« first day (2402 days earlier)      last day (2444 days later) » 

8:00 AM
Yes, I said proportionally hard to Main.
 
Ah I see
 
Like, CMCs are 3-4 times easier than Main, this is 3-4 times easier than the toroidal challenge
 
It is the kind of thing however that some langs have built ins for
 
^ cc @MATL
> No matching records found
Oh no, MATL doesn't
 
8:23 AM
you just need to permute the rows and columns :P
 
Thats pretty hard
for a CMC
 
8:45 AM
If two answers are of the same length, and the challenge author accepts answers, which one should be accepted?
 
@ASCII-only it seriously needs a 1-byte x - 1 builtin
 
@EriktheOutgolfer Why :P
 
personally I've needed to use it a lot of times
of course if it has x - 1 it must also have x + 1
@WheatWizard CMC may mean chat mega challenge :p
 
I'm running out of codepage space though :/
 
well, you can use some of the unused chars there
 
8:55 AM
Maybe reserving 96 chars for printables was a bad idea though
 
well, otherwise we would have quoted strings
for example do you use and in charcoal?
@ASCII-only
 
@EriktheOutgolfer They're all used
But hmm
Maybe Charcoal needs a version 2 :P
 
@ASCII-only even and ?
 
I only add them because they have a use
lol
 
0
Q: Find out how Abstractly Capacious™ someone is!

JazzachiAt the Code-Golf™ Psychological Society, you are delighted to present a new method of measuring mental age: Abstract Capacity™. To get the Abstract Capacity of a person, you take their actual age divided by 4, take their IQ divided by 9, then add the two values together. Then, round this sum dow...

 
9:11 AM
@NewMainPosts plagiarism suspected
 
9:25 AM
@EriktheOutgolfer Hmm if there's a way to eliminate variables from Charcoal there'd be way more free space
Not sure if making it stack based is a good idea
 
stack-based? isn't it canvas-based?
(btw pings lately come to my inbox immediately before I even hear the ping sound)
 
@EriktheOutgolfer O_o what
@EriktheOutgolfer what ._.
 
are you sad?
 
???
 
9:29 AM
whoops
@ASCII-only * .-.
@EriktheOutgolfer what do you mean
 
"._." means depression or something
not sure how "canvas-based" would depress you
 
@EriktheOutgolfer It depresses me that you don't know the difference between memory model and output model lol
wait a second fixed i guess
 
@ASCII-only so canvas isn't memory? isn't it stored somewhere?
 
@EriktheOutgolfer Well it is
But your variables aren't on the canvas >_>
 
they're part of the whole memory
 
9:33 AM
I guess no. wat
it's just output
 
9:47 AM
0
A: Sandbox for Proposed Challenges

Qwerp-DerpIs this a valid Takuzu board? code-golf decision-problem Takuzu is a logic game in which you have to complete a grid with 0s and 1s. The grid must follow 3 rules: No three horizontal or vertical grids in a row can be the same. There must be an equal number of 0s and 1s in each row and column. ...

 
hi
 
10:29 AM
3
Q: Braille graphics

ngnCut a boolean matrix in 4x2 blocks and render them as Braille characters U+2800...U+28FF. [[0,1,0,0,1,0], [1,0,0,0,0,0], [1,0,0,0,1,0], [1,1,1,1,0,0]] ⣎⣀⠅ Pad with 0-s if dimensions are not multiples of 4 and 2. [[0,1,0], [1,0,0], [1,1,1]] ⠮⠄ Usual golfing rules apply, flexible on in...

 
10:43 AM
0
Q: Circularly moving sum

Luis MendoInspired by a question at Stack Overflow. Given a non-empty array of integers x and a positive integer n, compute the sum of each sliding block of length n along the array x, circularly filling the missing values at the left with values from the right as follows: the first block contains the f...

 
@officialaimm 32*2 and 2*10 are as long as 1024 but 4*5 is shorter. Good catch.
 
11:03 AM
is the programming language that is run on the ti-nspire cas different than one run on for example TI BASIC on the ti84
do they count as different golfing languages
 
@HusnainRaza Probably
@HusnainRaza If they're different then yes :P
 
i think one has more features than the other but all resources I hqve seen pnlije say that both run tibasic
prilije? i meant online
 
@HusnainRaza why are you asking?
 
i wanted to write a program for the nspire for the oeis question, but ti basic is talen
 
@Adám Oh, there is not a square builtin in APL?
 
11:12 AM
@Mr.Xcoder Here I am
 
imo theyre different, different features and different syntax
 
@LuisMendo I don't really understand how the array is circularly shifted. Can you explain?
And maybe give an example of circular shifting such that it is clearer?
 
Imagine you have copies of the array (as many as needed) attached to its left. Does that help?
 
Oh, ok
 
The first block contains the first entry of the original array, and whatever is needed to its left
 
11:14 AM
@LuisMendo And then how do you get those subarrays? Get each contiguous sublist of length n?
 
Yes, sliding one entry at each time. I hoped that would be clear from example 1 :-(
 
Yep, I was confused by ex 2 in fact :)
 
@Mr.Xcoder That's there to show you may need more than one copy of the original array attached to the left
 
@LuisMendo I now have understood. +1 thanks!
 
Thanks to you!
 
11:18 AM
I want to do it in Pyth. It's a great challenge
 
Stack Overflow is a good source of ideas :-)
 
@Pavel TIL your avatar is dirty bomb logo
 
@HusnainRaza I consider them to be different, so go ahead and use them
 
11:54 AM
@HusnainRaza Vastly different. IIRC
 
12:31 PM
@officialaimm No, * (power) when used with only one argument (the right side one) uses e as left argument. However, ×⍨ is "times-selfie", i.e. multiply by itself, which is of course the same as power. Remember, APL isn't a golfing language. J has *: for square, however.
 
@Adám Oh, I see.
 
12:42 PM
It's always fun when your cursor disappears
From every tab in your browser
 
12:54 PM
noice
 
1:27 PM
;-; *args is a tuple
 
@totallyhuman cast to list with list :D
 
ye but it makes my code look dumb
list list list
 
@totallyhuman That's what MY looks like, list(map(...)), list(range(...)).
 
@Zacharý that is the problem with Python 3 :/
 
well yeah that's python 3's dumb obsession with custom objects
 
1:35 PM
ninja'd
 
with more detail :P
 
I probably should make MAP=lambda *args:list(map(*args)) or something
 
:P
from __past__ import saveusfromthecustomobjects
 
from __future__ import braces gives the best error message.
 
@Zacharý isn't something like Nice try?
 
1:44 PM
Not a chance
 
25
A: Shortest Error Message

totallyhumanPython 2, 87 79 bytes -8 bytes thanks to Zacharý and Erik the Outgolfer. from __future__ import braces #i am most surely seriously actually totallyhuman Try it online! Error message, 78 bytes: Assuming the code is stored in a file named a. File "a", line 1 from __future__ import br...

 
@Zacharý If you make a better version of a builtin, the correct name should be map_ with a trailing _. Its in pep8 and everything
 
still waiting for from __past__ import print_statement :P
 
@totallyhuman That would be pointless, but from __past__ import division, range, map would be good.
 
@Zacharý division ಠ_ಠ
why do you want default integer division back ಠ_ಠ
 
1:49 PM
Codegolf, why would you want print w/o parens back?
Mainly range and map.
 
it's just a byte extra to do a//b
 
A is a lot in code-golfing.
 
05ab1e has the motherlode of built-in constants
for some reason
 
Some of žv - žJ could be done using a 2**x builtin.
o. That's a waste of bytes.
4o => 16.
 
...oh no fair i out-golfed by brute force
 
2:01 PM
o_o
 
@Zacharý ಠ_ಠ
 
2:13 PM
@totallyhuman If you think that's bad, look at the first few lines of Deorst D:
 
@totallyhuman Brain-flak has one built-in constant
 
() for 1? :P
 
2:34 PM
@totallyhuman that reminds me to add numeric constants to Deorst
 
2:44 PM
Hey, @Adám, how many bytes would and be in Dyalog Classic? (Or anyone else that knows the answer)
 
@Zacharý use TIO
 
@Zacharý ⍤⌸⊆⍸⌺ are replaced by ⎕Uxxxx where xxxx is the Unicode hexadecimal code point, so six or seven, depending on whether you need a trailing space to separate from an identifier on the right.
@cairdcoinheringaahing Yes. If a language is registered as SBCS, TIO counts chars, even if your code has more than 256 (257!) distinct characters.
 
@Adám So that's why Jelly thought that was 1 byte on TIO
@Adám That's how I'm replying from now on :D
 
@cairdcoinheringaahing Yeah, by mistake, I edited the old answer instead of replying with a new one. My bad.
@cairdcoinheringaahing No, TIO does not count bytes in SBCSs, it only counts characters.
 
I won't actually, just wanted to try it
 
2:50 PM
@cairdcoinheringaahing Yes. If a language is registered as SBCS, TIO counts chars, even if your code has more than 256 (257!) distinct characters.

Sandbox

Where you can play with chat features (except flagging) and ch...
 
Well, TIO hates me.
 
@Zacharý Btw, can be replaced by ⎕OPT saving you two bytes. However, my most common usecase for is ⍠1 for ⎕R or ⎕S, in which case you can prepend (?i) to the first regex.
 
@Zacharý What seems to be the problem?
 
NVM! I never realized you said hex.
Okay, I still have the problem. WHy is this giving me a translation error? tio.run/…
 
3:01 PM
@Zacharý ⎕UCS 10240 is not in ⎕AV
 
So, it won't work in classic at all then, right?
 
@Zacharý Classic cannot store characters which are not in ⎕AV at any time, because it stores characters as indices of ⎕AV and only one byte is set aside for each character.
 
Well, I'm just going to use classic encoding on a unicode answer then ... >_<
 
@Zacharý I'm not sure that's valid. If the code cannot actually be run on Classic. Is this Nick's Braille graphics challenge?
 
Does anyone know what time Dennis is usually active?
 
3:07 PM
@cairdcoinheringaahing Click on his avatar.
 
@Adám Yep. >_<
 
> Seen 1m ago, talked 12h ago
 
@Zacharý On classic, you can substitute b/⍳⍴b← for in most cases. And you probably don't need as it give you overlapping cuts. You need non-overlapping ones.
 
I have no idea how to do that for a 2d array, is my big problem.
 
Oh, and is {(,⍵)/,⍳⍴⍵} for even special cases, and {(,~~⍵)/,⍳⍴⍵} if you need catching of non-Booleans.
@Zacharý We'll then you could ask in the APL room…
 
3:12 PM
Does anyone know who I should ask about chatexchange?
 
@muddyfish help or bug?
 
both?
 
sorry wasn't very clear. Help or bug?
 
I'm trying to get it working in a script that uses asyncio
 
@muddyfish @ATaco knows a lot about userscripts, you could ask him
 
3:14 PM
i think muddyfish is working in python
 
I am
 
yea CE is python
 
which ataco hates :P
looks valid to me
 
Basically, chatexchange doesn't work in scripts using asyncio because it doesn't use asynchronous methods
 
@totallyhuman Ask Martin, he deleted it
 
3:18 PM
@totallyhuman There is a same answer with lesser (or corrected) byte-count, 4 upvotes and from the same answerer :D codegolf.stackexchange.com/a/54010/59523
 
@totallyhuman What post is that?
 
48
Q: First code golf decathlon

DennisTasks All competitors try to solve the following list of 10 tasks: math Read a positive integer n from input and return the sum of the cubes of the first n non-negative integers. For input 1, this should return 0. Read a positive integer n from input and return a truthy value if and only if ...

 
Also, has anyone else noticed the mass set of edits in the review queue?
 
3:22 PM
It's just some new user making http become https. The road to edit badges are easy :D
 
💂 is this offensive to Russians or not?
@totallyhuman lol, unicode-table.com/en/#1FFFF
 
lolwat
 
> <Not A Character>
 
3:36 PM
@totallyhuman Does ,,, not have that builtin?
 
...no?
 
0
A: Sandbox for Proposed Challenges

ckjbgamesHelp Me Catch Up With xkcd! code-golfinternet I definitely enjoy xkcd, and you, as a code golfer, most likely enjoy it as well. However, I often find that I won't check the website for a while and there will be a lot to read! So, I want to make a solution for that. In proper code golf fashion. ...

Had this idea after visiting xkcd
 
that's just wasting a byte
 
I'm about to post this, any last feedback?
 
I think its a dupe
can't find what of
@cairdcoinheringaahing Wait I read it again, and now I'm confused as to how a cop can make a submission that cannot be instantly cracked
 
3:48 PM
0
A: Sandbox for Proposed Challenges

ckjbgamesHelp Me Catch Up With xkcd! code-golfinternet I definitely enjoy xkcd, and you, as a code golfer, most likely enjoy it as well. However, I often find that I won't check the website for a while and there will be a lot to read! So, I want to make a solution for that. In proper code golf fashion. ...

 
@WheatWizard Did I not mention that the cops riffle shuffle their own program?
 
riffle shuffling is a process that can instantly be reversed
 
@WheatWizard how?
 
@NewSandboxedPosts dupe notification... :D
 
@cairdcoinheringaahing Ok I see it is not possible to isntantly reverse it. Your description of a riffle is rather confusing.
However now I'm unclear what exactly cops hidden programs are supposed to do
 
3:52 PM
@WheatWizard riffle shuffle a string.
Guess it needs a bit more clarification
 
Thoughts on this? This isn't a dupe right? codegolf.meta.stackexchange.com/a/13791/59523
 
@officialaimm ask the entire room, not just Wheat Wizard :)
 
@cairdcoinheringaahing ;)
 
4:06 PM
@cairdcoinheringaahing sorry my internet went out there
 
16 hours ago, by Leaky Nun
> My internet went off for 3 minutes
I met my family
They were nice people
 
@cairdcoinheringaahing I also don't get the "you only get one guess" thing. How could someone guess wrong?
 
@WheatWizard thinks of excuse I have no idea. I'll add that to the list of thinks that need fixed
Q: I have a builtin in a new language that solves one of my challenges, but requires and extra number pushed to the stack, in order to work. Does it count as violating the new standard loophole?
 
What new standard loophole
 
@Zacharý adding a builtin to solve the challenge in 1 byte
 
4:16 PM
that's a old standard loophole tho
 
36
A: Let's allow newer languages/versions for older challenges

MD XFTrust the community and add a standard loophole I think the community can handle the add-a-builtin problem (I may be very wrong). I think that if someone adds a feature to their language that is clearly meant to only solve one particular challenge, the community will probably downvote that answe...

 
That never specified "1 byte".
 
@Zacharý Does this count as a loophole for this‌​?
 
It should. That seems like a very cheap tactic since you're probably never going to use it again.
 
@Zacharý M changes its functionality, depending on its 3rd argument, so it could be used again.
 
4:27 PM
I'm talking about with the same 3rd argument. If it were something like lambda n:range(len(n)), then it would make perfect sense to add it. But something like a challenge that you can probably never reuse, that seems a little cheap.
 
@Zacharý it's element wise operation, where the 3rd argument determines the operation, such as addition and miximum
 
So, why did you add THAT to the builtin?
 
@Zacharý add what?
 
Oh, I thought it was a different multiply-two-strings challenge. That actually doesn't seem cheap at all then. I think it's fine.
 
@Zacharý demands upvote ;)
 
4:31 PM
2.6k, I'm not upvoting you -_-
 
@Zacharý half of that is movies.Se
 
SOrry, I modified it.
 
2
Q: Primus-Orderus Primes

Bill SteihnWho are they? Primus-Orderus Primes (POP) are primes which contain their order in the sequence of primes. So the nth prime, in order to be POP, must contain all the digits of n in a certain way which I'll explain. Examples Let's get things clearer: All digits of n must appear among the digits...

 
@NewMainPosts this bill guy's addicted to primes
 
How does converting diacritics to ascii art (for lack of a better term) sound for a challenge?
 
4:41 PM
@totallyhuman He has 7 questions, all relating to primes
@Zacharý VTC as dupe (can't find it)
 
@cairdcoinheringaahing What?
 
@Zacharý I've seen that challenge before, can't find it though
 
It probably does not exist.
  /
Adam

      /
Zachary
^ Something like that ^
 
@Zacharý Ok, haven't seen that before.
 
4:49 PM
"
U
 
Would probably be solved with some basic string translation though
 
@totallyhuman Why are you sending that?
 
no
@Zacharý it's an emoticon :P
 
@cairdcoinheringaahing Adding a new line with string translation is generally far from trivial
 
4:49 PM
@cairdcoinheringaahing it's multi-line
it's much harder than regular string replacement
 
@TuxCopter Unless you're Charcoal.
 
Or V
 
Question for those who know Haskell and have read "Learn You a Haskell" and "Real World Haskell": Which of these books should I start with in learning Haskell? I'm thinking of using Haskell in my thesis in some way
 
@Zacharý how about Ǖ?
 
@TuxCopter Wait, what exactly is the task?
 
4:51 PM
@cairdcoinheringaahing Maybe modify it and make it this (in a following message)
@DJMcMayhem hears V
 
Or do you have another book recommendation? Books about other functional programming languages are also acceptable
 
Exactly :P
 
-
"
U
@Sherlock9 Is self-learning not an option? :P
 
@Sherlock9 I read Learn you a Haskell and it was rather helpful in my opinion
 
Oh I get it. That's a vim built-in
 
4:53 PM
@DJMcMayhem Huh? Vim has a builtin for that? Should I use both precombined AND combining marks?
 
11 messages moved to Trash
 
I'm not sure what that means
 
block select, maybe?
 
I just know that you can do i<C-k>:A and get Ä (for example)
 
0
Q: Stretch an array

Wheat WizardEarlier I defined the process of crushing an array In a crush we read the array left to right. If at a point we encounter two of the same element in a row we remove the first one and double the second one. Now crushing arrays is easy, whats hard is uncrushing them. Your task is to take an...

 
4:54 PM
@DJMcMayhem Not the challenge I'm thinking of.
 
Well give me an example than
 
ÃÇÈ =>

~ \
ACE
 ,
 
Happy 26th birthday Linux!!! Here's your fucking cake, go ahead and compile it yourself. https://t.co/18aZL45m28
7
 
@Zacharý test cases: Ŀ and
 
Ŀ => wouldn't be input
Ṣ =>

S
.
 
5:00 PM
@Zacharý its a precombined diacritical though
Also: Œ and Æ
 
Not everyone would do that, of course. There has to be SOME limits. Maybe just leave it?
Ŀ, Œ, and Æ will probably just remain the same.
 
Ŀ => L.
Π=> OE
 
Then L. and OE would be confused with them.
 
@Zacharý Ah, so it's the reverse of what I did
 
@Zacharý yeah but its only a one way conversion, so it doesn't matter
 
5:03 PM
I don't think vim has that
 
TABS!
Ŀ => L<TAB>.
Π=> O<TAB>E
 
test case: ĿL.ŒOE => L.L.OEOE
 
ĿL.ŒOE would be LL.OOE<TAB>. E <-- two trailing spaces and leading spaces around the final E
This is definitely going in the sandbox. But a few questions: should it also handle combining-diacritics? And should input like Ǖ be handled?
 
Are you sure it's OE, not CE?
 
I'm sure.
Would this be easy in Charcoal/V/Vim?
 
5:13 PM
lambda l:h(map(g,l))
g=lambda n:n%2<1and[g(n/2)]*2or n
h=lambda l:[l]*(l*0!=[])or sum(map(h,l),[])
lolwat
 
hi
is it possible in C to specify compiler flags by some #pragma magic maybe?
 
@Zacharý I don't know charcoal, but it seems like it would be a major PITA in charcoal. However, it would also definitely be a PITA in V. So I have no idea
 
@totallyhuman What is this to?
 
2
Q: Stretch an array

Wheat WizardEarlier I defined the process of crushing an array In a crush we read the array left to right. If at a point we encounter two of the same element in a row we remove the first one and double the second one. Now crushing arrays is easy, whats hard is uncrushing them. Your task is to take an...

i'm not sure it's that hard o0
 
@DJMcMayhem SOGL?
 
5:15 PM
9
A: Add GCC options to top of C source file

Matt JoinerYes it is, at least for some flags. You can push and pop diagnostic settings like this: #pragma GCC diagnostic error "-pedantic" #pragma GCC diagnostic warning "-Wall" This is also possible for optimization levels on a per-function level: #pragma GCC optimize ("string"...) See these other q...

 
interesting!
 
@Zacharý Yeah, I bet SOGL and 05AB1E would probably both kick butt compared to V/charcoal
 
#pragma GCC optimize ("string"...) is intriguing
 
update: after trying to program next oeis on the ti nspire cas, there just isnt enough computing power
shouldnt have expected much from a graphing calculator
 
@DJMcMayhem 05AB1E I'm not so sure about, but SOGL would probably destroy everybody.
 
5:18 PM
@HusnainRaza it only has to work theoretically (as with all code golfs)
 
CMC: Given a string, surround it in lines of spaces of the same length. For example, "Hello" --> " \nHello\n "
Shoot, spaces
"Hello" --> " "*5+"\nHello\n"+" "*5
 
@DJMcMayhem str.center should be able to do that in Python, I just can't be bothered to make it work :D
 
s=>('\n'+s+'\n').center(len(s)*3)
should work i think
 
Crap I forgot Newlines sorry
 
@totallyhuman Proton?
 
5:21 PM
yes
 
@totallyhuman See edit
 
Proton is Python + Javascript. Makes it much better for golfing
Does anyone here know regex?
 
V, 6 bytes: ÄÒ YGp
@cairdcoinheringaahing Most people in here, I'd assume
 
s=>('\n'+s+'\n').center(len(s)*3+1)
 
@totallyhuman Do you see the issue with your answer? If not I can try to help.
 
5:26 PM
i see the issue but i'm not sure i understand the process of stretching
 
Streching is not a process, we are looking for a solution rather than applying a process
 
> jeez man y u so salty ;-;
Didn't mean to be salty
Sorry :-)
 
lol i was joking
although maybe that didn't come across well enough
 
No, I got the joke :P
 
The problem with your answer was it was too eager to separate even numbers. For example if you want to stretch [1,2] you could try [1,1,1] however because crushing is left to right that would crush back to [2,1] meaning it isn't a valid stretch.
 
5:29 PM
Hence the :-)
 
@DJMcMayhem Jelly, 12 bytes Try it online!
Waaay too long
 
@cairdcoinheringaahing What does it do?
 
@Mr.Xcoder check the reply
 
Oh yeah, sorry
Yep, it's quite awfully long
 
@Mr.Xcoder Jelly doesn't like strings
 
5:33 PM
still, I think I can golf it
 
@Mr.Xcoder go ahead, I'm grappling working with Deorst
 
@cairdcoinheringaahing try
 
@Mr.Xcoder tried it. The problem is that it returns z
(z is the argument)
 
CMC: flatten an array of positive integers and arrays
 
@totallyhuman Jelly, 1 byte: F
 
5:35 PM
@totallyhuman Pyth, 2 bytes: .n
 
@totallyhuman Deorst, 1 byte: ]
 
@totallyhuman Python 2, 39 bytes: lambda l:eval(`l`.translate(None,'[]'))
 
0
A: Sandbox for Proposed Challenges

StephenArray Factorization code-golf array-manipulation primes Given an array of positive integers, output a stable array of the unique prime factors of these integers. In other words, for each integer in the input, add all prime factors of that integer not already in the output array to the output ar...

 
@totallyhuman lambda x:[i for k in x for i in k]
 
ಠ_ಠ I forgot to add a str.count method in Deorst
 
5:38 PM
@Mr.Xcoder uh what
 
No, I'm serious
 
@Mr.Xcoder Only works to depth 2
 
Oh crap
 
@Mr.Xcoder Do you know if it's possible to link binaries of newer libraries (e.g. ARKit for iOS 11+) statically to an app so you can use them in older versions (e.g. iOS 10)?
 
and only works if there are no numbers
 
5:38 PM
Or do they use some new kernel/really-low-level things
 
Letter for str.count? C is gone, as is c
 
@Downgoat Not entirely sure about that, but I think they use new kernels.
 
O_o huh ok
 
@cairdcoinheringaahing :P
 
ik but cas doesnt privide enough precision
like my typing skills
 
5:40 PM
what?
 
I guess some GPU stuff might of been updated for thinks like ARKit but I guess I'll try statically linking MapKit
 
@HusnainRaza wrong chat perhaps?
 
@totallyhuman ascii only
@HusnainRaza you can reply to messages by clicking the arrow nex to them and clicking reply
@totallyhuman He was replying to one of my earlier messages (I think)
One last message to see my rep
 
yes
thank for tip
 
For anyone who knows Python flavour regex, feast your eyes on this
Kk|(E[^\d ]([\da-f]+)?)|([?SmFvW][^\d]([\da-f]+)?)|([gr].)|([tU]..)|(o.)|('[^']+')|([^\da-f? ]([\da-f]+)?)
 
5:44 PM
what
>.> curse regex
 
what is that supposed to do?
 
@totallyhuman its the command regex for Deorst
I'm not that good at regex, but it works ¯\_(ツ)_/¯
 
Oh regex is... killing
o/
 
Question: should f().a be an lvalue?
 
3px looks very, very huge
I think I'll go with 2px
 
5:58 PM
OK, why in the world does everyone love regex parsers so much
 

« first day (2402 days earlier)      last day (2444 days later) »