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
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
|
package moe.nea.firmament.features.mining
import java.util.*
import kotlin.time.Duration
import moe.nea.firmament.util.TimeMark
class Histogram<T>(
val maxSize: Int,
val maxDuration: Duration,
) {
data class OrderedTimestamp(val timestamp: TimeMark, val order: Int) : Comparable<OrderedTimestamp> {
override fun compareTo(other: OrderedTimestamp): Int {
val o = timestamp.compareTo(other.timestamp)
if (o != 0) return o
return order.compareTo(other.order)
}
}
val size: Int get() = dataPoints.size
private val dataPoints: NavigableMap<OrderedTimestamp, T> = TreeMap()
private var order = Int.MIN_VALUE
fun record(entry: T, timestamp: TimeMark = TimeMark.now()) {
dataPoints[OrderedTimestamp(timestamp, order++)] = entry
trim()
}
fun oldestUpdate(): TimeMark {
trim()
return if (dataPoints.isEmpty()) TimeMark.now() else dataPoints.firstKey().timestamp
}
fun latestUpdate(): TimeMark {
trim()
return if (dataPoints.isEmpty()) TimeMark.farPast() else dataPoints.lastKey().timestamp
}
fun averagePer(valueExtractor: (T) -> Double, perDuration: Duration): Double? {
return aggregate(
seed = 0.0,
operator = { accumulator, entry, _ -> accumulator + valueExtractor(entry) },
finish = { sum, beginning, end ->
val timespan = end - beginning
if (timespan > perDuration)
sum / (timespan / perDuration)
else null
})
}
fun <V, R> aggregate(
seed: V,
operator: (V, T, TimeMark) -> V,
finish: (V, TimeMark, TimeMark) -> R
): R? {
trim()
var accumulator = seed
var min: TimeMark? = null
var max: TimeMark? = null
dataPoints.forEach { (key, value) ->
max = key.timestamp
if (min == null)
min = key.timestamp
accumulator = operator(accumulator, value, key.timestamp)
}
if (min == null)
return null
return finish(accumulator, min!!, max!!)
}
private fun trim() {
while (maxSize < dataPoints.size) {
dataPoints.pollFirstEntry()
}
dataPoints.headMap(OrderedTimestamp(TimeMark.ago(maxDuration), Int.MAX_VALUE)).clear()
}
}
|