« first day (4826 days earlier)   

1:00 AM
Sandbox posts last active a week ago: Weave two lists, cycling if necessary
 
1:32 AM
Wow I sure love having to specify that STDIN should actually come from STDIN when using Gradle
great build tool yes
tasks.named<JavaExec>("run") {
    standardInput = System.`in`
}
that's really great stuff
could you imagine having a build tool that actually used STDIN?
that'd be ridiculous. silly, even
(if you don't do ^, you get Exception in thread "main" java.util.NoSuchElementException: No line found when trying to read from a Scanner)
 
 
1 hour later…
3:07 AM
@RydwolfPrograms ...i did not realize this was a thing i needed to do by midnight tonight
whoooooooooo
i just realized i need to do my dad's for 2021 too to be able to finish mine for 2021 which i need to be able to triple check mine for 2022 which i probably need for 2023
 
Tbw
3:22 AM
I've had people complaining about the explanation of the 1-2-3-5-7... challenge so much that I've changed it twice and it's still confusing people. To those that do understand it, is there a better way of writing it? Maybe like the comment I just left?
 
3:34 AM
Idk how I forgot to mention this but I went to prom!
It was awesome
I went with a group of friends, and we took pictures and ate at a restaurant beforehand and it was so amazing
We were buddy-system for transport and not getting lost in the big dark room full of neon lights, and I was paired up with one of my trans guy friends and he was dressed in this black velvet dress with snake bite piercings and he looked like a total vampire and it was so cool lol
And the clothes my dad had for me turned out to be a full on tuxedo and not some gross old dress shirt like I half expected which was cool
 
Tuxwolf programs
 
I'd post one or two of the pics 'cause they came out AWESOME (one of our friends lives with a photographer) but they all have other people in them
 
If you have a photo app that supports subject selection you could just copy you onto a blank photo or something
Like that tap to select feature that uses ai to determine an outline of what to copy
Or you could crop/apply a blur or something like that
 
Ooh wait I do have some of just me (most of which are me trying to climb a tree lol), I just don't have access to them right now
Here's me cropped out of one lol
(we were doing the charlie's angels pose thing)
And then since I was in New York I had no idea my roommates planned a whole entire after party, so when we got back the apartment was full of people, a DJ, lots of illegal stuff, two guys knife fighting, just an average day living on your own at 18 lol
 
4:01 AM
@RydwolfPrograms rip
 
4:17 AM
@Tbw an example sequence with an explanation might be nice, but i can't see anything confusing about the spec
 
Tbw
Ok thanks. The only issue with that is it is somewhat involved to prove that any such sequence actually is one, since all the submitted answers involve mapping the positive integer to an abelian group, such that the primes with 1 are in bijection with the group, and the sequence is the elements that map to the identity. Then you have to prove that this actually makes a correct answer.
 
4:46 AM
 
 
1 hour later…
6:05 AM
Is there an efficient way to calculate the smallest Y such that Y! > X, where X is large? Specifically I'm interested for X=10^1000
Seems brute forcing is fast enough, Y=450
 
I was thinking of Newton's method.
x! = Γ(x+1)
 
6:26 AM
Q: Given three sorted arrays, find the k-th smallest element using limited access to the array.
I: [Read the three arrays bypassing API](https://uoj.ac/submission/685752), [rob the arrays and feed main a fake input](https://uoj.ac/submission/396637), or just [totally skip main](https://uoj.ac/submission/616101)
 
6:43 AM
lol
testing system that links user code and tester into a single program always has issues like that
you can pass pretty much every JS problem in Codewars with console.log('<PASSED::>'); process.exit() (but please don't, they know it and you'll be penalized heavily)
 
@DLosc, not sure if you have already done this but managed to prove Turing-completeness for Acc!! by making a bf -> Acc!! translator
 
Neat, I worked out the theory for that a while ago when trying to make a (transpiled multiple times) tinylisp interpreter in acc!! for one of DLosc's bounties
 
@Bubbler Note the 88be24b2c5f6af2bc828c1c5101034a5. It was designed such that the first line in output must match a secret string. But we found way to find it out
 
wtf lol
 
Is that... encoding the nth bit of the secret string as a pass/fail on the testcase?
 
6:54 AM
No it's designed to anti userwriteoutput
 

« first day (4826 days earlier)