Advent of Code 2020 - BaseA Style (Day 11)

[Jean-Balteryx]

Discussion thread for day 11 of the Advent of Code : https://adventofcode.com/2020/day/11

[jdunkerley79]

Not the fastest or cleanest for part 2...

Spoiler

Iterative macros all the way...

Part 2 (slow and messy :():

[AdamR]

I've not solved it yet, but if anyone wants more background reading this is a modified version of Conway's Game of Life which I did play once before in Alteryx 🙂

https://www.youtube.com/watch?v=hcrVS5YOrm4&feature=emb_title

[dsmdavid]

I'm almost embarrassed by those ugly generate rows. Almost.

Spoiler

 Same iterative solves both, just update the rows that enter it. 

[AdamR]

At last a spatial problem!

Spoiler

Base workflow

Iterative Macro

Part 2 is going to need a refactor!

[AdamR]

And here it is after the refactor 🙂

Spoiler

Pretty spatial data

Parent workflow

Iterative Macro

[Balders]

One macro for both parts for me. My hint would be to work out your seat pairings list before iterating through the occupancy changes.

Spoiler

Main workflow. Creating the seat pairs was for part 2 was long a took a little debugging. One missed tickbox on a multi-row grouping had me stumbling for a while.

A few notes on the solution:

- I used row and col fields to find seat pairings, but then combined those into a single seatid field before going into the macro.
- I added ring of empty "." spaces around the room to avoid wraparound issues on the edges.
- Once spaces have been given an ID, empty "." are not needed in the iteration checks, they always stay "." and don't affect seating choices so they're filtered out before going into the macros or working out the next seat in part 2. 



Iterative Macro. Half of the tools here are just creating a proper escape for the iterations as opposed to pair matching.

[patrick_digan]

Mine was a mess today, but I got there in the end.

Spoiler

Part 2. My macro isn't complicated, but the price I pay is time. Part 2 took 5 minutes in its entirety. I spilt it into 2 parts. 1 part in the main workflow figures out where each seat needs to check (brute force), and then the iterative macro does all of the iterations (again, very brute force). 

i used my normal approach with these AOC problems using a single long string. It makes it fairly easy to work with and bebug, but It just isn't performant when you have to replicate the string a bunch. Glad this one's behind me solved. I'm sure the Alteryx Abacus would make it much more performant.

[NedHarding]

Mine was way to messy and hacked together for me to post...  Wasn't that hard in theory, in practice it drove me crazy debugging it.

[peter_gb]

Quite happy with mine, my final solution makes up for the nightmare mis-read of part 1! Didn't resort to reddit or here today 🙂
I had to leave this for a while to go and work after part 1 - wasted an hour by misinterpreting the part one as look 8 seats in each direction, not 8 seats, 1 in each direction.... DOH!

Iterative macro for part 1, then an iterative to prepare for part 2 by looking for the next seat in each direction for each seat as this doesn't change.

Then a modification of my part 1 macro to solve part 2. <1 min so not too horrible either.

Full

Spoiler

    

macro1

Spoiler

macro2 to prep part 2

Spoiler

 

 

macro 3 to solve part 2

Spoiler

    

 

  Edit - also had fun seeing how the seats filled up:-

Spoiler

 

https://public.tableau.com/profile/petergb#!/vizhome/Day11_16077110205010/AoCDay11