Advent of Code 2023 Day 4 (BaseA Style)

This one was frustrating, because I had the right idea from the beginning but made a lot of mistakes along the way. I'll let this serve as my annual reminder that generating a lot of rows is often not the best approach, especially with Advent of Code!

Spoiler

Part 1: Found the points for each card by taking the 2^(num_matches - 1)

Part 2: This was a journey. I started by spinning my wheels on an iterative macro that worked on the sample data AFTER I corrected a lot of silly mistakes. I wish I could recall what those were now, but a lot of time has passed since then 😅

Once I got my macro working on the sample data, I tried the production data. I quickly realized that my approach was flawed, because the iterations were taking a long time to process. My initial approach was to use Generate Rows to create a brand new card for each copy won. I realized that I didn't actually need to create a copy of the card - I just needed to know how many would be generated. This led me to my final solution, which was to process each card individually and determine how many of the other cards it would generate (I'm confusing myself even explaining it).

I agree with @clmc9601 that there HAS to be a way to do it without a macro, but I might leave that pondering until tomorrow.

Main Workflow:

Macro:

First Macro Using Generate Rows: This was going to take forever to run. Very brute force approach in hindsight.

AOC2023_Day4_KelseyKincaid.yxzp

AncientPandaman

Spoiler

Day 04.yxzp

AkimasaKajitani

My solution.

Part 1 was very easy. But it took so much time for part 2 because I didn't understand the question correctly and wasted my time. But I am satisfied in the execution time(less than 1 sec).

Spoiler

Day04_.yxzp

IraWatt

My first macro in AoC. Definitely learned a bit from the challenge today.

Spoiler

Day 4.yxzp

Ladarthure

My solution with a quick working macro

04.yxzp

Samantha_Jayne

Spoiler

Only have time for star 1 at the moment and I have dance later... so may have to catch up on this one. Will update my pic & file when I do.

AoC_2023_SamiJayne_Day4.yxmd

caltang

Spoiler

No luck with a non-macro solution... tried doing a many-to-many join but the idea fell flat. Would love to see a different appraoch.

AncientPandaman

non-macro version.

I just random test and boom, hit the result.

Spoiler

then use multiple-rows formula and get above row based on the max winning number (which is 10).
(change the [count] to 1, as it just referred to original copy)

1+
iif([Row-1:Sum_match]>0,[Row-1:Count2],0)+
iif([Row-2:Sum_match]>1,[Row-2:Count2],0)+
iif([Row-3:Sum_match]>2,[Row-3:Count2],0)+
iif([Row-4:Sum_match]>3,[Row-4:Count2],0)+
iif([Row-5:Sum_match]>4,[Row-5:Count2],0)+
iif([Row-6:Sum_match]>5,[Row-6:Count2],0)+
iif([Row-7:Sum_match]>6,[Row-7:Count2],0)+
iif([Row-8:Sum_match]>7,[Row-8:Count2],0)+
iif([Row-9:Sum_match]>8,[Row-9:Count2],0)+
iif([Row-10:Sum_match]>9,[Row-10:Count2],0)&#8203;

then sum the count2

Screenshot 2023-12-04 181313.png

DataNath

Day 4 done and dusted! Took me a little while to debug my part 2 macro but we got there in the end!

Spoiler

Day 4.yxzp

mbarone

Part I only. Hope to work on Part II without a macro (not sure it's possible, but going to try) later.

Spoiler

2023-12-04 09_08_00-Alteryx Designer x64 - Day04.yxmd.png

Part II: Gave up on the non-macro approach, so here it is with the macro.

Spoiler

Perfect use case for a simple iterative macro - I feel good getting this one done quickly after I was a mess on yesterday's puzzle!

Spoiler

day4.yxzp

jarrodthuener

I accidentally mislabeled my batch macro, so had the opportunity to use "Find Missing Macros" function, which worked beautifully. I went the route of a batch macro with a nested iterative macro - not sure both were needed, but it gets the job done.

Spoiler

Day4.yxzp

NicoleJ

Another fun one! AND it's macro time... I do love a good iterative macro in the morning!!

Spoiler

Part 1 was pretty straightforward solve, used my new GetPart formulas for the initial parsing into cards > my hand > winning hand, and then fun to find an excuse to use the POW function (ka-POW!!!!)

Part 2 ended up being simpler than expected once I paused and made sure to annotate a few things and understand my steps first before launching into my iterative macro. I love a good iterative macro use case, and this was a perfect one!!

Day4.1.png

Day4.yxzp

mmontgomery

Day4. I Macro is brute force to solve it 'faster'

Spoiler

2023_Day4.yxzp

patrick_digan

Similar non-macro as @PangHC . I'm forever grateful to @danilang who showed me the power of the multi-row formula tool for the AOC challenges many years ago.

Spoiler

Day 4.yxmd

Kenda

Spoiler

attempted to go the non-macro route at first then threw that idea out the window

4.yxzp

DavidP

Not my best work, but it's done! Iterative macro took 13 minutes to run, but I don't care!!!

Spoiler

Day4.yxmd

AoC 2023 day4 iterate.yxmc

ccanterman

Oof - I predict I'll be good at Iterative Macros by the time the challenge is finished. Thanks to @phottovy for posting your solution file. After my attempts failed miserably, yours was super helpful! We'll see how it goes next Iterative Macro challenge, but I feel like I understand them way better than I did 2.5 hours ago

Spoiler

afv2688

This one took a while to think how to do it with no macro

Spoiler

AoC_2023_4.yxmd

FrederikE

Spoiler

04.yxzp

mceleavey

Ok, so my iterative macro was being extremely naughty, and so I set it on fire and used a macro-free brute force approach.

Don't judge me.

Spoiler

The first part was really easy. I simply stripped out the component parts, split to rows so I had a winning number per row and filtered to where one of my numbers was the same as a winner. I then simply multi-rowed a double calculation per card and boom.
Part two was a bit frustrating as I probably messed up something very straightforward with my iterative macro. The initial design simply ran through for each card number and generated the rows for each one, I then stripped out the originals and unioned the output back to the initial stream but it kept giving me extra rows and I couldn't figure out why, so I threw my hands up in the air in exasperation and brute forced it with an inelegant multi-row, using a look back for the number of rows representing the possible number of wins (10):
(if [Row-10:Count]-10>=0 then 1 else 0 endif) * [Row-10:Wins] +
(if [Row-9:Count]-9>=0 then 1 else 0 endif) *[Row-9:Wins] +
(if [Row-8:Count]-8>=0 then 1 else 0 endif) *[Row-8:Wins] +
(if [Row-7:Count]-7>=0 then 1 else 0 endif) *[Row-7:Wins] +
(if [Row-6:Count]-6>=0 then 1 else 0 endif) *[Row-6:Wins] +
(if [Row-5:Count]-5>=0 then 1 else 0 endif) *[Row-5:Wins] +
(if [Row-4:Count]-4>=0 then 1 else 0 endif) *[Row-4:Wins] +
(if [Row-3:Count]-3>=0 then 1 else 0 endif) *[Row-3:Wins] +
(if [Row-2:Count]-2>=0 then 1 else 0 endif) *[Row-2:Wins] +
(if [Row-1:Count]-1>=0 then 1 else 0 endif) *[Row-1:Wins] + 1

...which made my brain hurt so I need to lie down.