Skip to content

Latest commit

 

History

History
123 lines (100 loc) · 3.57 KB

Reflection 2023.md

File metadata and controls

123 lines (100 loc) · 3.57 KB
Raw

Reflection

Day 01

150/611; points: 0, --th place

  • digits was amazing, sucks I misread the question
  • Had some issues with overloads, will have to look into that.
  • Had issues with recording because I still had "0" bound as start stop recording.

Day 02

265/125; points: 0, --th place

  • My project was a bit too cursed, so I tried fixing it 15 minutes before the contest.
    • This was dumb; I broke it fully and had to use Python instead

Day 03

84/125; points: 17, 275th place

  • First points
  • My code was slow, cause I used a list instead of a set (I wanted to use a set, but messed up)
  • make code use asSet where applicable, and make that method warn if a collection is passed multiple times, but isn't a set
  • Apparently the slowdown has from the fact that SimpleGrid had a hashcode and equals

Day 04

201/103; points: 0, 324th place

  • I really should run my code on the test input first
  • I accidentally use python syntax instead of kotlin syntax

Day 05

41/2; points: 159, 86th place

  • LET's FUCKING GO
  • Fix overload issue between stdlib and utils (min/max)

Day 06

221/127; points: 0, 104th place

Day 07

218/141; points: 0, 121st place

Day 08

45/48; points: 109, 75th place

Day 09

164/107; points: 0, 88th place

Day 10

33/168; points: 68, 71st place

Day 11

5/47; points: 150, 47th place

  • Nice points: TOP 50
  • Didn't use GitHub copilot today,
  • Convert a list of BoundedGridPoints to Points easily

Day 12

197/20; points: 81, 41st place

  • Nice
  • People used more bruteforce for day one to go faster but that cost them for part 2
  • Could have used yComb
  • Add memoize functions
  • Allow yComb to be memoized easily

Day 13

53/81; points: 78, 39th place

  • A bit slow I think?
    • Didn't sleep that well
    • I didn't do my finger stretched today, idk if that's related
  • I don't think using a grid here helped me
  • find IS firstOfNull, that's why there is no findOrNull

Day 14

267/167; points: 0, 43rd place

  • Didn't sleep well

Day 15

232/172; points: 0, 50th place

  • Slept better, but I don't think I was fully awake. I really need to use a full hour to wake up and not just lie in my bed

Day 16

14/24; points: 164, 36th place

  • Still didn't sleep well, but better morning routine might have helped
  • Grids again??
  • leftEdge and friends were broken
    • lines of length 0 raise errors where they kinda shouldn't (maybe do add a warning?)
    • bounds' top left and family were wrong

Day 17

17/17; points: 168, 30th place

  • 17, 17, 17, 17, 17, 17, 17, 17, 17, 17,
  • Dijkstra result has start and end switched.

Day 18

4/48; Points 150, 28th place

  • I forgot about pick's smh, also messed up my shoelace :[
  • I think using the floodfill for p1 was still the better call, based on the assumption I'd need to use the wall colors

Day 19

341/41; Points 60, 27th place

  • using a map instead of a list would have avoided problems, or just converting the char to an index in 1 location would have been great
  • expand yComb

Day 20

40/52; Points 150, 23rd place

Day 21

662/9; Points 90, 24th place

  • Didn't sleep well, probably why I made so many mistakes in p1
  • Accidentally counted the evens in p2 instead of the odds

Day 22

67/72; Points 63, 23rd place

  • Didn't sleep well, probably why I made so many mistakes in p1 and p2

Day 23

125/53; Points 48,

  • Didn't sleep well
  • yComb can have stack overflow, using coroutines would be better
  • roi should be poi
  • roi/poi is slow. at least make a special case for removing nodes with 2 neighbours
  • longest no backtrack path? idk