Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Advent of Code 2023 is nigh

adventofcode.com

Post reply on HN
Print view View original

111–120 of 319 posts

Re: Advent of Code 2023 is nigh

#111

Post by frantathefranta »

jjice wrote: post #101
shever73 wrote: post #92

As others have said, part 2 of today's was really difficult. I finally solved it using Python regex `overlapped=true`, but it was very tricky. The irritation of having all of the test cases passing, but it failing for my challenge input! I hope it doesn't scare off newcomers, but I already know a few who have given up on part 2.

Same here. I would've really like if the spec specifically mentioned the possibility of that one edge case ahead of time instead of having to sift through the 1000 lines of input. No hate on AOC though, I really respect all the hard work that goes into it.

Yeah I was sort of lucky that my edge case was on the last line, still took a full hour of wondering why my solution was wrong though.

Re: Advent of Code 2023 is nigh

#112

Post by mqus »

Frummy wrote: post #73
sebtron wrote: post #65

Earlier quoted context omitted.

Ok, I'll be that guy... what edge case? My part 2 was 4-line addition to part 1 (see code below, if inappropriate let me know and I'll remove it), and it worked first try. On the other hand, I made a mistake in part one and got it right only on a second attempt... (definition of nums[] omitted) for (j = 1; j

Edge cases such as "1oneight", "3sevenine", "sevenine3" (I made these up to highlight)

Thanks! I thought I had the most straight-forward code there is and there were no edge cases that I hit. But I didn't use regex or something, just manual pattern matching like gp.

Imho regexes are overused for such stuff, precisely because they might do a lot of things you don't think about and are usually a lot slower than manually implementing what you want.

Re: Advent of Code 2023 is nigh

#113

Post by shever73 »

jjice wrote: post #101
shever73 wrote: post #92

As others have said, part 2 of today's was really difficult. I finally solved it using Python regex `overlapped=true`, but it was very tricky. The irritation of having all of the test cases passing, but it failing for my challenge input! I hope it doesn't scare off newcomers, but I already know a few who have given up on part 2.

Same here. I would've really like if the spec specifically mentioned the possibility of that one edge case ahead of time instead of having to sift through the 1000 lines of input. No hate on AOC though, I really respect all the hard work that goes into it.

> No hate on AOC though, I really respect all the hard work that goes into it.

100% with you on this. Last year was the first year I'd had time to complete it, and I always love the challenge.

Re: Advent of Code 2023 is nigh

#114

Post by RodgerTheGreat »

bryancoxwell wrote: post #25

I think I’d like to try this year’s in a language I haven’t touched before. What languages should I consider if I want something paradigmatically different from Go, Python, etc?

You could try Lil:

https://beyondloom.com/tools/trylil.html

Lil is a synthesis between ideas from functional languages, array-oriented languages like APL, and relational languages like SQL. For example, it has a first-class database-style "table" type, a query syntax that generalizes across lists, dictionaries, and tables, and arithmetic operators automatically "spread" between scalar and listy values.

Lil is available as a standalone CLI or browser-based interpreter, but it is also distributed as part of Decker, a graphical rapid prototyping environment:

https://beyondloom.com/decker/index.html

Re: Advent of Code 2023 is nigh

#115

Post by Mountain_Skies »

pciexpgpu wrote: post #104
rrishi wrote: post #74

Genuinely curios why people are so into AoC ... Feels leetcode-y

I wonder if there is a more productive form of it that helps learn/practice difficult technologies in a short amount of time. Say a combo of Ray tracing in a weekend, a game with reinforcement learning etc that helps people learn disparate technologies that they may not be familiar with while staying grounded on practical yet difficult problems. FWIW AoC is great and anything that keeps the core CS spirit going seems…

Advent of Cyber is that for infosec. It covers a broad range of security topics but is designed primarily to teach the basics of each topic rather than to be a test of advanced skills.

Re: Advent of Code 2023 is nigh

#117

Post by sokoloff »

Frummy wrote: post #73
sebtron wrote: post #65

Earlier quoted context omitted.

Ok, I'll be that guy... what edge case? My part 2 was 4-line addition to part 1 (see code below, if inappropriate let me know and I'll remove it), and it worked first try. On the other hand, I made a mistake in part one and got it right only on a second attempt... (definition of nums[] omitted) for (j = 1; j

Edge cases such as "1oneight", "3sevenine", "sevenine3" (I made these up to highlight)

I’d wager that most implementations would be fine on “sevenine3”, as they’d likely get 73 for the row, whether or not they found the 9.

Re: Advent of Code 2023 is nigh

#118

Post by wjholden »

frantathefranta wrote: post #111
jjice wrote: post #101

Earlier quoted context omitted.

Same here. I would've really like if the spec specifically mentioned the possibility of that one edge case ahead of time instead of having to sift through the 1000 lines of input. No hate on AOC though, I really respect all the hard work that goes into it.

Yeah I was sort of lucky that my edge case was on the last line, still took a full hour of wondering why my solution was wrong though.

What was this edge case you encountered? My code worked...after I finally read the problem closely enough.

Re: Advent of Code 2023 is nigh

#119

Post by Chobilet »

Without going into spoilers, its interesting that people jumped to regex to solve this. For me that was a fairly non intuitive when I first saw the problem(both parts).

What jumped to me is the problem statement indicated a finite number of states and I crafted a solution based on that information. But its really cool to see how we all jump to different implementations.

Re: Advent of Code 2023 is nigh

#120

Post by Mountain_Skies »

kristaps wrote: post #8

Last year there were people solving the puzzles with LLMs, but I don't think I saw anyone get past day 5 or so. I'm interested in how well it goes this year. Please reply if you are trying yourself or can link to public attempts by others

Part two of today's problem makes me wonder if they're trying to come up with puzzles that aren't easy for LLMs to complete but might end up making things that also discourage humans from playing.
Post reply on HN