#AdVENt of Code
tranquil vapor
sullen fiber
i change this to crystalline?
sullen fiber
ohh
native pewter
tranquil vapor
compiling this takes longer than compiling crystal itself lol
way longer
its still compiling
warped dust
vee cannot wait two minutes
tranquil vapor
THIS IS A TERRIBLE NAME https://github.com/samueleaton/sentry
FLOSS devs will stop naming their shit so that it seems like telemetry
tranquil vapor
vee cannot wait two minutes
yop
warped dust
FLOSS devs will stop naming their shit so that it seems like telemetry
vee not everyone is a webdev
sullen fiber
why isnt this showing me an error
warped dust
she's having a reality shock
sullen fiber
is this valid
sullen fiber
holy shit
tranquil vapor
it might split it into chunks of 3 chars
native pewter
is this valid
tranquil vapor
/run ```cr
print "explosionsssssoooooooo".split 3
nah
split heartBOT
Here is your cr(0.36.1) output @tranquil vapor
["explosionsssssoooooooo"]```i only need intellisense
tranquil vapor
qhar
sullen fiber
tranquil vapor
wtf does it do then
/run ```cr
print "explosio3nsssssoooooooo".split 3
does it just cast to string
split heartBOT
Here is your cr(0.36.1) output @tranquil vapor
["explosio3nsssssoooooooo"]```Makes an array by splitting the string on any ASCII whitespace characters (and removing that whitespace).
tranquil vapor
???
tranquil vapor
/run ```cr
print "i am going to explode".split 3
split heartBOT
Here is your cr(0.36.1) output @tranquil vapor
["i", "am", "going to explode"]```ohhhh so real
sullen fiber
/run cr print "abc def ghi jkl def iek".split 3
tranquil vapor
fire
split heartBOT
Here is your cr(0.36.1) output @sullen fiber
["abc", "def", "ghi jkl def iek"]```wait thats actually fire
native pewter
Makes an array by splitting the string on any ASCII whitespace characters (and r...
mb i saw too late!
split heartBOT
Here is your cr(0.36.1) output @sullen fiber
["abc def ghi jkl def iek"]```/run ```cr
print "real::banana::lover".split "::", 2
Here is your cr(0.36.1) output @tranquil vapor
["real", "banana::lover"]```thats so real
sullen fiber
who needs DOCS
when you have LSP
im definitely going to never use the parentheses-less syntax
tranquil vapor
we need to write eval func in crystal
tranquil vapor
im definitely going to never use the parentheses-less syntax
bluepilled...
why not
its soo based
crystal/ruby naming convention is so funny
you suffix your method names with
? to indicate a boolean func
! to indicate a dangerosu action
/run ```cr
print " banana".strip!.includes? "ban"
@tranquil vapor I received cr(0.36.1) compile errors
Showing last frame. Use --error-trace for full trace.
In file0.code:1:17
1 | print " banana".strip!.includes? "ban"
^-----
Error: undefined method 'strip!' for String
Did you mean 'strip'?
/piston/packages/crystal/0.36.1/run: line 4: ./out: No such file or directory
! means inplace
warped dust
ur mom dangerous
native pewter
/run ```cr
print "ven".strip!
this is the worst part of crystal
sullen fiber
wouldnt it mean mutation
split heartBOT
@native pewter I received cr(0.36.1) compile errors
Showing last frame. Use --error-trace for full trace.
In file0.code:1:13
1 | print "ven".strip!
^-----
Error: undefined method 'strip!' for String
Did you mean 'strip'?
/piston/packages/crystal/0.36.1/run: line 4: ./out: No such file or directory
no!!!!
tranquil vapor
i hate when u always have to include lang in ur searches
sullen fiber
is this a real lsp
tranquil vapor
LOL
its kinda bad lol
wait did u enable intellisense in settings
but yeah naming convention is cool
sullen fiber
i have these
tranquil vapor
/run ```cr
Array(Int64).new.unsafe_fetch 10
@tranquil vapor I only received cr(0.36.1) error output
Invalid memory access (signal 11) at address 0x50
[0x557638b392f6] *Exception::CallStack::print_backtrace:(Int32 | Nil) +118
[0x557638b2cbac] __crystal_sigfault_handler +316
[0x7fbe5fe70730] ???
[0x557638b95336] *Pointer(Int64) +6
[0x557638b9531a] *Array(Int64) +10
[0x557638b1ec50] __crystal_main +1120
[0x557638b96396] *Crystal::main_user_code<Int32, Pointer(Pointer(UInt8))>:Nil +6
[0x557638b9620c] *Crystal::main<Int32, Pointer(Pointer(UInt8))>:Int32 +44
[0x557638b2a0b6] main +6
[0x7fbe5fa4109b] __libc_start_main +235
[0x557638b1e72a] _start +42
[0x0] ???
FIRE
why does it have that method lmaoo
whats the point of unsafe array methods
tranquil vapor
crystal is strange cause it has like everything you could ever ask for and still didnt become famous
it has so much shit
native pewter
yerah,
tranquil vapor
fire language
native pewter
so many features and almost c speed too
sullen fiber
ok now how run crystl
tranquil vapor
crystal banana.cr
tranquil vapor
LSP so bad that it's faster to run the compiler to see if there are errors than use the IDE
sullen fiber
this istrue
tranquil vapor
for some reason compile takes like a second but error reporting is fucking instant
is it cause the slow part is llvm and it first does type checking then llvm
i think llvm might be the slow part cause rust is also fucking slow to compile
tranquil vapor
WRONG CHAT GIRLIE
native pewter
`crystal banana.cr`
when u gonna shards
steady fog
@tranquil vapor hi
tranquil vapor
hiii
steady fog
Crystal so funny
@tranquil vapor my day 5 solution uses so much ram
systemd oom killer kills it
tranquil vapor
is day 5 the one with the cards
steady fog
nop
tranquil vapor
systemd oom killer kills it
on the normal input? or large input
steady fog
okay
native pewter
I'm making massive arrays
issue
tranquil vapor
soon
tranquil vapor
did u write that
native pewter
<@959816030216151140>
i don even get whats going on
solemn python
my day 6 solution is a oneliner
Makes sense, it's just math which is easy to oneline
Unless you did it the boring way
native pewter
Unless you did it the boring way
i did it the boring way because theres nothing stopping you from doing it
kinda dum
like at least yesterdays brute force took a few minutes
tranquil vapor
you love
solemn python
i did it the boring way because theres nothing stopping you from doing it
I did too at first but then I did the smart way to deal with the Large Input™
tranquil vapor
0ms
steady fog
tranquil vapor
fedora
its literally instant
give biggerrrrr input @steady fog
generate 100mb input
steady fog
howw
steady fog
yop
steady fog
@tranquil vapor did you use crystal conccurrencyu
tranquil vapor
i thought it was larger input
tranquil vapor
<@343383572805058560> did you use crystal conccurrencyu
no
synchronous
its just algorithm i came up with
do u brute force just do the entire range
i tried that and it took like 10 minutes and my result was wrong anyway
steady fog
i dont know how i would brute force
tranquil vapor
brute force would just be manually testing every single seed number in the ranges
xd
i did that first it was sooo slow
native pewter
how did nice make it worse lmfao
tranquil vapor
wtf is nice
native pewter
WHAT
tranquil vapor
what now
native pewter
how do u not know nice...
tranquil vapor
why does nice make it seem slower
native pewter
i have no idea.
tranquil vapor
nice seems slightly slower
native pewter
nice seems slightly slower
go look at man page
just doing nice lowers its priority i think?
tranquil vapor
nuh uh
tranquil vapor
VENCORD IS INAPPROPRIATE
native pewter
yes.
we had CFS and niceness in uni exercise recently :3
(and then they changed scheduler)
sullen fiber
if it adds 20 to the niceness doesnt that mean lower priority
native pewter
it goes -20 to 19
sullen fiber
oh i was reading the args wrong
i thought -n means put the number wher n is
and that nice -20 means add 20
native pewter
honestly idek because nothing i did seemed to make it faster
solemn python
@sullen fiber you love
solemn python
off by how much
solemn python
do you want the reason
native pewter
ur kinda silly
solemn python
waiting
solemn python
- Your plugged in distance should actually be distance + 1
- Floor the higher solution and ceil the lower one before subtracting
- Add one to the final result
thats the method that works for me after cheating by looking at reddit
sullen fiber
are u not using like bigints
solemn python
I am
sullen fiber
what do u mean "ceil" and "floor"
solemn python
solemn python
what do u mean "ceil" and "floor"
ceil: round up
floor: round down
sullen fiber
how do u round an integer
solemn python
the equation is continuous
the solutions aren't integers
ok well to be specific
I am using Irrational BigInts
from some library I found
sullen fiber
oh ur using bigfloats
solemn python
what lang are you using
sullen fiber
ts
solemn python
ah
sullen fiber
i was thinking
solemn python
well just check the rounding on your bigint math, it probably defaults to floor or something
sullen fiber
maybe i can calculate sqrt(x) mod 1
of a bigint x
im using newton's method for approximating sqrt(x) i'd probably have to like
reimpl it to somehow retain that information
meh too much brainage i dont wanna think about modifying newton's method to retain sqrt(x) mod 1
u know
i could probably implement bigfloats
but brain fried
proper totem
too much coding of advents for the vap
solemn python
ok making p1 large-input-able was ez, just had to replace my let acc = 1usize; with a let acc = BigUInt::from(1u8);
the actual race #s arent big enough to justify replacing all my floats with arbitrary-precision stuff
steady fog
@solemn python
native pewter
arent you like torturing yourself with rust
also you dont need to feel bad 
as long as you learned something you finished aoc!
ill probably give up too some time bc im not that smart xd
solar tide
Unless you did it the boring way
daring marlinBOT
https://github.com/taskylizard/adventofcode/blob/trunk/2023/6/day.cr#
**day.cr: **
input = File.read("./input.txt").strip.lines
def part1(input : Array(String))
time, distance = input.map(&.split[1..])
time.zip(distance).product { |time, distance| (0_i64..time.to_i).count { |i| i * (time.to_i - i) > distance.to_i } }
end
def part2(input : Array(String))
time, distance = input.map(&.split[1..].join)
puts (0_i64..time.to_i64).count { |i| i * (time.to_i64 - i) > distance.to_i64 }
end
puts part1(input)
puts part2(input)
CRYSTAL YUSS
COUNT LMAO
you did the stupidest possible brute force I love it
i at least binary search only the beginning and end of the winning numbers and then add them together O.O
checking every single number is funny
native pewter
theres no penalty so its intended 
tranquil vapor
https://github.com/taskylizard/adventofcode/blob/trunk/2023/6/day.cr#
ur aoc runner seems cool
i would totally make smth similar but i use many different languages so it's pointless
native pewter
wth so much effort
solemn python
In my opinion, your AoC setup is half the event :)
If you rotate languages then of course that has its own merits, but if you don't, you better make it count
native pewter
i mean i get wantin to make proper projects but i wouldnt say aocs goal is to finish as fast as possible
its just gamified like that
warped dust
In my opinion, your AoC setup is half the event :)
If you rotate languages then ...
in my opinion ur insane
solemn python
In my opinion I agree
solemn python
i mean i get wantin to make proper projects but i wouldnt say aocs goal is to fi...
No not in terms of speed
warped dust
anyone making an elaborate setup to finish it faster is a loser
solemn python
Just in terms of making fun projects that aren't just 25 files in a row
solemn python
anyone making an elaborate setup to finish it faster is a loser
So true
solar tide
ur aoc runner seems cool
thanks, it's also published on npm
solar tide
i would totally make smth similar but i use many different languages so it's poi...
it has templates support so you can do this
steady fog
@tranquil vapor lol it finally finished
my brute force took 21 minutes
works though
solemn python
for what
steady fog
day 5 p2
solemn python
p2 normal input?
steady fog
wdym
native pewter
they really out here misspelling advertisements
steady fog
crystal
native pewter
my brute force took 21 minutes
so slow,
steady fog
i could make it go faster by using threads
but then i need to find a way to set a max amount
native pewter
cheater
steady fog
nuh uh
steady fog
damn the adrenaline goes crazy tho if you actually plan to do it fast
youre like 14
solemn python
this is literally a fucking physics problem
the most basic physics problem ever
but sure
anyways d7 time
warped dust
why are you all brute forcing then
solemn python
why are you all brute forcing then
because it takes less than a second to do and its the easy copy paste solution
steady fog
less than a second to write you mean?
warped dust
but its lame
steady fog
easy to understand
solemn python
I did the maths after initially solving it w/ brute force
proper totem
vap did the maths
warped dust
🙏
proper totem
they solved it in desmos iirc
warped dust
my goat
warped dust
they're references
steady fog
same thing
warped dust
what do you expect
steady fog
you
warped dust
every string method works on them like normal
solemn python
hm this one doesn't look too bad but im sure p2 is hell
proper totem
guh so much info to parse
hard for LLM and hard for human
probably doesn't help that I've never played poker
steady fog
error[E0277]: the size for values of type str cannot be known at compilation time
what
die
i regret this already
warped dust
rewriting my website in crystal
warped dust
nvm it isnt
steady fog
kill rini kill
steady fog
@warped dust
solemn python
Full house, where three cards have the same label, and the remaining two cards share a different label: 23332
does this mean the 2 remaining cards have to be the same label or is 33321 a full house
delicate crane
super sick and cant think right now even if I tried
skipping aoc today
real hinge
It is
steady fog
java is better
real hinge
Guh I have to have breakfast and go to school in 30 mind
steady fog
thread 'main' panicked at ./day-07-rust.rs
44:
called Result::unwrap() on an Err value: ParseIntError { kind: InvalidDigit }
note: run with RUST_BACKTRACE=1 environment variable to display a backtrace
shut the fuck
delicate crane
super sick and cant think right now even if I tried
skipping aoc today
https://github.com/CodeF53/AdventOfCode/blob/main/src/2023/07.ts
heres how far I got before realizing brain no worky
sullen fiber
i just woek up fgrom tha nap
real hinge
I dont know how to play that game
sullen fiber
<@454072114492866560> tell me u did the maths
yes i did but
i did it in desmos first
then in uiua after submitting
felt dirty...
this looks fire
teal mountain
today was ok
sullen fiber
im do in uiua
native pewter
what the fuck it doesnt build anymore
WHAT THE FUCK I JUST DID 2015
im gonna kms
this is so rarded
solemn python
angy
native pewter
fuck me.
solemn python
oh wait im an idiot
native pewter
2015 day 7```crystal
require "log"
def one(input : String) : Int32 | Nil
wires = Hash(String, String).new
input.split("\n").each do |line|
spl = line.split(" -> ")
wires[spl[1]] = spl[0]
end
calcOp(wires, Hash(String, Int32).new, "a")
end
def calcOp(wires : Hash(String, String), known : Hash(String, Int32), op : String) : Int32
if known.has_key?(op)
return known[op]
end
if /\d+/.match(op)
return op.to_i32
end
if wires.has_key?(op)
spl = wires[op].split(" ")
if spl.size == 1
known[op] = calcOp(wires, known, spl[0])
elsif spl.size == 2
known[op] = ~calcOp(wires, known, spl[1])
elsif spl.size == 3
if spl[1] == "AND"
known[op] = calcOp(wires, known, spl[0]) & calcOp(wires, known, spl[2])
elsif spl[1] == "OR"
known[op] = calcOp(wires, known, spl[0]) | calcOp(wires, known, spl[2])
elsif spl[1] == "LSHIFT"
known[op] = calcOp(wires, known, spl[0]) << calcOp(wires, known, spl[2])
elsif spl[1] == "RSHIFT"
known[op] = calcOp(wires, known, spl[0]) >> calcOp(wires, known, spl[2])
end
end
return known[op]
end
return 0
end
def two(input : String) : Int32 | Nil
wires = Hash(String, String).new
input.split("\n").each do |line|
spl = line.split(" -> ")
wires[spl[1]] = spl[0]
end
a = calcOp(wires, Hash(String, Int32).new, "a")
known = Hash(String, Int32).new
known["b"] = a
calcOp(wires, known, "a")
end
why do I keep doing this
steady fog
lina moment
harsh marsh
ballmer peak experiment failed
sullen fiber
guh ok back time to code for real
proper totem
that could be neat
teal mountain
harsh marsh
the example was really weak this time
i submitted like 3 times wrong before getting it
all passed the example
teal mountain
part 1 ||251136060||, part 2 ||249400220||
teal mountain
the example was really weak this time
agreed
teal mountain
i had one wrong submission on p2, but i realised that there was so many conditions that the example wasn't touching
teal mountain
You dont need to spoil
i know, i do just in case because some people don't even want a ballpark
native pewter
time for bed.
teal mountain
@harsh marsh what the fuck 34 second delta
damn
oh wait that was yesterdays
oops
sullen fiber
lole
teal mountain
i actually got the p2 logic like 98% fully correct on my first write
but i spent like 2 mins debugging smth stupid
my approach was basically to ||have a frequency table of hands, and add J's frequency to the most frequent hand, but i wasn't accounting for 5x J's||
harsh marsh
I would have gotten like a 30 second delta but i forgot about some really dumb standard library behavior
||the 5Js|| gave me an exception the first time I ran so I caught that
proper totem
dkfjndkfjnkjnKjnKjndksjnvkDJNKJNVKDJFNKDFJVDKFJVDFVKJN
||my answer was wrong because I wrote a new sorting function for the different strengths with joker but forgot to use it instead of my original one||
nvm that was apparently just one of many issue 
blehhh
I don't know what I changed but it works now
sullen fiber
i love uiua updating and then failing at the last second after 10m of fetching an dcompiling
proper totem
||my answer was wrong because I wrote a new sorting function for the different s...
okay so this did actually solve it i just ran with my input but looked at aeth's answer instead and thought i had it wrong because of that
my brain's smartness knows no bounds
sullen fiber
huge
solemn python
aaaaaaaaaaaaaAAAAAAAAAAAAAAAAAAAAa my example is correct but not actual input
and my code is absolute hell to debug
ok found my issue
success
my code is shit but it works so don't judge me
solar tide
where is your code
solar tide
my code is ok
solar tide
had to write a frequency and a sort function though
did today in Typescript
actually I was thinking, maybe I'll rewrite this in civet or coffee
sullen fiber
are u gonna pr it or just make for fun
solemn python
ill make a large input and you can choose to add it or not idc
I just want to test my code
sullen fiber
real
i wont cus i want the generators in the repo
i wonder what a large input looks like
solemn python
hm
well probably just a crap ton of ranomized cards + inflate the bids some fun number of digits for comedic bignum effect
I guess I won't do it myself then
I should sleep
if I had more time I would pr it in ts
solemn python
how so
sullen fiber
what happend
hazy sinew
i was COOKING in part 2
sullen fiber
eagely playing ranked aoc
hazy sinew
instead of just doing ||cards.replace("J", "23456789TQKA")|| i tried to ||do some + or - depending on the amount of js and the amount of other uniuqe characters and get it that way (horrible idea)||
hazy sinew
eagely playing ranked aoc
i at least wanna win on some private leaderboards
im #3 here
steady fog
@solemn python tyminky
solemn python
@steady fog zeeter
solemn python
i at least wanna win on some private leaderboards
Well take pride in knowing you are definitely winning MB
Not that it's much of a competition
I'm like the only one who does days directly after release
solemn python
Competitive advent of code participator (derogatory)
hazy sinew
ok for the next 30 minutes im gonna be deleting code and seeing if anything changes
steady fog
@solemn python what lang are you doing
solemn python
<@487443883127472129> what lang are you doing
Rust
teal mountain
but eyy you finally got it, gj
solemn python
I love rust as a language and this is my first AoC so I'm not doing anything crazy
Just sticking to what I know for now
hazy sinew
I would have gotten like a 30 second delta but i forgot about some really dumb s...
i had a solution implemented for part 2 in about 30 seconds too except it was wrong and i shouldve just went with the simpler one (||replacing every J with 23456789TQKA||)
wouldve gotten me like #200
excited for tomorrow tho because its gonna be super easy
steady fog
i keep forgetting to push my code
proper totem
instead of just doing ||cards.replace("J", "23456789TQKA")|| i tried to ||do som...
||you could just add the number of J to the count of the card that was most common in hand||
steady fog
@proper totem vscode plugin that reminds me in a way I notice
solemn python
||you could just add the number of J to the count of the card that was most comm...
What about like full houses
proper totem
that doesn't change anything
hazy sinew
||you could just add the number of J to the count of the card that was most comm...
||i didnt calculate the most common cards in the hand, i just did hand.groupingBy { it }.eachCount { it.value == the number im checking rn }||
steady fog
I'm using rust too
proper totem
having a J will never get you a full house because it could get you a four of kind instead which is higher
hazy sinew
what about 23J32
solemn python
tyminky are you using regex to identify the type of hand
||I convert the hand into a vec of card repetitions (KK111 = 3,2 62881 = 2,1,1,1)||
solemn python
||I convert the hand into a vec of card repetitions (KK111 = 3,2 62881 = 2,1,1,1...
Then match exact formats
proper totem
||just sort by most cards add j to first in array||
and you'll get a full house but it wouldn't change it because it doesn't matter which card you have three or two of
hazy sinew
heres my refactored approach that i ended up using (not the orginal one)
||
private fun valueOf(s: String): Int {
val hand = s.map { if (it == 'J') "23456789TQKA" else it.toString() }.joinToString("")
val uniques = s.replace("J", "").distinct().count()
val counts = hand.groupingBy { it }.eachCount()
val score = when {
counts.any { it.value == 5 } -> 6
counts.any { it.value == 4 } -> 5
uniques == 2 && counts.any { it.value == 3 } -> 4
counts.any { it.value == 3 } -> 3
uniques == 3 && counts.any { it.value == 2 } -> 2
uniques == 4 -> 1
else -> 0
}
return score
}
```||if i realized that i can just ||replace all Js with 23456789TQKA|| then i wouldve had part 2 in like 22 minutes
proper totem
nvm wouldn't make it more performant probably since I'd need to add more logic for it
hazy sinew
also how did u guys compare by strength
solemn python
Wdym?
proper totem
||const strengths = "23456789TJQKA" compare indexof card and sort`||
hazy sinew
bruh
hazy sinew
||
fun compareHands(hand1: String, hand2: String): Pair<String, String> {
val cardsHand1 = hand1.map { charToCardRank(it) }
val cardsHand2 = hand2.map { charToCardRank(it) }
for (i in 0 until 5) {
if (cardsHand1[i] != cardsHand2[i]) {
return if (cardsHand1[i] > cardsHand2[i])
Pair(hand1, hand2)
else
Pair(hand2, hand1)
}
}
return Pair(hand1, hand2)
}
fun charToCardRank(ch: Char): CardRank {
return when (ch) {
'A' -> CardRank.ACE
'K' -> CardRank.KING
'Q' -> CardRank.QUEEN
'T' -> CardRank.TEN
'9' -> CardRank.NINE
'8' -> CardRank.EIGHT
'7' -> CardRank.SEVEN
'6' -> CardRank.SIX
'5' -> CardRank.FIVE
'4' -> CardRank.FOUR
'3' -> CardRank.THREE
'2' -> CardRank.TWO
'J' -> CardRank.JOKER
else -> throw IllegalArgumentException("Invalid card character: $ch")
}
}
enum class CardRank(val displayName: String, val value: Int) {
ACE("Ace", 14),
KING("King", 13),
QUEEN("Queen", 12),
TEN("10", 10),
NINE("9", 9),
EIGHT("8", 8),
SEVEN("7", 7),
SIX("6", 6),
FIVE("5", 5),
FOUR("4", 4),
THREE("3", 3),
TWO("2", 2),
JOKER("Joker", 1);
}
```||
this is where i spent most of part 1
steady fog
I need to make another enumm
solemn python
I love enums but that's not the way to use them 
steady fog
||
```
fun compareHands(hand1: String, hand2: String): Pair<String, String> ...
if you wanna be extra fancy you can implement Comparable
hand > hand2
then
hazy sinew
wtf bro Exception in thread "main" java.lang.IllegalArgumentException: Comparison method violates its general contract!
i didnt sign no contract
hazy sinew
i think its cuz i did ||"23456789TQKA".indexOf(hand) and not .indexOf(card)||
teal mountain
heres my refactored approach that i ended up using (not the orginal one)
||
```k...
holy pattern matching 😩
teal mountain
||
```
fun compareHands(hand1: String, hand2: String): Pair<String, String> ...
i just had an array of strings and compared indices
hazy sinew
so ur telling me i couldve done this instead of making an enum class, a converter and a comparison with a for loop?
||```kotlin
fun compareHands(hand1: String, hand2: String): Pair<String, String> {
val cardOrder = "J23456789TQKA"
return hand1.zip(hand2)
.find { it.first != it.second }
?.let { if (cardOrder.indexOf(it.first) < cardOrder.indexOf(it.second)) Pair(hand1, hand2) else Pair(hand2, hand1) }
?: Pair(hand1, hand2)
}
😨
hazy sinew
i was staring at my input for an hour trying to figure out whats wrong, manually comparing everything to find the edge case and then ended up just ||replacing J with 23456789TQKA||
teal mountain
||```ts
const HANDS = [
'A', 'K', 'Q', 'J', 'T', '9', '8', '7', '6', '5', '4', '3', '2'
];
const getType = (card: Card): number => {
const freq: Record<string, number> = {};
for (const hand of card.hand) {
freq[hand] ??= 0;
freq[hand]++;
}
const k = Object.keys(freq);
const v = Object.values(freq);
let num = k.length;
if (num === 2 && v.includes(1)) return num;
if (num === 2 && v.includes(2)) num += 0.5;
if (num === 3 && !v.includes(3)) num += 0.5;
return num;
};
const compare = (a: Card, b: Card): number => {
const aType = getType(a);
const bType = getType(b);
if (aType < bType) return 1;
if (aType > bType) return -1;
for (let i = 0; i < 5; i++) {
const aRank = HANDS.indexOf(a.hand[i]);
const bRank = HANDS.indexOf(b.hand[i]);
if (aRank < bRank) return 1;
if (aRank > bRank) return -1;
}
return 0;
};
its so nice seeing all of these be gray
sullen fiber
just try not doing that
wait u were actually right
hazy sinew
so what was the edge case
sullen fiber
i still have edge cases
i just misread the text
||i was sorting hands like "KTJJT" -> "JJTTK" to do the strength comparisons||
teal mountain
today's text didn't have too much bullshit which i'm glad for. it was a lot to read in general tbh
sullen fiber
||but i was supposed to be comparing strength in the order it was dealt||
really unsure what the edge cases are tho
hazy sinew
really unsure what the edge cases are tho
have fun
its all gonna be correct
except 1 single mismatch
because all the cards are different
so one of them might not be handled correctly but only that specific scenario
sullen fiber
yea i just have no way of telling which is being handled wrong
teal mountain
yeah the sample for today sucked, i posted my big input + answer somewhere above if u need it
hazy sinew
have fun
sullen fiber
true
hazy sinew
u can send me ur input and ill give u the correct list back
so u can find the edge case
sullen fiber
ya that would help actually
hazy sinew
dm
solar tide
today's text didn't have too much bullshit which i'm glad for. it was a lot to r...
yeah it was straightforward but long as fuck
sullen fiber
apparently ive been doing it right the whole time
and im summing wrong
but i get the example right
so HWO
HOW
hazy sinew
me when vap sorts the list correctly, sorts the bids correctly but doesnt know elementary school math
we should create a better example input
sullen fiber
LMAO
@warped dust help
i spent 2 hours
on part 1 today
because i didnt know the difference between indices and ordinals
guys rate my part 1
hazy sinew
2345A 2
2345J 5
J345A 3
32T3K 7
T55J5 17
KK677 11
KTJJT 23
QQQJA 19
JJJJJ 29
JAAAA 37
AAAAJ 43
AAAAA 53
2AAAA 13
2JJJJ 41
JJJJ2 31
can yall run this and make sure part 2 is sorted ascending
and the answer is 3667
i forgot about the fact that 2 prime numbers can have the same lcm
sullen fiber
true u need coprimes
sullen fiber
because what if u read it and it spoils
😦
actually
you need prime coprimes
set of primes that is setwise coprime
teal mountain
nerd
hazy sinew
you need prime coprimes
im instead just gonna hope nobody encounters this issue
sullen fiber
true
hazy sinew
sullen fiber
everything is a one liner
hazy sinew
wtf is this guys i did 2022 in november 2023 and im first
sullen fiber
well its kinda obvious look at the stars
hazy sinew
how did nobody here complete 2022
real hinge
I just got bored after like day 13-14
problems were getting very hard and because ı was not enjoying I told myself why continue
sullen fiber
and the answer is 3667
yea
sullen fiber
SHUT UP
tranquil vapor
but its lame
MATH IS EVEN LAMER
hazy sinew
i helped like 30 people with my example input
maybe i should start making one every day if the one provided is shit
real hinge
how does that work
do we only check first different letter
???
@hazy sinew tell
real hinge
okay
real hinge
That's not the right answer; your answer is too high. If you're stuck, make sure you're using the full input data; there are also some general tips on the about page, or you can ask for hints on the subreddit. Please wait one minute before trying again. [Return to Day 7]
WHY PART 2 IS WRONG
native pewter
@rotund moth
hazy sinew
day 6 part 2 should have been
As the race is about to start, you realize the piece of paper with race times and record distances you got earlier actually just has very bad kerning. There's really only one race - ignore the spaces between the numbers on each line.
You realize that your boat would move much faster if you also manually pushed it.
To manually push your boat, you need to let go of the button for 5 milliseconds, and then give it a good shove. Pushing your boat will double its speed for the first 50 milliseconds.
How many ways can you beat the record in this one much longer race, while also being able to push the boat?
stop dreaming
real hinge
bad
hazy sinew
what
real hinge
supposed to get 6839
hazy sinew
print the list and compare
real hinge
doing
tranquil vapor
it's cause you didn't end ur print with a newline
your terminal auto inserted a newline
real hinge
bad
tranquil vapor
yop
hazy sinew
hear me out
tranquil vapor
eagely has gone fully insane from advent of code
tranquil vapor
HEAR ME OUT
mantika's favourite language
hazy sinew
mantikas favorite langugae is probably whatever windows uses
real hinge
nıp
tranquil vapor
fix
hazy sinew
I am getting wrong answer
send list from my input
real hinge
JAVA IS LOOOOVELY
real hinge
send list from my input
it looks so different
hazy sinew
send
real hinge
2345A 1
J345A 2
2345J 3
32T3K 5
KK677 7
T3Q33 11
Q2KJJ 13
T3T3J 17
Q2Q2Q 19
2AAAA 23
T55J5 29
QQQJA 31
KTJJT 34
JJJJJ 37
JJJJ2 41
JAAAA 43
2JJJJ 53
AAAAJ 59
AAAAA 61
hazy sinew
dont replace the most common number with j
cuz then ur gonna end up failing when sorting
joker is only here to pretend to be another number, not actually be another number
hazy sinew
print the non replaced sorted one
real hinge
this shows non reoplaced ones
wait
somethings off
oaky here
I see the problem I think
hazy sinew
send ur logic for 4 of a kind
real hinge
I just need to figure out how to fix it
hazy sinew
send ur entire eval function
hazy sinew
ok
real hinge
also this is my logic
hazy sinew
send the thing that actually checks what a hand is
real hinge
thank god
the problem was
I was first calculating character counts, and putting them into some vector
then I was removing "J" element and add its count to element at 0
problem is if J has the highest count, I was making J 2x of it and removing it
then any J would disappear and therefore boom
hazy sinew
wait cant u u just do part2Score(hand) = part1Score(hand.replace("J", "")) + hand.count("J")
someone try this im too lazy
hazy sinew
not with ur logic
only later
private fun valueOf(hand: String): Int {
val uniques = hand.replace("23456789TQKA", "").distinct().count()
val counts = hand.groupingBy { it }.eachCount()
val score = when {
counts.any { it.value == 5 } -> 6
counts.any { it.value == 4 } -> 5
uniques == 2 && counts.any { it.value == 3 } -> 4
counts.any { it.value == 3 } -> 3
uniques == 3 && counts.any { it.value == 2 } -> 2
uniques == 4 -> 1
else -> 0
}
return score
}
p2score = valueOf(hand.replace("J", "")
- if (valueOf(hand) != 6) hand.count("J") else 6
tho i did basically the same thing except p2score = valueOf(hand.replace("J", "23456789TQKA")) and then removed that string again for the uniques
real hinge
@tranquil vapor dont do todays aoc
tranquil vapor
<@343383572805058560> dont do todays aoc
too late
real hinge
|| ```
1:parse inputs
2: count all characters, then order them by length of each character; if character count is same order them by strength
3:loop over game modes and convert character count to game type
4: sort the results by game type, if game type is same loop over characters and order by strength
5: calculate the output
real hinge
answer
tranquil vapor
nooob
real hinge
I will kill you
tranquil vapor
jk i didnt
but it shouldnt be too hard tbh
i will code now
i wonder what the best way to store data is
maybe class
real hinge
You made me lose fatih on myself
tranquil vapor
very useful docs
real hinge
class Person
end
tranquil vapor
class Hand
getter :cards, :bid, :rank
def initialize(cards : String, bid : Int32)
@cards = cards
@bid = bid
end
def self.from(line : String)
cards, bid = line.split " "
self.new(cards, bid.to_i)
end
end
cursed syntax
INSANE
/run ```cr
puts "explode".succ
Here is your cr(0.36.1) output @tranquil vapor
explodf
s u c c
hazy sinew
```cr
class Hand
getter :cards, :bid, :rank
def initialize(cards : String, ...
booilerplate
tranquil vapor
do u ever write code, run it, and it immediately works
wow part 2 is easy as fuck with my approach
i love my solution so much
its clean
tranquil vapor
its clean
def solve(file)
hands = File.read_lines(file).map do |line|
Hand.from line
end
puts file, hands.sort!.map_with_index { |h, i| h.bid * (i + 1) }.sum
end
i just implemented comparison operator in my class
@real hinge today is very easy and cozy
wdym
real hinge
<@287555395151593473> today is very easy and cozy
ı had pain
tranquil vapor
part 2 done
so easy
with oop
easy imo
mmmm
@real hinge did u finish day 7
check my solution
SO CLEAN
one of the rare times OOP is op
cause i just swapped out the card list and rank calculate func via inheritence
real hinge
<@287555395151593473> did u finish day 7
I did
tranquil vapor
it really is easy
i cleaned up further
now its really readable
WAIT I FORGOT TO SUBMIT MY ANSWER LOL
oh not right actually
maybe i forgot smth
i think i know why
yep
small flaws where i might || count jokers twice ||
native pewter
🤡
tranquil vapor
WHY DOES THIS RETURN RETURN FROM PARENT
harsh marsh
because its just in a block not a function
tranquil vapor
how do u return from block...
harsh marsh
you don't they're blocks
tranquil vapor
RETURN VALUE
real hinge
https://github.com/mantikafasi/advent-of-code-2023/blob/master/day07/src/main.rs @tranquil vapor read my solution
tranquil vapor
nop
tranquil vapor
first i need to finish mine
tranquil vapor
GUHH I HAVE TO WAIT 5 MINUTES
u will send me ur input and part 2 solution @real hinge
do
real hinge
u will send me ur input and part 2 solution <@287555395151593473>
okay
249748283
248029057
tranquil vapor
GUH
tranquil vapor
small flaws where i might || count jokers twice ||
no lmao xd
tranquil vapor
nop im close
real hinge
okay
real hinge
you werent this happy 6 minutes ago
tranquil vapor
well i take it back it wasn't that easy, but once i realised my mistake of || counting jokers multiple times || it was easy to fix
/run ```cr
puts "u love this method"
print "399Q9".chars.tally, "\n"
print "399Q9".chars.tally.values.sort! { |a, b| b <=> a }
Here is your cr(0.36.1) output @tranquil vapor
u love this method
{'3' => 1, '9' => 3, 'Q' => 1}
[3, 1, 1]```@real hinge we need to add proper threading and async await syntax to crystal
tbh u can probably easily implement async await with a library
tranquil vapor
tbh u can probably easily implement async await with a library
promise = async do
9 + 10
end
puts await promise # 21
doubt
dont you need to implement threads in compiler level
tranquil vapor
crystal already has coroutines
u can just use those to implement async await
for real threading yes you'd likely need to edit compiler or make a C library
tranquil vapor
malware
sullen fiber
Under trans
real hinge
https://cdn.discordapp.com/attachments/1180052078631071827/1182247083038744656/S...
people should write in uiua instead of obfuscating
this is so unreadable fr
teal mountain
new malware bundles the uiua compiler and uiua source programs
sullen fiber
TRUE
sullen fiber
Soon I will start writing Uiua byte code
teal mountain
you should do aoc in ijvm bytecode
sullen fiber
What is ijvm
teal mountain
a subset of the jvm ISA that works exclusively on integers
you get a stack and you get integer ops
teal mountain
real hinge
https://raw.githubusercontent.com/alchzh/advent-of-code/master/2021/day01/1.ws
websocket language
hazy sinew
https://en.wikipedia.org/wiki/IJVM#IJVM_Instructions
2019
harsh marsh
intcode was fun
hazy sinew
Soon I will start writing Uiua byte code
its probably easier than writing actual uiua
sullen fiber
It's probably about the same
Worse cus some of the utilities u get in Uiua don't exist in the byte code cus they're inlined
I will miss fork
hazy sinew
can u do 🎄🎄x in bytecode
hazy sinew
RISE
hazy sinew
ur subconciously convincing me to use uiua
sullen fiber
😁
tranquil vapor
solve @sullen fiber https://adventofcode.com/2024/day/8
solar tide
<@287555395151593473> we need to add proper threading and async await syntax to ...
there exists a shard for this for the latter
sullen fiber
solve <@454072114492866560> https://adventofcode.com/2024/day/8
🙁
solemn python
I did one or two AoC days in whitespace
Hear me out: do part 1 in a normal language and part 2 in whitespace, all in one file
harsh marsh
ooh that's a good idea
...maybe the other way around though
it'd have to be a language that doesn't care about whitespace very much like C
sullen fiber
I'm gonna have to do the next week late since I'm in Florida
I'mma have to like walk to a library to do them
warped dust
uiua mobile keyboard
warped dust
vim on termux (do not
sullen fiber
proper totem
I'm gonna have to do the next week late since I'm in Florida
my condolances
solemn python
vim on termux (do not
)
steady fog
@solemn python
@solemn python @solemn python @solemn python @solemn python @solemn python @solemn python @solemn python
real hinge
Can someone graph their input
wdym
sullen fiber
Like a node graph
real hinge
oh
real hinge
I have zero idea how I am gonna do this
last year tried brute force and ended up with billions of searches
steady fog
i am using java today
real hinge
good