this post was submitted on 01 Dec 2023
136 points (95.9% liked)

Advent Of Code

770 readers
1 users here now

An unofficial home for the advent of code community on programming.dev!

Advent of Code is an annual Advent calendar of small programming puzzles for a variety of skill sets and skill levels that can be solved in any programming language you like.

AoC 2023

Solution Threads

M T W T F S S
1 2 3
4 5 6 7 8 9 10
11 12 13 14 15 16 17
18 19 20 21 22 23 24
25

Rules/Guidelines

Relevant Communities

Relevant Links

Credits

Icon base by Lorc under CC BY 3.0 with modifications to add a gradient

console.log('Hello World')

founded 1 year ago
MODERATORS
 
all 26 comments
sorted by: hot top controversial new old
[–] [email protected] 23 points 11 months ago

It's not a good sign that this meme appeared on day 1 already...

[–] stifle867 12 points 11 months ago (1 children)

It's great! I've had community members point me in the right direction after already "solving" it incorrectly. It really makes you think about it. You have to expand your test cases and really come up with a better solution.

Not to say it isn't difficult especially if you expect the problem to be described perfectly accurately.

[–] [email protected] 11 points 11 months ago (1 children)

I don't need a perfect instruction set, but dang if the examples couldn't be better sometimes. Like sixteen was in there to show it only counted for 6, but nothing with overlapping text.

[–] csh83669 8 points 11 months ago (1 children)

The example for me immediately showed my overlap bug with “eightwo”. There aren’t too many other ways to make this ten words overlap. 🙂

[–] stifle867 7 points 11 months ago (2 children)

The problem is the example is actually eightwothree which comes out as 83 so if you replace from start to finish the example passes but the solution is incorrect.

[–] [email protected] 1 points 11 months ago (1 children)

At this point you're just complaining that the edge case is not highlighted in red.

I think it's the right amount of pointers to make you aware of the issue without straight up telling you.

[–] stifle867 5 points 11 months ago

I'm not really sure how to interpret your comment but I'll try my best. The edge case that causes some solutions to fail does not have any definition on how to handle it on the problem page. In other words, it does not state anywhere whether the correct interpretation of 1threeight is meant to be 18 or 13. If your solution replaces the words to numbers from left to right you end up with 13 as the value but it's meant to be 18.

The example answers don't cover this but you will realise something is wrong if you run it against your full problem. Community has been very helpful on providing pointers.

[–] [email protected] -1 points 11 months ago* (last edited 11 months ago) (1 children)

Just don't replace, or replace only the first letter with the numeral

[–] DaleGribble88 1 points 11 months ago (2 children)

You can't just replace the first letter either, because depending on the order of your replacements, you could be replacing the end of another number. (Encountered this exact problem trying to optimize my solution.)

[–] CameronDev 4 points 11 months ago (1 children)

I replaced the second letter, none of them overlap 2 letters.

[–] [email protected] 0 points 11 months ago
[–] [email protected] 1 points 11 months ago (1 children)

As long as you replace any spelled out numbers from left to right it should work

[–] stifle867 1 points 11 months ago (1 children)

2oneight - if you replace from left to right you get 21ight or 21. This doesn't work for part 2 as the answer should br 28.

[–] [email protected] 1 points 11 months ago* (last edited 11 months ago)

We're talking about just the first letter, so

2oneight —> 21neight —> 21n8ight

[–] [email protected] 9 points 11 months ago (1 children)

I've solved part 2 multiple times now, but small intricacies keep popping up every time.

[–] Cyno 2 points 11 months ago (1 children)

I'm stuck on one of these intricacies now! I saw some of the ideas on how other people did it but i honestly have no clue what i did wrong. Got any tricky examples to share?

[–] [email protected] 3 points 11 months ago (1 children)

I just solved it. I printed every set of integers in the format "Result: [theTwoDigits], from string: [inputString]" and mannually checked all of them. Eventually found the edge case that was causing troubles. It was a typo.

[–] Cyno 4 points 11 months ago (1 children)

Oh god, sorry to hear that 😅i'm feeling desperate enough to try that, i just wrote a different implementation and i get the same (wrong) result. At this point I just want to know what i misunderstood or mistyped cuz its driving me crazy

[–] [email protected] 2 points 11 months ago (1 children)

The bug is some strings can have overlapping characters. onEight threEight fivEight. There are more cases. So if you do a search and replace your string becomes 1ight 3ight and the second number does not get found.

Possible fixes: Search and replace and add the extra letter: oneEigh then search and replace. Search and replace words to numbers but put some extra letters in just in case.

[–] Cyno 1 points 11 months ago

Thanks, I managed to find the culprit in the end however - I was using a forward-look of 5 characters for finding if it matches a word so in a string of a3two for example, it would register the two before the 3. It was an easy fix once i found the issue.

[–] Matthew 6 points 11 months ago

I was having a bitch of a time with day 1's part 2 until someone pointed out that sometimes two numbers share a first/last letter (eightwo).

[–] [email protected] 6 points 11 months ago (1 children)

Those overlapping spelled out numbers were rough

[–] purplemonkeymad 2 points 11 months ago

It was for sure to catch people that were taking the easy route and using a replace. I had to rework my solution to get the start positions for the matches. That method made both parts easy to get together.

[–] [email protected] 1 points 11 months ago

I managed the overlap logic just fine (if in a clumsy way), but spent over an hour trying to debug my crappy parser-combinator code when the issue all along was a mis-spelling in my lookup table.