Derived-property census of GetElementProperties against the Pilot
Los mantenedores suelen responder en 1 día
Nadie ha tomado este issue todavía.
Evaluación
- Dificultad
- 4/5
- Tiempo estimado
- 3-5 días
- Aptitud para principiantes
- 55/100
Línea de trabajo
Start with test-opensysml-go/census_test.go and data/gen_census.py, then read the linked census snapshot and the rules.go property rules to understand how the measurements map to served properties. The issue suggests adding a measured column using the embedded examples; done means the census can report how often served rules return values. The issue gives no specific test command for that addition.
Escrito por el modelo de indexación a partir del texto del issue.
Descripción
Following up on #2012 and #2021: once GetElementProperties made every metamodel property readable by name, we could count, across whole models, which derived properties come back with values. We added a census to our test suite that does this for each MDK in the same way, with the Pilot as the reference column.
How it works
- The property list is every feature marked
derived="true"in the Pilot'sSysML.ecore(328 properties), per concrete metaclass. - Two corpora: our matrix models (23 files) and the Pilot's examples and training models (195 files, loaded together as one model).
- For each element, as written, the census reads every derived property its metaclass carries and records a value, an empty answer, or "not derived".
- OpenSysML is read in process through the Go client, at
071880026e.
Snapshot: census-SNAPSHOT.md. Design and notes: DISCUSSIONS.md, Step 13. It can be rerun with ./run-census.sh from test-suite/.
Summary (properties with values on at least one element / answered but always empty / not derived):
| corpus | Pilot | sysml-toolkit | OpenSysML |
|---|---|---|---|
| matrix models | 194 / 71 / 0 | 194 / 69 / 2 | 142 / 86 / 33 |
| Pilot examples | 300 / 22 / 0 | 297 / 22 / 3 | 180 / 82 / 54 |
The "not derived" group largely matches what rules.go declares as omitted, and the service's explanations for those are clear. The census suggests another group may be the most useful one to look at: properties that are listed as served but come back empty on every element where the Pilot has values. On the Pilot examples there are 66 of them. Some of the most frequent:
| property | Pilot has values | OpenSysML |
|---|---|---|
Feature::ownedTypeFeaturing |
18,843 of 20,294 | empty on all 13,953 |
InstantiationExpression::instantiatedType |
1,456 of 1,456 | empty on all 1,107 |
Type::endFeature |
782 of 21,346 | empty on all 14,751 |
Connector::association |
605 of 605 | empty on all 447 |
Import::importedElement |
408 of 408 | empty on all 325 |
Behavior::step |
220 of 220 | empty on all 169 |
Function::result |
127 of 127 | empty on all 90 |
Flow::sourceOutputFeature / targetInputFeature |
85 of 127 | empty on all 106 |
ConjugatedPortDefinition::originalPortDefinition |
79 of 79 | empty on all 39 |
SendActionUsage::payloadArgument |
34 of 35 | empty on all 23 |
The same holds for the <kind>Definition of several usage kinds (enumeration, flow, interface, case, use case, view, rendering, allocation...) and for the filtered owned/nested usage lists of Definition and Usage (ownedFlow, nestedTransition, ownedRequirement, ...). The full list is in the snapshot under "Differences from the reference".
A second group fills at a much lower rate than the Pilot, e.g. Step::behavior 5% vs 100%, Expression::function 8% vs 100%, Usage::isReference 3% vs 100%, Connector::sourceFeature 32% vs 94%. Part of this group probably reflects implied relationships: the Pilot includes them after transformation, while GetElementProperties answers with isImpliedIncluded = false (e.g. Feature::type, Type::ownedSpecialization, Type::inheritedFeature). So these numbers are a pointer rather than a verdict. Usage::isReference and the connector ends are the cases we already described in #2012.
A possible use on your side: rules.go already records, per property, the rule that serves it. A measured column next to it, filled the way this census fills it (how often a served rule returns a value over the embedded examples), would show these cases as the rules are written. You're welcome to take the census or any part of it (data/gen_census.py, test-opensysml-go/census_test.go). The Go adapter is about 230 lines and uses only the public client.
Not derived on the Pilot examples (54)
ActionDefinition::action, AllocationDefinition::allocation, AnalysisCaseDefinition::resultExpression, AnalysisCaseUsage::resultExpression, AnnotatingElement::owningAnnotatingRelationship, AssertConstraintUsage::assertedConstraint, AssignmentActionUsage::referent, AssignmentActionUsage::targetArgument, AssignmentActionUsage::valueExpression, Association::associationEnd, Association::relatedType, Association::sourceType, Association::targetType, BooleanExpression::predicate, CalculationDefinition::calculation, ConnectionDefinition::connectionEnd, Documentation::documentedElement, EnumerationDefinition::enumeratedValue, EventOccurrenceUsage::eventOccurrence, ExhibitStateUsage::exhibitedState, Expression::isModelLevelEvaluable, FlowDefinition::flowEnd, ForLoopActionUsage::loopVariable, ForLoopActionUsage::seqArgument, Function::isModelLevelEvaluable, IfActionUsage::elseAction, IfActionUsage::ifArgument, IfActionUsage::thenAction, IncludeUseCaseUsage::useCaseIncluded, InterfaceDefinition::interfaceEnd, LoopActionUsage::bodyAction, MetadataFeature::metaclass, Package::filterCondition, PerformActionUsage::performedAction, RenderingDefinition::rendering, SatisfyRequirementUsage::satisfiedRequirement, SatisfyRequirementUsage::satisfyingFeature, TerminateActionUsage::terminatedOccurrenceArgument, TextualRepresentation::representedElement, UseCaseUsage::includedUseCase, VerificationCaseDefinition::verifiedRequirement, VerificationCaseUsage::verifiedRequirement, ViewDefinition::satisfiedViewpoint, ViewDefinition::view, ViewDefinition::viewCondition, ViewDefinition::viewRendering, ViewUsage::exposedElement, ViewUsage::satisfiedViewpoint, ViewUsage::viewCondition, ViewUsage::viewRendering, ViewpointDefinition::viewpointStakeholder, ViewpointUsage::viewpointStakeholder, WhileLoopActionUsage::untilArgument, WhileLoopActionUsage::whileArgument
- Lenguaje dominante
- Go
- Estrellas
- 56
- Forks
- 17
- Merge medio
- 12 h 45 min
- PR fusionados (30 d)
- 910
Preparar el entorno
- Incluye un Dockerfile o un archivo de Docker Compose
- Tiene una plantilla de pull request
- Leer la guía de contribución
Primeros pasos
- Lee el issue completo y luego la guía de contribución del proyecto.
- Comenta en el issue que vas a ocuparte — evita que dos personas hagan lo mismo.
- Haz un fork del repositorio y trabaja en una rama.
- Abre un pull request que haga referencia al número del issue.
Más de Open-MBEE/OpenSysML
-
dependencies
Dificultad 2/5 1-3 horas Aptitud para principiantes 65/100
Open-MBEE/OpenSysML#1951 · 3 comentarios · 1 reacción ·
Los mantenedores suelen responder en 1 día
-
Dificultad 4/5 3-5 días Aptitud para principiantes 48/100
Los mantenedores suelen responder en 1 día
-
bug
Dificultad 4/5 3-5 días Aptitud para principiantes 48/100
Open-MBEE/OpenSysML#2029 · 2 comentarios ·
Los mantenedores suelen responder en 1 día
-
Java client: no member enumeration, no GetElementProperties, no instance pathsPosiblemente ocupada @devin-ai-integration la tomó hoy. Abiertoenhancement
Dificultad 4/5 3-5 días Aptitud para principiantes 42/100
Open-MBEE/OpenSysML#2022 · 2 comentarios ·
Los mantenedores suelen responder en 1 día
-
Symbol API fields answer differently from GetElementProperties, with no sign that the answer is incompletePosiblemente ocupada @devin-ai-integration la tomó hoy. Abiertodocumentation enhancement
Dificultad 4/5 3-5 días Aptitud para principiantes 48/100
Open-MBEE/OpenSysML#2021 · 2 comentarios · 1 reacción ·
Los mantenedores suelen responder en 1 día
Todos los issues de Open-MBEE/OpenSysML
Issues similares
-
area: metrics area: tracing contribfest enhancement good first issue
Dificultad 2/5 1-3 horas Aptitud para principiantes 83/100
open-telemetry/opentelemetry-ebpf-instrumentation#4188 ·
Los mantenedores suelen responder en 1 día
-
Dificultad 2/5 1-3 horas Aptitud para principiantes 74/100
blinklabs-io/dingo#6074 ·
Los mantenedores suelen responder en 1 día
-
kind/bug remote
Dificultad 2/5 1-3 horas Aptitud para principiantes 68/100
podman-container-tools/podman#30143 ·
Los mantenedores suelen responder en 1 día
-
component/mesheryctl kind/bug language/go
Dificultad 1/5 Menos de una hora Aptitud para principiantes 88/100
meshery/meshery#22993 · 2 comentarios ·
Los mantenedores suelen responder en 1 día
-
enhancement
Dificultad 2/5 1-3 horas Aptitud para principiantes 64/100