algo was, rewriting it brought unit test time from like 11 seconds down to 3 iirc)

> defining your comments as variables

looking at you, Python

also. that one meme where you can see git not installed :|

if anyone hasn't muted and moved chat to another desktop from this wall of chat:

~~say you have set_property foo [bar baz] quux as a function (that's half symbolic i guess?). what type would the second parameter [bar baz] be~~ wait nvm. have an idea

if anyone wants to give feedback, it's here. be warned though, it is in JS, and very very all over the place

3:38 AM

@MilkyWay90 i put some detailed notes in a wiki

chat room for the new cicada challenge

4:27 AM

@DJMcMayhem half a decade in the making

what on earth

Anonymous

4:59 AM

Moon is real red outside

Sherlock9

Ah yes, the lunar eclipse tonight

Complete other side of the earth from my hemisphere, but I hope the Americas have fun

5:12 AM

@Mego I did not even know about this until now

I missed the peak, but it's still purty

Anonymous

@DJMcMayhem It's super cool

And there's not a cloud in the sky (here)

Sooo if there were people living on the moon, they would be experiencing a solar eclipse right now. And if we were experiencing a solar eclipse, the moon-people would have a... "Geolar" eclipse?

Anonymous

It would probably be called a terrestrial eclipse

5:29 AM

Oh yeah, that's way smarter

7:20 AM

@ASCII-only wait nvm, since it's basically variadic/there's no set length for the array. is there any nicer way than this? (below)

generic
  t
  function set_property [
    value_of_type identifier_node name
    conditional_argument # placeholder name
      value_of_type array_literal_node properties
      all
        properties
        lambda [property] is_instance property identifier_node
    value_of_type t value
  ] [
    # body here
  ]

feersum

What is it supposed to be?

just do foo.bar.baz = 1, in a prefix language. set_property foo [bar baz] 1

feersum

But, the stuff. Is it supposed to be sample code in <some language>? A grammar? Code in the compiler?

mostly sample code, i guess?

feersum

I don't get it, oh well.

7:29 AM

in a normal language it'd be:

function setProperty<T>(
  name: IdentifierNode,
  properties: ArrayLiteralNode
    where properties.children.all(child => child is IdentifierNode),
  value: T
) {
  // function body
}

also @anyone what conditionals/branching/looping should be in the core of the language (as in, the part used to create the standard library)? goto+if? while? for_each? should C-style for loops be in the core (they are no more than sugar for a rearranged while loop)? and should exponentiation be a core feature too?

flawr

10:29 AM

@Riker Just noticed that the drawings in /r/redditgetsdrawnbadly/ are usually a lot better than the ones in /r/redditgetsdrawn/

12:02 PM

anyone musically literate here? do you see any meaning in this: B♭ A# A A G D♭ B♭ B♭ D D D? is X# (the sharp of some note) the same as Yb (the flat of another)? does the notation: 0:17 - 3^7, 0:29 - 2^5, 0:45 - 9^4, ... mean anything in music? (i believe the : is for minute:second, but what about x^y?)

12:19 PM

@ngn I'm not musically literate at all, but I can tell you that the sharp of a not is the same as the flat of the next one. 'Sharp' means a note one semitone (or a half-step) higher in pitch, and 'flat' means it's one semitone lower in pitch. It's also worth noticing that E and B don't have a sharp accidental, while C and F don't have a flat accidental.

^ That only works for 12-tone tempered instruments, btw. No idea about untemepered ones.

@J.Sallé "tempered" = fixed-pitch?

@ngn I meant "temperament" actually. English is hard.

@ngn So in this case: "An equal temperament is a musical temperament, or a system of tuning, in which the frequency interval between every pair of adjacent notes has the same ratio." 12-tone temperament is a type of equal temperament.

@J.Sallé facepalm. so musical notes don't have definite frequencies? and i thought programming was messy... :)

@J.Sallé ok, but assuming no 12-tone "temperament" (english is not one of my strengths either), the only exception to X# = Yb is when there's a semitone gap between X and Y, isn't it?

(X and Y being consecutive notes)

@ngn I'm not sure. As I said, I'm not very musically literate, so that's as far as I've studied.

@ngn and yeah, the frequency of a note fluctuates slightly. The thing is, some of those frequencies sound weird for most people, so temperaments were created to account for that.

@J.Sallé thanks anyway. i think i found what i need - a table of piano frequencies

3:23 PM

@ngn So are we gonna put them in an mp4 or another audio file (maybe midi) and see what happens?

@MilkyWay90 well, it's gonna be just a bunch of notes with unknown durations. i doubt it would be a recognizable melody

i suspect these notes' frequencies might have something to do with the steganography

@ngn Let's talk over on the cicada chat, that was made for discussing things like this

sure

Riker

3:54 PM

@flawr "Flawr's Theorem"? :p also, have you payed attention to the recursive drawings people have posted?

Neil

5:18 PM

@Riker recursive drawing makes me think of mcescher.com/gallery/back-in-holland/drawing-hands

A: Sandbox for Proposed Challenges

Duplicate of this?

@EriktheOutgolfer ^

New Sandboxed Posts

0

Binary self-rotation Given a binary 3D array, for each layer, cyclically rotate up each of its columns as many steps as indicated by the binary encoding of the columns of the layer above it, and then cyclically rotate left each of its rows as many steps as indicated by the binary encoding of the...

Riker

5:41 PM

@Neil it's people painting pictures of people who painted pictures of people painting pictures

@Riker so basically a recursive function without an exit condition

7:33 PM

@ngn lol theoretical studies are the entire opposite of logic for us straightforward people :P

hm, Deadcode himself has offered a bounty... almost 26.9% of his rep

@Adám hm, doesn't seem like it

major differences: number of dimensions, handling of both rows and columns

in fact, the way you represent the binary arrays has inspired me to post an unrelated challenge... :P

@EriktheOutgolfer Apparently, my challenges are inspiring.

7:48 PM

@Adám really? ;P

@EriktheOutgolfer Bu in essence, it is the same, adjacent major cells determine rotation amount.

@Adám you have to think outside array manipulation languages as well

@EriktheOutgolfer E.g.

the majority of languages doesn't even know what a 2D array is (Jelly included, even though it has built-ins with "multidimensional" in their name, although it does have vectorization to an adjustable depth)

@EriktheOutgolfer Yeah, that's a bit hard for me.

@EriktheOutgolfer Nothing prevents solving this with vectors of vectors, though I guess few languages built on such arrays support rotation across rows.

restricted-source printable-ascii

7:51 PM

@Adám in fact, BASIC knows what a 2D array is very well lol

although they still use two indices instead of a vector

Downgoat

8:51 PM

@ASCII-only how would global variable initialization order be determined

flawr

11:01 PM

@Riker if only the margins were a little bit larger...

@Riker I saw them a few days ago I think, is it still an ongoing thing?

New Main Posts

0

Q: No-alphanumeric code exec

Introduction In most interpreted languages, there is an operation that will run code generated at runtime. It generally requires alphanumeric characters to refer to this operation. Write a program, using as few alphanumeric characters as possible, that is able to download and execute any string ...

dzaima

tfw you've written a beautiful bit-based algorithm and finally implementing it realize that you've swapped LSB and MSB in half the places D:

@JonathanAllan so... yeah, I thought I outgolfed you over there, until I realized that four-in-a-row and no more moves is pretty legit (and actually victorious for one... :P)

Jonathan Allan

@EriktheOutgolfer It still looks so long!

11:17 PM

mine was 40 bytes, but it invalidated every four-in-a-row, not just the ones where a move happened afterwards

haven't trashed the code yet though, might be able to improve it and outgolf you tomorrow :P

(it's way different than yours, btw)

@JonathanAllan yeah, for some reason it looks like your code was written by a psychopath, no offense ;P

like, how do you even fit %15 in there???

Jonathan Allan

@EriktheOutgolfer Ḅo1%15 tests for [2,2,2,2] or [1,1,1,1] when lists of length four contain any of {1,2,0}.

...good luck with the golfing, I'm almost sure it must be possible in less; and, yes, the code looks like a nightmare!

(tests by yielding 0)

... although actually :ṀSn4 is shorter.