:snowflake: Day 9 of Advent of Code is here! Join ...

# advent-of-code

[Alyona Chernyaeva]

❄️ Day 9 of Advent of Code is here! Join @Sebastian Aigner and @kropp as they solve today’s puzzle in Kotlin. Tune into the livestream today at 17:00 UTC!
🔗 https://www.youtube.com/live/7ct8eBdpavQ

[Anirudh]

I've reloaded a few times but I can't see YT chat

[Jonathan Kolberg]

Please enable livestream in YT

[Anirudh]

ok, now chat is working\

[ephemient]

it just came up in the stream but I think I'm not allowed to post links:

[ephemient]

I did implement a Java agent which causes arithmetic overflows to throw exceptions, https://github.com/ephemient/kotlin-numeric/tree/main/agent

[Jonathan Kolberg]

Yeah can be I just mentioned your agent @ephemient, not sure if it's visible

[ephemient]

it's not really packaged up for public use though

[Jonathan Kolberg]

I have a personal rule to always use Long and if that overflows you'll find out

[ephemient]

https://kotlinlang.slack.com/archives/C87V9MQFK/p1672167437730299?thread_ts=1671963784.115849&cid=C87V9MQFK

[ephemient]

still haven't looked into making a compiler plugin…

[Jonathan Kolberg]

Now I remember why I remembered it

[Anirudh]

fwiw, I think I know why the descriptions are so "bad" - it's not by mistake. ppl who paste that into LLMs will fail until they break it down to a "real" problem

[Jaap Beetstra]

I think part2 works for Sebastian because he doesn’t join consecutive empty blocks together. Then his algorithm doesn’t consider that empty part long enough. See https://github.com/SebastianAigner/advent-of-code-2024-behind-the-scenes/issues/1

[Sebastian Aigner]

Whoa, thanks for the investigation @Jaap Beetstra! Appreciate you diving into this! 🫶🏼