I did the first one easy enough, but I completely misunderstood how the problem dampener worked and wasted too much time on it

This is my kotlin script for part 2

import java.io.File
import kotlin.math.abs

fun safeP(increasing: Boolean, pair: Pair<Int, Int>): Boolean =
    (abs(pair.second - pair.first) in 1..3) && (increasing == pair.second - pair.first > 0)

val lines: List<String> = File("day02-input").readLines().map { it.trim() }.filterNot { it == "" }
val sequences: List<List<Int>> = lines.map { it.split(" ").map { it.toInt() } }

fun getProgressions(seq: List<Int>) =
    generateSequence(-1) { if (it + 1 != seq.size) it + 1 else null }
        .map { progVariation: Int ->
            seq.filterIndexed { index, _ -> index != progVariation }
                .let { alternateSeq: List<Int> ->
                    alternateSeq.mapIndexed { i: Int, level: Int ->
                        if (i != alternateSeq.lastIndex) level to alternateSeq[i + 1] else null
                    }
                }
                .filterNotNull()
        }

val answer = sequences.count { sequence: List<Int> ->
    getProgressions(sequence).any { prog: List<Pair<Int, Int>> ->
        prog.all { safeP(true, it) } || prog.all { safeP(false, it) }
    }
}

println(answer)

I'm still only learning kotlin. In my opinion it's pretty shitty code due to the long call chains. I imagine to make it better I should probably break up those chains, assigining the intermediate values to different variables, to make it more readable. Do you agree and is there other stuff that you think ought to be improved?