refactor: extracted processing of child scores #8
Conversation
…CalculationStrategy in a separate method. - removed redundant loops - added documentation
| ) | ||
| ) | ||
| } | ||
| hasIncompleteResults = true |
There was a problem hiding this comment.
instead of setting this explicitly you could make it an auto-computed property on the SeparatedKpiResults and check if either list does not has elements. (or if missingEdgeWeights is > 0.0 [if that's a valid constraint])
There was a problem hiding this comment.
actually no.
Neither the failed list nor missingEdgeWeights tells us this information.
This flag should indicate if there was any incomplete (Incomplete, Failed, Error) element in the initial childScores. However, in non strict mode, we cast Incomplete elements to Success elements and continue calculation. Thus, the failed list can be empty in this case, the missingEdgeWeights is 0.0, however, there was an incomplete input.
This happens for this input childScores = listOf(KpiResult.Incomplete(..), 1.0) calculateKpi(childScores, strict = false). Here we want to continue our calculation due to strict = false, however, we still want to indicate that the calculation is based on an incomplete result. Thus the flag.
This is something to maybe improve on later or somehow make it more explicit.
extracted processing of child scores in CalculationStrategy in a separate method.