Last active
February 11, 2024 22:33
-
-
Save ccocchi/02db5921bf8da4cffa78 to your computer and use it in GitHub Desktop.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
case class Metric(id: String, aggregationOperation: AggregationOperation) | |
case class Dimension(id: String) | |
object AggregationOperation extends Enumeration { | |
type AggregationOperation = Value | |
val Sum, Average, Min, Max, Count = Value | |
} | |
/** | |
* Think of a RowSelection as a typed SQL table. We use this selection to | |
* modelize OLAP cubes. | |
* | |
* Headers are the names of each column, with an additionnal information about | |
* wether we are dealing with a dimension or a metric. | |
* | |
* Rows are double dimensions arrays. | |
* Each row is viewed as 2 concatened sequence: one of strings for dimension values, the others of numerics. | |
* | |
* Ex: | |
* ["ga:origin", "ga:browser"] ["ga:visits", "ga:timeOnSite"] | |
* ["France", "Chrome"] [20.0, 345.5] | |
* | |
*/ | |
case class RowSelection( | |
metricHeaders: Seq[Metric], | |
dimensionHeaders: Seq[Dimension], | |
metricRows: Seq[IndexedSeq[Double]], | |
dimensionRows: Seq[IndexedSeq[String]]) { | |
def take(n: Int): RowSelection = { | |
/// To be implemented | |
} | |
def ++(other: Selection): RowSelection = { | |
/// To be implemented | |
} | |
def filter(col: Dimension, f: (String) => Boolean): RowSelection = { | |
/// To be implemented | |
} | |
def groupByColumns(cols: Seq[Dimension]): RowSelection = { | |
///To be implemented | |
} | |
} | |
class RowSelectionSpec extends FunSpec with ShouldMatchers { | |
val metric1 = Metric("m1", AggregationOperation.Sum) | |
val metric2 = Metric("m2", AggregationOperation.Average) | |
val dimension1 = Dimension("d1") | |
val dimension2 = Dimension("d2") | |
val selection = RowSelection( | |
Seq(metric1, metric2), | |
Seq(dimension1, dimension2), | |
Seq(IndexedSeq(1.0, 3.0), IndexedSeq(2.0, 9.0)), | |
Seq(IndexedSeq("dimVal11", "dimVal12"), IndexedSeq("dimVal21", "dimVal22")) | |
) | |
describe("RowSelection") { | |
it should "take the n first rows" in { | |
val taken = selection.take(1) | |
taken.metricRows should have size(1) | |
taken.dimensionRows should have size(1) | |
} | |
it should "add a selection to another selection" in { | |
val other = selection | |
val added = selection ++ other | |
added.metricHeaders should equal(selection.metricHeaders) | |
added.dimensionHeaders should equal(selection.dimensionHeaders) | |
added.metricRows should equal(Seq(IndexedSeq(1.0, 3.0), IndexedSeq(2.0, 9.0), IndexedSeq(1.0, 3.0), IndexedSeq(2.0, 9.0))) | |
added.dimensionRows should equal(Seq(IndexedSeq("dimVal11", "dimVal12"), IndexedSeq("dimVal21", "dimVal22"), IndexedSeq("dimVal11", "dimVal12"), IndexedSeq("dimVal21", "dimVal22"))) | |
} | |
it should "filter the selection" in { | |
val f: (String => Boolean) = dimVal => (dimVal == "dimVal11") | |
val filtered = selection.filter(dimension1, f) | |
filtered.metricRows should equal(Seq(IndexedSeq(1.0, 3.0)))) | |
filtered.dimensionRows should equal(Seq(IndexedSeq("dimVal11", "dimVal12"))) | |
} | |
it should "group by columns" in { | |
val s = selection.copy(dimensionRows = Seq(IndexedSeq("dimVal1", "dimVal2"), IndexedSeq("dimVal1", "dimVal2"))) | |
val grouped = s.groupByColumns(Seq(dimension1, dimension2)) | |
grouped.dimensionRows should equal(Seq(IndexedSeq("dimVal1", "dimVal2"))) | |
grouped.metricRows should equal(Seq(IndexedSeq(3.0, 6.0))) | |
} | |
} | |
} | |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment