KSLCore/ksl.app.comparison

Package-level declarations

Types

Analysis types the comparison analyzer can produce. Each one has its own validation rule against the current selection — see ComparisonSelectionModel.validateForResponse.

BatchCompletedComparisonSource

class BatchCompletedComparisonSource(result: RunResult.BatchCompleted, val sourceLabel: String = defaultLabel(result)) : ComparisonDataSourceIfc

Scenario-host adapter that exposes a completed scenario sweep — the RunResult.BatchCompleted surfaced through whichever controller owns lastResult — as a ComparisonDataSourceIfc for the comparison analyzer UI.

ComparisonDataSourceIfc

interface ComparisonDataSourceIfc

Read-only access to a collection of "experiments" whose per-replication response observations can be pulled for cross-experiment comparison analyses (box plots, multiple- comparison analysis, confidence-interval plots).

ComparisonSelectionModel

class ComparisonSelectionModel(val sources: List<ComparisonDataSourceIfc>)

Shared mutable state for the comparison analyzer's experiments-first workflow.

ExperimentRow

data class ExperimentRow(val name: String, val modelIdentifier: String, val numReplications: Int, val responses: List<ResponseRow>)

One experiment exposed by a ComparisonDataSourceIfc.

InMemoryComparisonSource

class InMemoryComparisonSource : ComparisonDataSourceIfc

Test fixture implementation of ComparisonDataSourceIfc backed by an in-memory map. Used by the substrate's own tests and available to any host (Swing, web, CLI, headless) that wants to drive a comparison UI without spinning up a real run.

KSLDatabaseComparisonSource

class KSLDatabaseComparisonSource(db: KSLDatabase, val sourceLabel: String = defaultLabel(db)) : ComparisonDataSourceIfc

Database-backed adapter exposing the experiments stored in a KSLDatabase as a ComparisonDataSourceIfc — the third member of the comparison-source family alongside InMemoryComparisonSource (test fixture / programmatic use) and BatchCompletedComparisonSource (live scenario-sweep result).

ResponseCategory

enum ResponseCategory : Enum<ResponseCategory>

Distinguishes the three kinds of measurable model elements whose per-replication values are eligible for cross-experiment comparison.

ResponseRow

data class ResponseRow(val name: String, val category: ResponseCategory)

One response / counter exposed by an ExperimentRow.

ValidationResult

data class ValidationResult(val ok: Boolean, val reason: String? = null)

Result of pre-flight validation for a given (selection, analysis) pair. When ok is false, reason carries a short user-facing explanation; the host's Generate affordance disables with this text shown as a tooltip / status line.

Functions

recordingResponse

fun List<ExperimentRow>.recordingResponse(responseName: String): List<ExperimentRow>

Experiments in this list that record a response named responseName. Used to drive the "Recorded by N of M" column in the response table and to define the participant set for each render.

unionOfResponses

fun List<ExperimentRow>.unionOfResponses(): List<ResponseRow>

Union of responses across this list of experiments, sorted by name with no duplicates. When two experiments record a response under the same name but different categories (rare; usually indicates a model schema collision), the first occurrence wins — the UI can surface the discrepancy separately if needed.