day02.md

Day 2

all / 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

Available as an RSS Feed

Prompt / Code / Rendered

Day 2, not too bad for Haskell either :)

There is some fun in parsing here:

data Policy = P
    { pIx1  :: Int
    , pIx2  :: Int
    , pChar :: Char
    , pPass :: String
    }

parsePolicy :: String -> Maybe Policy
parsePolicy str = do
    [ixes,c:_,pwd] <- pure $ words str
    [ix1,ix2]      <- pure $ splitOn "-" ixes
    P <$> readMaybe ix1
      <*> readMaybe ix2
      <*> pure c
      <*> pure pwd

I used one of my more regular do-block tricks: if you pattern match in a Maybe do-block, then failed pattern matches will turn the whole thing into a Nothing. So if any of those list literal pattern matches failed, the whole block will return Nothing.

In any case, we just need to write a function to check if a given policy is valid for either criteria:

countTrue :: (a -> Bool) -> [a] -> Int
countTrue p = length . filter p

validate1 :: Policy -> Bool
validate1 P{..} = n >= pIx1 && n <= pIx2
  where
    n = countTrue (== pChar) pPass

validate2 :: Policy -> Bool
validate2 P{..} = n == 1
  where
    n = countTrue (== pChar) [pPass !! (pIx1 - 1), pPass !! (pIx2 - 1)]

And so parts 1 and 2 are just a count of how many policies are true :)

part1 :: [Policy] -> Int
part1 = countTrue validate1

part2 :: [Policy] -> Int
part2 = countTrue validate2

Back to all reflections for 2020

Day 2 Benchmarks

>> Day 02a
benchmarking...
time                 55.69 μs   (55.61 μs .. 55.78 μs)
                     1.000 R²   (1.000 R² .. 1.000 R²)
mean                 55.89 μs   (55.82 μs .. 56.03 μs)
std dev              323.1 ns   (232.5 ns .. 422.3 ns)

* parsing and formatting times excluded

>> Day 02b
benchmarking...
time                 42.96 μs   (42.88 μs .. 43.06 μs)
                     1.000 R²   (1.000 R² .. 1.000 R²)
mean                 43.11 μs   (43.07 μs .. 43.19 μs)
std dev              196.8 ns   (94.94 ns .. 332.4 ns)

* parsing and formatting times excluded