« first day (4184 days earlier)      last day (956 days later) » 

00:01
@Neil it's a balancing system :P
00:19
@LeakyNun Pip, 23 bytes but not quite right... FiU,tRLt{P(U,tRLt)*i%t}
just thought this outcome was funny lol
oh what i thought it linked the result
FiU,t{P(U,tRLt)*i%t} is closer but no cigar lol
oh hey wait i think i got it
@LeakyNun Pip, 17 bytes: FiU,t{P(U,t)*i%t}
cc: @DLosc >:)
@thejonymyster I see you have done 100x of the work
what can i say im a go-getter
17 bytes one actually works lol
00:38
@DLosc Stdin doesn't work, but that's a general problem that's more of a pyodide bug
00:55
If you can trace my online accounts, you can probably find my real name :P
I believe user knows it
@Steffan Same here
tho for me its 100% you can find my real name
@thejonymyster Pip -l, 12 bytes: {Ua*Ub%t}MCt
@Seggan well, I can't find it so it's not 100% :P
>having a real name
Sandbox posts last active a week ago: Generate a Tiefling's Traits
01:00
@Zionmyceliaadamancy NGL, I'd be happy if M$ went bankrupt.
01:12
i kinda honestly prefer Jo to my actual name
i have to remind myself that while it would be hilarious to change my name to a bad pun, it would definitely make my life a lot harder
01:28
flashbacks to Beavis and Butthead "joe mamma" scene
 
1 hour later…
02:37
@Steffan For some reason it shows up in the github activity graph thing
02:54
huh what
@emanresuA where on earth in github did you find that
I can't find it
The network bit
network?
in vyxal?
03:16
0
Q: All Possible Ties in Tic-Tac-Toe

YousernameYour task is to output all possible ways to end a game with a tie (all rows, columns, and diagonals are completely filled and do not have 3 X's or 3 O's, and there are exactly 5 X's and 4 O's in total) in Tic-Tac-Toe (assuming X goes first): OXO XXO XOX OXX XOX OOX XOX XOO XOX XXO OXO OX...

03:38
working on a really wonky challenge idea; writing this is so much though lol
its just in draft form rn so i wont post but the idea is this:
write a compressor and a decompressor; the inputs to compress are binary trees
specifically of the form:
@NoHaxJustRadvylf it is
Tree: [] | [Tree, Tree]
not that strict ofc but like, my point is that there arent any trees with singular branches, nor data on the tree other than structure
cherry on top: programs must decompress to valid trees
not sure how to score, i feel like compression ratio and the source layout thing are both super significant to the challenge
maybe i'll just have code golf as a secondary scoring metric, tie breaker n all that
03:55
@Seggan yeah everyone knows lyxal's real name is Rick Astley
Yeah, duh
@PyGamer0 isn't Lyxal's name in their bio?
You got me, good one
04:37
@NewPosts Is there an OEIS sequence for "number of possible ties on a nxn board"?
04:49
n in a row?
3 in a row seems to not exist yet. and I think it'll grow much slower because you have no choice but use honeycomb pattern in the middle
@Bubbler how did they compute the result for n=7
05:07
OEIS is a secret organisation with access to all of Google's computers.
of course it can't be an efficient algorithm
Oops
never stop avoiding quadruple negatives, people
05:45
@DLosc Here's what'd be equivalent to Regenerate's (A!B) in mainstream regex: (A|(?!A)B) or (?(?=A)A|B) (or (?(A)A|B) in .NET). As a universal drop-in replacement for (A!B) (ignoring that backreferences need renumbering), they must duplicate A. But my regex engine with -xlcnd,ml, it'd be (?(?*(A))\1|B) where \1 is whatever group number (A) captures, where (?(?*condition)yes|no) is a molecular lookahead conditional.
 
1 hour later…
07:02
CMC: Create a file called *
@emanresuA huh what about the extension????
@emanresuA Zsh -F, 2 bytes: >*
CMC#2: Delete a file called *
@AidenChow file extensions are a social construct
@emanresuA risky business
I tried ;)
07:03
wait so files dont need extensions?
Zsh -F (-F is VERY IMPORTANT), 4 bytes: rm *
why is -F so important?
otherwise the * matches any file, so you'll delete every file in the current directory
oh damn ok
07:06
IIRC some OSes just won't let you
relatedly, Windows won't let you create a file called nul (and a few other names)
Is it possible to create a file named . or ..?
No
also impossible (on Unix-like OS's) are: empty string, names containing / or null bytes
I believe it's possible through reformatting
whats reformatting
07:08
if you use a different filesystem?
I don't think the OS will like you if you try to manually make a file called .
This is ingenious :)
Idea: SE but if you click the button enough you can upvote twice
That just sounds like normal SE with extra steps
07:35
@emanresuA Medium basically has a system like that. You can "clap" as many times as you like
@mousetail SE does too
On SE it's a bug
You can upvote multiple times if you refresh quickly enough
220
Q: Duplicated votes are being cleaned up

Yaakov EllisIt was reported on MSO that it is possible for a user to upvote or downvote more than once on a single post (resulting in reputation changes for the author for each vote). After some data examination, we were able to confirm that this vulnerability exists (it is happening due to race conditions s...

@mousetail that's the joke :p
You can also answer a question once it's been closed for 30 more minutes. Just need to remove the "disabled" atribute on the button
08:24
For 3 in a row, it is very boring unfortunately:
https://ato.pxeger.com/run?1=jVXBbqMwEL3tIV8xUrWSSXAbSA4VWvY39hChihqnjbZr0ODsavsre-ml_ah-zY5tMBBCCFLC-Hnm-c3YY_59VH_1c6ne3t6Pes_vP7_8KOQeHnPxUyP9sccyxyKE-jUEFEGyAHoOe1ClpnECKPURFdTHX8z8sPwTwL5EIAMOCmx0AGlKBLBcQgwrb93dQWzpMAQBKdHdVmXFAodJTdDa2oZPGbYohNgpaFSgYV53UCvOwHbpHa6ibCeyPhAbIAFRKn1QRznk-572lzjDx0_5-AU-Ma-PgkniCRBP8s3qo3B-ysen-BC-0XbwGHJVELkbzJdzKNgWdFKyK-n1C_DxAnxmgUEOswWyGfBxBheK1MvguvPBx_rP03d7RAGgPG6O_yqdbMPFONa1CvVQXlVSFYyZrgp8M5kmpddi4ZsJc_Uk2SaEKGra2hIS1W63zpbKtt1D56mCzHoJpWvjlPnerKUsOr9t0NXnVWLZc24DsM86rKaZFtPTfktMX4H1bS2LKR55lOz
That link doesn't decode properly?
Oh wait I have to click see full text
 
2 hours later…
10:24
A year or so, then claim you're living in the Oort Cloud
10:47
@Zionmyceliaadamancy but i have to give $500, doesnt that defeat the point of "free" laptops? :P
CMC: Install arch linux :P
@emanresuA What are you trying to accomplish?
Various hax
Ok, you need to open the file in write mode at least if it's intended to work in python
also that link appears to select Pip, not Python
@pxeger Exactly
I'm messing with a bizarre bug in DSO
 
1 hour later…
12:29
0
A: Sandbox for Proposed Challenges

Command MasterInfinite quote escaping sequence sequence string code-golf Related If we start with the string a, and repeat the following process forever: surround the string with quotes, escaping existing quotes by doubling them connect the original string with this, separated by a a We get the following inf...

13:03
@emanresuA I'll need you to remind me on the day
 
2 hours later…
14:43
@hyper-neutrino Why are you all voting for Vyxal if you don't even play on the site ?
why are you singling me out in specific
besides, I've never heard of the site before, so since I was pinged with a link I'm going to check it out and possibly play on it, is that a problem?
What voting?
Nope, not a problem at all. I just wonder why is every body suddenly voting for Vyxal ;-;
Alright, fair enough. (Just wondering why you're pinging me in particular lol, I'm like the least active Vyxal community participant)
14:46
@TKirishima some mutually beneficial advertising is why :p
@hyper-neutrino Because I saw your username on WeekGolf and 5 people just created an account to vote for Vyxal, so I was just wondering, why? It's not you in particular..
@TKirishima as I said, mutually beneficial advertising :p
in Vyxal, Jul 11 at 8:45, by lyxal
Hey guys so there's this website called week golf and each month a new language is added by choosing the most popular suggested language. Vyxal is on that list and y'all should come vote for it to be added lol: https://week.golf/newLanguage.php
@lyxal Ohhhhhhhh XD Ok I see I see! Thank's!!
the screenshot is from the vyxal discord
that's where your new users are from :p
I see I see x) Vyxal will probably win..
14:50
@TKirishima Why do I have to click twice on each solution in the leaderboard for its code to appear?
Your on Mozilla ?
you're****************
Firefox 103.0b4
Also it would be nice if only the shortest solution of each user counted, so we could see other users' solutions on the leaderboard instead of just 5 variations of the winner's solution. (see for example dv_man's Bash answers)
@pxeger Yes, that's right. I will try to find out why. This was sent to me yesterday and I haven't had time to look into why.
Lyxal is nowadays more influential than Jeff Bezos.
@TKirishima where do you play? i cant find the chellenges
14:52
@mathcat XDDDDDDD
@Seggan In the "Problems" section
@Seggan I've been told before that it's not very clear but I don't really know what to replace it with. Do you have any ideas?
maybe make it on the header
@TKirishima I think if you change the navigation arrow to something more commonly known as a menu button (like three horizontal lines) it might be more obvious that it has links
14:54
@TKirishima The "start golfing now" button should take you to challenges, not your profile
I'll do it!
> @mousetail Almost finished the page about evolution
What does "upgrades" on the leaderboard mean?
When Solutions are revealed, you can upgrade them by doing fewer bytes.
For example, if a solution is 75 bytes and you do something like 74, it's an upgrade of 1 byte. (An upgrade is made when solutions are already revealed)
14:59
Ah I see
@pxeger For example, you just made an update of 1 byte on Bash!
What's the shortest way to convert a string to binary in python?
Solution are already shown on Fibonacci & Binary.
You can see them by clicking on the little bulb
How do I change my country?
@TKirishima ah thanks
15:10
- You can use a VPN
- You can try to analyse how you can do it
- You can reload your country by clicking on Enable - Disable
- You can hide it by clicking on the button below the flag
Oh, so it's based on your IP
in that case, it isn't detecting my IP's location correctly
My country is "xx"
@mathcat Each week there is a challenge and when the week is over, the solutions are revealed. Then a new challenge appears. This week's challenge is Ulam of Ulam
XD
XX is for "No country"
@pxeger wow is xx in Europe? my Europe geography is trash :P
Didn't you know, xx is above Latvia
not north, but vertically above
15:16
I thought it was below Slovenia
it's both
Europe is a torus, you see
(struggling to determine which shape can be both mostly-2d and also have slovenia above latvia)
It seems like europe is switching between a torus and a mobius strip.
It's a Klein bottle
15:18
European topology was vastly simplified by Brexit
4
but it's still pretty confusing
@mathcat Btw, I think that it should look something like that: [bin(ord(c))[2:]for c in s]
the open(0,"rb") trick used by the winner is pretty clever tbh
@pxeger Yup
@pxeger Lyndon is usually reaally good
15:39
@thejonymyster Nice! Some golfing tips: 1) \, does inclusive range starting with 1, which wouldn't save any bytes except 2) you're computing the same range twice, so you can yank the first instance with Y\,t and then use y in the second instance, which also means 3) you can drop the parentheses; finally, 4) since the loop body is one statement, you don't need the curly braces either.
Woo! :-)
thats cool makes sense ty
wow that takes it down to 12 bytes very nice
is there any way to get "last element of" other than "tail of length 1"?
i think order of operations isnt on my side but "last digit" here could be shorter than "mod 10"
@thejonymyster Last element is usually "element at index -1," which is two bytes using v for -1
But then yes, the precedence doesn't work nicely in this case
Oh, actually, in this case you really do have to use math because you're applying the operation to every element of a list. So "last element" would just get the last element of the list, not the last element of each element of the list.
@pxeger I dunno, there's still a ton of islands. I'd say Europe is topologically equivalent to a bunch of polka dots. Brexit just removed a few dots.
16:07
oops, site down
@DLosc ah right lol
16:53
@TKirishima I'd also suggest making everything not so huge on the site. I usually have my browser zoom at about 50% when at the site
^ lol
It feels like one of Ginger's websites
17:12
@mathcat ?
just thought (at)Ginger's websites look very similar to weekgolf
btw I've submitted an answer to the ulam of ulam challenge and it says TIME-OUT, although it seems to do fine on ATO.
Try to optimize your program.
ATO has only one test case, while on WeekGolf there are 12 test cases if I'm not mistaken (some of them hidden).
So it takes logically more time than on ATO.
In what language did you execute it?
Python
17:30
oh some are hidden. I was wondering why my hardcoded solution didn't work :P
I was trying to figure out how you had implemented such an elaborate cheat detector :P
@Steffan XD
I just had random + hidden validators some hard code shouldn't work anymore
You can pretty trivially find the hidden test cases with cat /mnt/in/* in Bash, right?
Or is there something more sophisiticated to hide them
@NoHaxJustRadvylf You can do ls and cat and it works too
But hidden test cases are random
I wonder if caird's name has anything to do with the Irish word "corcairdhearg". There is a passing similarity.
17:54
28
A: The many users of CGCC

Zion mycelia adamancy"caird coinheringaahing" comes from this Jelly answer. Given that the answer is written in Jelly (my primary language on the site), and the challenge is very much about the site, I thought that it'd be a fun, unique username that references the site. And it should be pronounced "cared co-in-hear-...

Yes, I wonder though if the jelly compression algo has the Irish color names and thus caird's name is influenced by that.
I would guess that's not the reason, since the color name appears to be corcair + dhearg rather than cor + caird + hearg.
Hm does Jelly know that though?
It's caird + coinhering + aahing
Caird being some shoe related job IIRC
18:10
@WheatWizard What I'm saying is, I'd be surprised if Jelly's dictionary included "words" that were not words in their own right but merely substrings from the middle of larger words.
It would be like having "rnatio" in there because "international" is a word.
Tbh I don't know how it works. I figured it was some sort of NN continuous space stuff going on.
It just glues together ASCII characters and/or words from a dictionary
Ah ok, then seems unlikely.
Okay, any bets on the random R/W speed of my new RAID 10 SAS SSD array?
Oh, it's being held up by either the 13 year old CPU or the 13 year old RAID controller. I guess I'll have to wait for the new server before getting an accurate measurement
19:13
@NoHaxJustRadvylf which version of SAS?
what kind of R/W? Sequential?
(very approximate) guess: 50MBps
19:25
hmm maybe I shouldn't have moved that yet
because now I want to reply to it and ask for VsTC
i mean you could always move it back lol it's fine i gave it the 5th vtc
@pxeger Ask anyway--it's easy enough to click through to the Bakery transcript and get to the question from there (unless only high-rep chat users can see the Bakery?)
the Bakery is readable to everyone (right?)
(although only ROs can send messages there)
@hyper-neutrino not that it mattered that it was the 5th for you
bakery is readable to everyone yes; it's a gallery so only approved users (in this case only the TNB ROs) can send messages there
19:41
@pxeger @NoHaxJustRadvylf higher or lower?
1 hour ago, by NoHaxJustRadvylf
Oh, it's being held up by either the 13 year old CPU or the 13 year old RAID controller. I guess I'll have to wait for the new server before getting an accurate measurement
Hmm, I was still hoping for some kind of number
but I guess 50MBps random is probably not possible on any drive with a 13 yo CPU
I will say that the read and write speeds were in the double digits of megabytes, but no more precision than that until the R530 arrives :p
That could actually abe surprisingly high to me if you think the CPU is really holding it back
What would you expect with a fairly modern CPU and RAID controller?
19:44
50MBps was my rough guess for a modern CPU
I have no idea how different that would be on an old CPU
but if it's double-digits, then it's probably not that much
Okay fine I'll say what it was. 60 MiB/s for read and 20 MiB/s for write
my top-of-the-line Samsung 980 Pro will do about 80, and that's over PCIe 4.0 x4 (which is 8GBps theoretical maximum)
so then, taking SAS 12Gbps=1.5GBps, times 2 (because RAID 10), adjusting linearly, and then adding a bit more because SAS is probably more bandwidth-efficient on random IO than PCIe, that's why I guessed 50
but you should also bear in mind that I have no idea what I'm talking about
My initial solution to Extract Strings from Text in BQN weighs in at 53 bytes. Given that existing solutions include 17 bytes in Jelly and 31 in J, there's gotta be a better approach. :P
in fairness, J gets 31 partly because it can just use eval to unescape
don't know bqn but it's probably possibly to port my jelly
or something conceptually similar; there's a lot of wiggle room with it
19:54
"port my jelly" sounds pretty weird out of context
@UnrelatedString I think so (though I'm not sure if BQN has anything analogous to ƙ). I like to try to solve the problem without porting anyone else's submission at first.
@DLosc k comes from a J command, so it's possible
20:25
I'm not aware of any "group into blocks of identical elements" type builtin in BQN. There is a Group builtin, but it takes a left argument which is an array of integers representing which "bucket" to put each element into.
20:47
I believe this is an implementation of Jelly's ƙ in BQN (as _k). It's... rather long.
There might be a shorter way to do it, but I don't think it'll get very much shorter.
(And in specific situations, you wouldn't define a modifier, you'd just put the logic straight into your code.)
from documentation it seems like alone should do the trick but i can't entirely tell what it does
Monadic puts the indices in buckets according to the value at that index. Dyadic puts the right argument's elements in buckets according to the corresponding item in the left argument. Neither definition has any logic about consecutive identical elements, so grouping "aabbaaabbb" will put all the as in one bucket and all the bs in one bucket.
oh ƙ doesn't do that either
20:56
Wait what?
that's why i have that logic with cumsumming the zeroes
Hmmm
So what I implemented above is not ƙ but something a bit more specialized.
I'll have to play around with your solution a bit more, I think.
Trouble is, I'm not super familiar with Jelly's parsing rules still, so I don't think removing part of the solution did what I thought it would. :P
CMC: Spot what's wrong with this answer:
1
A: Print a 10 by 10 grid of asterisks

emanresu AVyxal, 4 bytes ₀₀×ŀ Try it Online! ŀ # Make a grid of size ₀ # Ten ₀ # By ten × # Filled with asterisks

i can sort of pseudo-diagram it as = ”' Ż (Ḃ +)\ (¬ Ä (× Ḃ)) (Ḋƙ Ż) Ḋ
@emanresuA Well, the Try it Online link goes to a localhost URL...
21:04
Yep.
There was a brief panicked moment when I saw that it wasn't going to any online interpreter I know of and assumed it was going to be a Rickroll :P
@pxeger where are you seeing that? I can't figure the site out :-(
@UnrelatedString ⊐⊸⊔ separates a list into groups of identical elements. (Which I think isn't exactly how ƙ works, but it's probably how you'd want to do it in BQN.)
tfw copilot suggests that Vyxal have a center of mass builtin
21:27
storm finally decided to let me get my wifi back o/
oh i just tried ⊐⊸⊔ dyadically and yeah ⊐⊸⊔ is exactly the same as ¹ƙ
Right, okay. So it's basically the same thing except that ƙ is a quick rather than an atom.
yep
for no reason whatsoever
in this case it was useful to map over the groups but 90% of the time it just gets used with the identity function
oh it also looks like you can actually delete elements by putting -1 in the array on the left of , which would skip a lot of the fancy dancing with prepending zeroes then removing the first group at the end
@UnrelatedString Yeah, I used that in my 53-byte answer.
21:40
The other thing that seems to be different is that BQN puts empty lists in if there are numbers that don't appear in the left argument of . I assume this isn't the case in Jelly because porting your solution directly would require me to filter out these empty lists.
ah
makes sense but unfortunate
ƙ does basically have an implicit
Here is a port without the filter
Could maybe be golfed more, but I'm going to try a different approach than the "count preceding zeros" instead.
yeah that sounds wise
21:48
Actually, since I was 95% of the way there already, here's the full ported version with the filter: 45 44 bytes
Anyone want to make an esolang themed around baby goats? Was tryna search for ais523's I/D machine and accidentally invented the "kid machine"
Sounds like it would be Gaot--
@NoHaxJustRadvylf lol
@UnrelatedString Initial version with reworked logic comes in at 40 bytes
22:01
so not too bad then
@DLosc This reminds me of my REDWOLF esolang
Oh yeah I vaguely remember that lol
IIRC it was inspired by "NP down? REEEDDDWOOOOOLLF!!!!11!11!1"
But now NP actually works. Why did you do that?
22:29
Copilot's favourite case: downpercase
@NoHaxJustRadvylf upgoat
@emanresuA Do you want me to explain your Pip code? ;P
@UnrelatedString Posted
@DLosc Sure
So `'.*?'` is a regex matching ', any number of characters (as few as possible), and another '.
+ applies the + quantifier to that regex as a whole, resulting in `(?:'.*?')+`
This is equivalent to Neil's regex.
a@ finds all matches, as before
TM trims the first and last character off of each match
22:40
Oh okay thanks
and then R"''"'' works as before
I feel like there's gotta be a shorter way to do that
(I can't think of a shorter way to do that replacement and it's bugging me a bit)
Ninja'd with style
R''X2'' is the same length
R`''`;@_ is longer because it needs the semicolon
Same for H_ and S_

« first day (4184 days earlier)      last day (956 days later) »