repo_name
stringlengths
1
62
dataset
stringclasses
1 value
lang
stringclasses
11 values
pr_id
int64
1
20.1k
owner
stringlengths
2
34
reviewer
stringlengths
2
39
diff_hunk
stringlengths
15
262k
code_review_comment
stringlengths
1
99.6k
timefold-solver
github_2023
java
564
TimefoldAI
triceo
@@ -0,0 +1,69 @@ +package ai.timefold.solver.quarkus.rest; + +import static org.junit.jupiter.api.Assertions.assertEquals; + +import ai.timefold.solver.quarkus.testdata.normal.constraints.TestdataQuarkusConstraintProvider; +import ai.timefold.solver.quarkus.testdata.normal.domain.TestdataQuarkusEntity; +import ai.timef...
No need to be so verbose. Enough to say that removing public will cause an exception. If someone decides to try even though they've been warned, they'll find out what exception it is. :-)
timefold-solver
github_2023
java
564
TimefoldAI
triceo
@@ -1,21 +1,32 @@ package ai.timefold.solver.quarkus.rest; +import static org.junit.jupiter.api.Assertions.assertEquals; + import ai.timefold.solver.quarkus.testdata.normal.constraints.TestdataQuarkusConstraintProvider; import ai.timefold.solver.quarkus.testdata.normal.domain.TestdataQuarkusEntity; import ai.time...
Dtto. I'm even considering if we need to warn about this at all; we typically don't.
timefold-solver
github_2023
java
564
TimefoldAI
triceo
@@ -0,0 +1,111 @@ +package ai.timefold.solver.quarkus.it.devui; + +import static org.assertj.core.api.Assertions.assertThat; + +import java.util.List; + +import jakarta.inject.Inject; +import jakarta.inject.Named; +import jakarta.ws.rs.Path; + +import ai.timefold.solver.core.api.solver.SolverManager; +import ai.timefol...
For sake of simplicity, I'd refactor this to avoid the lambda. Make the assertions be inside a method, have the string as an argument to the method, call the method twice. The stack trace in case of a failure will be nicer, and more self-explanatory.
timefold-solver
github_2023
java
564
TimefoldAI
triceo
@@ -0,0 +1,111 @@ +package ai.timefold.solver.quarkus.it.devui; + +import static org.assertj.core.api.Assertions.assertThat; + +import java.util.List; + +import jakarta.inject.Inject; +import jakarta.inject.Named; +import jakarta.ws.rs.Path; + +import ai.timefold.solver.core.api.solver.SolverManager; +import ai.timefol...
Dtto.
timefold-solver
github_2023
java
564
TimefoldAI
triceo
@@ -27,11 +28,11 @@ public class TimefoldProcessorHotReloadTest { @Test void solverConfigHotReload() {
This test (and its `MultipleSolvers` counterpart) seems to throw a lot of exceptions into the sysout, even though it is passing. Let's fix that.
timefold-solver
github_2023
java
564
TimefoldAI
triceo
@@ -229,18 +230,27 @@ SolverConfigBuildItem recordAndRegisterBuildTimeBeans(CombinedIndexBuildItem com ClassLoader classLoader = Thread.currentThread().getContextClassLoader(); Map<String, SolverConfig> allSolverConfig = new HashMap<>();
According to the convention, this would be a `solverConfigMap`.
timefold-solver
github_2023
java
564
TimefoldAI
triceo
@@ -268,9 +278,79 @@ SolverConfigBuildItem recordAndRegisterBuildTimeBeans(CombinedIndexBuildItem com return new SolverConfigBuildItem(allSolverConfig, generatedGizmoClasses); } - private SolverConfig generateSolverConfig(ClassLoader classLoader, IndexView indexView, - BuildProducer<Reflec...
`solverConfigWithoutSolutionClassList`
timefold-solver
github_2023
java
564
TimefoldAI
triceo
@@ -268,9 +278,79 @@ SolverConfigBuildItem recordAndRegisterBuildTimeBeans(CombinedIndexBuildItem com return new SolverConfigBuildItem(allSolverConfig, generatedGizmoClasses); } - private SolverConfig generateSolverConfig(ClassLoader classLoader, IndexView indexView, - BuildProducer<Reflec...
`solverConfigWithUnusedSolutionClassList`
timefold-solver
github_2023
java
564
TimefoldAI
triceo
@@ -268,9 +278,79 @@ SolverConfigBuildItem recordAndRegisterBuildTimeBeans(CombinedIndexBuildItem com return new SolverConfigBuildItem(allSolverConfig, generatedGizmoClasses); } - private SolverConfig generateSolverConfig(ClassLoader classLoader, IndexView indexView, - BuildProducer<Reflec...
What is a target? Nothing in the solver is called a target. The exception should not mention a "target" because the user does not know what that is.
timefold-solver
github_2023
java
564
TimefoldAI
triceo
@@ -268,9 +278,79 @@ SolverConfigBuildItem recordAndRegisterBuildTimeBeans(CombinedIndexBuildItem com return new SolverConfigBuildItem(allSolverConfig, generatedGizmoClasses); } - private SolverConfig generateSolverConfig(ClassLoader classLoader, IndexView indexView, - BuildProducer<Reflec...
I'd rephrase the first sentence. "Some solver configs (...) don't specify a @PlanningSolution class, yet there are multiple available (...) on the classpath."
timefold-solver
github_2023
java
564
TimefoldAI
triceo
@@ -268,9 +278,79 @@ SolverConfigBuildItem recordAndRegisterBuildTimeBeans(CombinedIndexBuildItem com return new SolverConfigBuildItem(allSolverConfig, generatedGizmoClasses); } - private SolverConfig generateSolverConfig(ClassLoader classLoader, IndexView indexView, - BuildProducer<Reflec...
Targets again.
timefold-solver
github_2023
java
564
TimefoldAI
triceo
@@ -121,10 +127,17 @@ private static void registerSpi(Class<?> serviceClass, BuildProducer<ServiceProv } @BuildStep - HotDeploymentWatchedFileBuildItem watchSolverConfigXml() { - String solverConfigXML = timefoldBuildTimeConfig.solverConfigXml + void watchSolverConfigXml(BuildProducer<HotDeploy...
Please be consistent; `solverCongigXmlFileSet`.
timefold-solver
github_2023
java
564
TimefoldAI
triceo
@@ -225,29 +218,330 @@ SolverConfigBuildItem recordAndRegisterBeans(TimefoldRecorder recorder, Recorder + "application.properties entries (quarkus.index-dependency.<name>.group-id" + " and quarkus.index-dependency.<name>.artifact-id)."); additionalBeans.produce(new...
`solverConfigMap`
timefold-solver
github_2023
java
564
TimefoldAI
triceo
@@ -225,29 +218,330 @@ SolverConfigBuildItem recordAndRegisterBeans(TimefoldRecorder recorder, Recorder + "application.properties entries (quarkus.index-dependency.<name>.group-id" + " and quarkus.index-dependency.<name>.artifact-id)."); additionalBeans.produce(new...
Since we started using the raw strings with `%s` patterns, please apply it here as well - the new code needs to be consistent. Over time, the old pattern will be replaced entirely, as we touch more and more of the old code.
timefold-solver
github_2023
java
564
TimefoldAI
triceo
@@ -225,29 +218,330 @@ SolverConfigBuildItem recordAndRegisterBeans(TimefoldRecorder recorder, Recorder + "application.properties entries (quarkus.index-dependency.<name>.group-id" + " and quarkus.index-dependency.<name>.artifact-id)."); additionalBeans.produce(new...
And here as well.
timefold-solver
github_2023
java
564
TimefoldAI
triceo
@@ -225,29 +218,330 @@ SolverConfigBuildItem recordAndRegisterBeans(TimefoldRecorder recorder, Recorder + "application.properties entries (quarkus.index-dependency.<name>.group-id" + " and quarkus.index-dependency.<name>.artifact-id)."); additionalBeans.produce(new...
```suggestion or remove the unnecessary solution classes from the classpath.""" ``` Be mindful that, if the `"""` is on a new line, the exception message will include an empty line. You don't want that.
timefold-solver
github_2023
java
564
TimefoldAI
triceo
@@ -225,29 +218,330 @@ SolverConfigBuildItem recordAndRegisterBeans(TimefoldRecorder recorder, Recorder + "application.properties entries (quarkus.index-dependency.<name>.group-id" + " and quarkus.index-dependency.<name>.artifact-id)."); additionalBeans.produce(new...
`unusedSolutionClassNameSet`.
timefold-solver
github_2023
java
564
TimefoldAI
triceo
@@ -225,29 +218,330 @@ SolverConfigBuildItem recordAndRegisterBeans(TimefoldRecorder recorder, Recorder + "application.properties entries (quarkus.index-dependency.<name>.group-id" + " and quarkus.index-dependency.<name>.artifact-id)."); additionalBeans.produce(new...
Another newline at the end.
timefold-solver
github_2023
java
564
TimefoldAI
triceo
@@ -563,40 +845,46 @@ private GeneratedGizmoClasses generateDomainAccessors(SolverConfig solverConfig, membersToGeneratedAccessorsFor.removeIf(this::shouldIgnoreMember); // Fail fast on auto-discovery. - var planningSolutionAnnotationInstanceCollection = + Collection<An...
`unusedSolutionClassNameSet`.
timefold-solver
github_2023
java
590
TimefoldAI
triceo
@@ -0,0 +1,115 @@ +package ai.timefold.solver.spring.boot.autoconfigure; + +import static java.util.Collections.emptyList; + +import java.util.ArrayList; +import java.util.Comparator; +import java.util.HashSet; +import java.util.List; +import java.util.Optional; +import java.util.Set; +import java.util.stream.Collector...
Why was introducing this necessary? What abstract classes do we need to find?
timefold-solver
github_2023
java
590
TimefoldAI
triceo
@@ -0,0 +1,115 @@ +package ai.timefold.solver.spring.boot.autoconfigure; + +import static java.util.Collections.emptyList; + +import java.util.ArrayList; +import java.util.Comparator; +import java.util.HashSet; +import java.util.List; +import java.util.Optional; +import java.util.Set; +import java.util.stream.Collector...
Let's avoid `Optional` if the only purpose of it is to avoid a local `null` check. I'd argue the following code reads much better: var implementingClassList = findImplementingClassList(targetClass); if (implementingClassList == null || implementingClassList.isEmpty()) { return null; } else ...
timefold-solver
github_2023
java
590
TimefoldAI
triceo
@@ -0,0 +1,115 @@ +package ai.timefold.solver.spring.boot.autoconfigure; + +import static java.util.Collections.emptyList; + +import java.util.ArrayList; +import java.util.Comparator; +import java.util.HashSet; +import java.util.List; +import java.util.Optional; +import java.util.Set; +import java.util.stream.Collector...
`packageSet`.
timefold-solver
github_2023
java
590
TimefoldAI
triceo
@@ -0,0 +1,115 @@ +package ai.timefold.solver.spring.boot.autoconfigure; + +import static java.util.Collections.emptyList; + +import java.util.ArrayList; +import java.util.Comparator; +import java.util.HashSet; +import java.util.List; +import java.util.Optional; +import java.util.Set; +import java.util.stream.Collector...
Please apply the new convention to formatting error messages.
timefold-solver
github_2023
java
590
TimefoldAI
triceo
@@ -0,0 +1,115 @@ +package ai.timefold.solver.spring.boot.autoconfigure; + +import static java.util.Collections.emptyList; + +import java.util.ArrayList; +import java.util.Comparator; +import java.util.HashSet; +import java.util.List; +import java.util.Optional; +import java.util.Set; +import java.util.stream.Collector...
Dtto.
timefold-solver
github_2023
java
590
TimefoldAI
triceo
@@ -0,0 +1,115 @@ +package ai.timefold.solver.spring.boot.autoconfigure; + +import static java.util.Collections.emptyList; + +import java.util.ArrayList; +import java.util.Comparator; +import java.util.HashSet; +import java.util.List; +import java.util.Optional; +import java.util.Set; +import java.util.stream.Collector...
Dtto.
timefold-solver
github_2023
java
590
TimefoldAI
triceo
@@ -0,0 +1,115 @@ +package ai.timefold.solver.spring.boot.autoconfigure; + +import static java.util.Collections.emptyList; + +import java.util.ArrayList; +import java.util.Comparator; +import java.util.HashSet; +import java.util.List; +import java.util.Optional; +import java.util.Set; +import java.util.stream.Collector...
Dtto.
timefold-solver
github_2023
java
590
TimefoldAI
triceo
@@ -64,62 +70,422 @@ @ConditionalOnMissingBean({ SolverConfig.class, SolverFactory.class, ScoreManager.class, SolutionManager.class, SolverManager.class }) @EnableConfigurationProperties({ TimefoldProperties.class }) -public class TimefoldAutoConfiguration implements BeanClassLoaderAware { +public class Time...
```suggestion // postProcessBeanFactory runs before creating any bean, but we need TimefoldProperties. // Therefore, we use the Environment to load the properties. ```
timefold-solver
github_2023
java
590
TimefoldAI
triceo
@@ -64,62 +70,422 @@ @ConditionalOnMissingBean({ SolverConfig.class, SolverFactory.class, ScoreManager.class, SolutionManager.class, SolverManager.class }) @EnableConfigurationProperties({ TimefoldProperties.class }) -public class TimefoldAutoConfiguration implements BeanClassLoaderAware { +public class Time...
I suggest we should have the same behavior in Spring Boot and Quarkus, where possible. What would be the reasons _not_ to do this here?
timefold-solver
github_2023
java
590
TimefoldAI
triceo
@@ -64,62 +70,422 @@ @ConditionalOnMissingBean({ SolverConfig.class, SolverFactory.class, ScoreManager.class, SolutionManager.class, SolverManager.class }) @EnableConfigurationProperties({ TimefoldProperties.class }) -public class TimefoldAutoConfiguration implements BeanClassLoaderAware { +public class Time...
Why the quote inside the string?
timefold-solver
github_2023
java
590
TimefoldAI
triceo
@@ -64,62 +70,422 @@ @ConditionalOnMissingBean({ SolverConfig.class, SolverFactory.class, ScoreManager.class, SolutionManager.class, SolverManager.class }) @EnableConfigurationProperties({ TimefoldProperties.class }) -public class TimefoldAutoConfiguration implements BeanClassLoaderAware { +public class Time...
Let's not leave the TODO here, let's fix it instead. What would be the pros and cons? What does Quarkus do?
timefold-solver
github_2023
java
590
TimefoldAI
triceo
@@ -1,12 +1,24 @@ package ai.timefold.solver.spring.boot.autoconfigure.config; +import java.util.Map; +import java.util.Set; + import ai.timefold.solver.core.api.domain.common.DomainAccessType; import ai.timefold.solver.core.api.score.stream.ConstraintStreamImplType; import ai.timefold.solver.core.config.solver.E...
It is a good idea to have a constant for this. However, I'd go further and have constants even for the names themselves - that way, you don't need to repeat them later in this class.
timefold-solver
github_2023
others
687
TimefoldAI
triceo
@@ -50,6 +50,12 @@ Defaults to `REFLECTION`. The other possible value is `GIZMO`. endif::[] +{property_prefix}timefold.solver.{solver_name_prefix}nearby-distance-meter-class:: +Enable the xref:enterprise-edition/enterprise-edition.adoc#nearbySelection[Nearby Selection] quick configuration. +If the Nearby Selection ...
```suggestion and automatically enables Nearby Selection for the compatible move selectors. ```
timefold-solver
github_2023
java
687
TimefoldAI
triceo
@@ -719,6 +720,27 @@ private void applySolverProperties(IndexView indexView, String solverName, Solve if (solverConfig.getDomainAccessType() == null) { solverConfig.setDomainAccessType(DomainAccessType.GIZMO); } + + Optional<String> nearbyDistanceMeterClass = + timef...
`Optional` has `ifPresent(...)` for just this use case.
timefold-solver
github_2023
java
687
TimefoldAI
triceo
@@ -60,6 +60,11 @@ public interface SolverBuildTimeConfig { */ TerminationRuntimeConfig termination(); + /** + * Enable the Nearby Selection quick configuration. + */ + Optional<String> nearbyDistanceMeterClass();
Shouldn't this return `Class<? extends NearbyDistanceMeter>` directly? The other properties also don't return `String`.
timefold-solver
github_2023
java
687
TimefoldAI
triceo
@@ -52,7 +55,7 @@ void solverProperties() { assertEquals(DomainAccessType.REFLECTION, solverConfig.getDomainAccessType()); assertEquals(null, solverConfig.getScoreDirectorFactoryConfig().getConstraintStreamImplType()); - + assertNotNull(solverConfig.getNearbyDistanceMeterClass(...
I obviously didn't notice before, but we use AssertJ, not JUnit assertions.
timefold-solver
github_2023
java
687
TimefoldAI
triceo
@@ -39,30 +40,27 @@ class TimefoldBenchmarkProcessorInheritedSolverBenchmarkTest { @Test void inheritClassesFromSolverConfig() { - Assertions.assertEquals(TestdataQuarkusSolution.class, solverConfig.getSolutionClass()); - Assertions.assertEquals(2, solverConfig.getEntityClassList().size()); - ...
`containsExactly` (and similar methods) is your friend.
timefold-solver
github_2023
java
687
TimefoldAI
triceo
@@ -39,30 +40,27 @@ class TimefoldBenchmarkProcessorInheritedSolverBenchmarkTest { @Test void inheritClassesFromSolverConfig() { - Assertions.assertEquals(TestdataQuarkusSolution.class, solverConfig.getSolutionClass()); - Assertions.assertEquals(2, solverConfig.getEntityClassList().size()); - ...
Also easier done with `containsExactly`.
timefold-solver
github_2023
java
684
TimefoldAI
triceo
@@ -280,4 +282,33 @@ private void inheritCommon(MoveSelectorConfig<?> inheritedConfig) { fixedProbabilityWeight, inheritedConfig.getFixedProbabilityWeight()); } + protected String addRandomSuffix(String name, Random random) {
Should be `static`, no?
timefold-solver
github_2023
java
684
TimefoldAI
triceo
@@ -280,4 +282,33 @@ private void inheritCommon(MoveSelectorConfig<?> inheritedConfig) { fixedProbabilityWeight, inheritedConfig.getFixedProbabilityWeight()); } + protected String addRandomSuffix(String name, Random random) { + StringBuilder value = new StringBuilder(name); + va...
I'd consider working without this. You can call `enableNearbySelection(...)`, which by default returns the config unchanged. (I'm driven by eliminating boilerplate code; the selectors already have plenty without adding more.)
timefold-solver
github_2023
java
684
TimefoldAI
triceo
@@ -280,4 +282,33 @@ private void inheritCommon(MoveSelectorConfig<?> inheritedConfig) { fixedProbabilityWeight, inheritedConfig.getFixedProbabilityWeight()); } + protected String addRandomSuffix(String name, Random random) { + StringBuilder value = new StringBuilder(name); + va...
Maybe this could return `false` by default, so that you only need to override it when `true` needs to be returned?
timefold-solver
github_2023
java
684
TimefoldAI
triceo
@@ -154,6 +154,18 @@ public void visitReferencedClasses(Consumer<Class<?>> classVisitor) { } } + @Override + public boolean hasNearbySelectionConfig() { + return (subListSelectorConfig != null && ((subListSelectorConfig.getNearbySelectionConfig() != null)
This condition is kinda insane. Maybe simplify it by introducing boolean variables for its components? The code will be longer, but people will actually be able to understand what the condition does. (There are similar conditions elsewhere in this PR.)
timefold-solver
github_2023
java
684
TimefoldAI
triceo
@@ -22,8 +23,24 @@ public UnionMoveSelectorFactory(UnionMoveSelectorConfig moveSelectorConfig) { @Override protected MoveSelector<Solution_> buildBaseMoveSelector(HeuristicConfigPolicy<Solution_> configPolicy, SelectionCacheType minimumCacheType, boolean randomSelection) { - List<MoveSelec...
Maybe explain _why_ in a comment? The problem with cartesian is that if there are two selectors, it will have A×B moves. So if there are four selectors (2 non-nearby + 2 nearby), it will be A×B×C×D moves. Whereas union will always be A+B+C+D.
timefold-solver
github_2023
java
684
TimefoldAI
triceo
@@ -22,8 +23,24 @@ public UnionMoveSelectorFactory(UnionMoveSelectorConfig moveSelectorConfig) { @Override protected MoveSelector<Solution_> buildBaseMoveSelector(HeuristicConfigPolicy<Solution_> configPolicy, SelectionCacheType minimumCacheType, boolean randomSelection) { - List<MoveSelec...
Cautious use of `var` wouldn't hurt here, I think. The name already says `list`.
timefold-solver
github_2023
java
684
TimefoldAI
triceo
@@ -0,0 +1,355 @@ +package ai.timefold.solver.core.config.heuristic.selector.move; + +import static org.junit.jupiter.api.Assertions.assertEquals; +import static org.junit.jupiter.api.Assertions.assertFalse; +import static org.junit.jupiter.api.Assertions.assertNotNull; +import static org.junit.jupiter.api.Assertions.a...
We use AssertJ, not JUnit assertions.
timefold-solver
github_2023
java
684
TimefoldAI
triceo
@@ -162,6 +166,31 @@ public void visitReferencedClasses(Consumer<Class<?>> classVisitor) { classVisitor.accept(selectorProbabilityWeightFactoryClass); } + @Override + public UnionMoveSelectorConfig enableNearbySelection(Class<? extends NearbyDistanceMeter<?, ?>> distanceMeter, + Random ...
Id argue that the following would read better, also coincidentally doesn't need to create an extra list or iterate twice: for (... all configs ...) { if (not nearby) { continue; } ... } Also, the `selectorConfig` variable is another nice use of `var` - all the necessar...
timefold-solver
github_2023
others
726
TimefoldAI
triceo
@@ -10,7 +10,7 @@ Assign each conference talk to a timeslot and a room, after the talks have been image::use-cases-and-examples/conference-scheduling/conferenceSchedulingMilestonesTimeline.png[align="center"]
I think this entire adoc file can disappear entirely. It is a description of an example which we no longer have. (In fact, I will be removing the entire "use cases and examples" chapter very soon.)
timefold-solver
github_2023
java
659
TimefoldAI
triceo
@@ -32,7 +32,8 @@ protected TriFunction<A, B, Score_, ConstraintJustification> getJustificationMap public <ConstraintJustification_ extends ConstraintJustification> BiConstraintBuilder<A, B, Score_> justifyWith( TriFunction<A, B, Score_, ConstraintJustification_> justificationMapping) { if (t...
Please convert to the new convention, using multi-line strings and placeholders.
timefold-solver
github_2023
java
659
TimefoldAI
triceo
@@ -47,7 +48,8 @@ protected BiFunction<A, B, Collection<Object>> getIndictedObjectsMapping() { @Override public BiConstraintBuilder<A, B, Score_> indictWith(BiFunction<A, B, Collection<Object>> indictedObjectsMapping) { if (this.indictedObjectsMapping != null) { - throw new IllegalStateExc...
Dtto.
timefold-solver
github_2023
java
659
TimefoldAI
triceo
@@ -32,7 +32,8 @@ protected PentaFunction<A, B, C, D, Score_, ConstraintJustification> getJustific public <ConstraintJustification_ extends ConstraintJustification> QuadConstraintBuilder<A, B, C, D, Score_> justifyWith( PentaFunction<A, B, C, D, Score_, ConstraintJustification_> justificationMapping) ...
Please convert to the new convention, using multi-line strings and placeholders.
timefold-solver
github_2023
java
659
TimefoldAI
triceo
@@ -48,7 +49,8 @@ protected QuadFunction<A, B, C, D, Collection<Object>> getIndictedObjectsMapping public QuadConstraintBuilder<A, B, C, D, Score_> indictWith(QuadFunction<A, B, C, D, Collection<Object>> indictedObjectsMapping) { if (this.indictedObjectsMapping != null) { - throw n...
Dtto.
timefold-solver
github_2023
java
659
TimefoldAI
triceo
@@ -32,7 +32,8 @@ protected QuadFunction<A, B, C, Score_, ConstraintJustification> getJustificatio public <ConstraintJustification_ extends ConstraintJustification> TriConstraintBuilder<A, B, C, Score_> justifyWith( QuadFunction<A, B, C, Score_, ConstraintJustification_> justificationMapping) { ...
Dtto.
timefold-solver
github_2023
java
659
TimefoldAI
triceo
@@ -47,7 +48,8 @@ protected TriFunction<A, B, C, Collection<Object>> getIndictedObjectsMapping() { @Override public TriConstraintBuilder<A, B, C, Score_> indictWith(TriFunction<A, B, C, Collection<Object>> indictedObjectsMapping) { if (this.indictedObjectsMapping != null) { - throw new Ill...
Dtto.
timefold-solver
github_2023
java
659
TimefoldAI
triceo
@@ -32,7 +32,8 @@ protected BiFunction<A, Score_, ConstraintJustification> getJustificationMapping public <ConstraintJustification_ extends ConstraintJustification> UniConstraintBuilder<A, Score_> justifyWith( BiFunction<A, Score_, ConstraintJustification_> justificationMapping) { if (this.ju...
Dtto.
timefold-solver
github_2023
java
659
TimefoldAI
triceo
@@ -47,7 +48,8 @@ protected Function<A, Collection<Object>> getIndictedObjectsMapping() { @Override public UniConstraintBuilder<A, Score_> indictWith(Function<A, Collection<Object>> indictedObjectsMapping) { if (this.indictedObjectsMapping != null) { - throw new IllegalStateException("Indi...
Dtto.
timefold-solver
github_2023
java
659
TimefoldAI
triceo
@@ -104,6 +125,17 @@ private void validateMatchWeighTotal(Number matchWeightTotal) { } } + private void validateJustification(ConstraintJustification... justifications) { + Objects.requireNonNull(justifications, "The justification must be not null."); + } + + private void validateIndictm...
Arguably, this should not be here. What if the user decided to indict with no indictments? We should be able to test for that.
timefold-solver
github_2023
java
659
TimefoldAI
triceo
@@ -135,6 +167,65 @@ private void assertImpact(ScoreImpactType scoreImpactType, Number matchWeightTot throw new AssertionError(assertionMessage); } + private void assertJustification(ConstraintJustification justification, String message) { + if (constraintJustificationCollection.isEmpty()) { +...
Hmm, I think I didn't think this through. A constraint can trigger multiple matches, therefore there may be multiple justifications. Therefore the user must be able to provide _all_ justifications that they expect to be there. So the code constraint verifier `justifiesWith(...)` needs to be a vararg.
timefold-solver
github_2023
java
659
TimefoldAI
triceo
@@ -32,7 +32,9 @@ protected TriFunction<A, B, Score_, ConstraintJustification> getJustificationMap public <ConstraintJustification_ extends ConstraintJustification> BiConstraintBuilder<A, B, Score_> justifyWith( TriFunction<A, B, Score_, ConstraintJustification_> justificationMapping) { if (t...
```suggestion """ Justification mapping already set (%s). Maybe the constraint calls justifyWith() twice?""" ``` Java's got cool features now. Use them, please. (Apply in other places in the PR as well.)
timefold-solver
github_2023
java
659
TimefoldAI
triceo
@@ -0,0 +1,25 @@ +package ai.timefold.solver.test.api.score.stream.testdata.justification; + +import ai.timefold.solver.core.api.score.stream.ConstraintJustification; + +public class TestSecondComparableJustification
Any particular reason for testing comparable justifications? What specifically is being tested here?
timefold-solver
github_2023
java
659
TimefoldAI
triceo
@@ -4,9 +4,55 @@ import ai.timefold.solver.core.api.score.buildin.hardsoftlong.HardSoftLongScore; import ai.timefold.solver.core.api.score.stream.Constraint; +import ai.timefold.solver.core.api.score.stream.ConstraintJustification; public interface SingleConstraintAssertion { + /** + * As defined by {@l...
I'm wondering, for the sake of consistency... the `penalizes(...)` method (and its reward counterparts) has the `message` as second argument. Maybe we deprecate those and introduce new ones where `message` goes first? It would look silly if some methods have message first, and other have message last. This is going ...
timefold-solver
github_2023
java
659
TimefoldAI
triceo
@@ -26,6 +29,9 @@ public Constraint[] defineConstraints(ConstraintFactory factory) { protected Constraint horizontalConflict(ConstraintFactory factory) { return factory.forEachUniquePair(Queen.class, equal(Queen::getRowIndex)) .penalize(SimpleScore.ONE) + .justifyWith((quee...
Isn't this the default if you don't provide any indictments?
timefold-solver
github_2023
java
659
TimefoldAI
triceo
@@ -39,36 +98,64 @@ default void penalizesBy(int matchWeightTotal) { * @throws AssertionError when the expected penalty is not observed */ default void penalizesBy(long matchWeightTotal) { - penalizesBy(matchWeightTotal, null); + penalizesBy(null, matchWeightTotal); } /** - ...
The next version of Solver is `1.8.0`. Please apply consistently.
timefold-solver
github_2023
others
659
TimefoldAI
triceo
@@ -0,0 +1,17 @@ +type: specs.openrewrite.org/v1beta/recipe +name: ai.timefold.solver.migration.ToLatestSingleConstraintAssertion +displayName: 'Use non-deprecated SingleConstraintAssertion methods' +description: 'Use `penalizesBy/rewardsWith(String, int)` instead of `penalizesBy/rewardsWith(int, String)` on `SingleCon...
I'd prefer we have a Java class for this. See how I created the `NullableRecipe`. The main reason is that if we did every such thing as a top-level YML recipe, we would soon have too many of them. I'd like to keep that folder relatively clean.
timefold-solver
github_2023
java
659
TimefoldAI
triceo
@@ -39,36 +98,64 @@ default void penalizesBy(int matchWeightTotal) { * @throws AssertionError when the expected penalty is not observed */ default void penalizesBy(long matchWeightTotal) { - penalizesBy(matchWeightTotal, null); + penalizesBy(null, matchWeightTotal); } /** - ...
If it starts with a capital letter, it is a sentence and therefore it ends with a `.`.
timefold-solver
github_2023
java
515
TimefoldAI
zepfred
@@ -213,6 +217,59 @@ void solveGenerics() throws ExecutionException, InterruptedException { solverJob.getFinalBestSolution(); } + @Test + void solveWithOverride() throws InterruptedException { + CountDownLatch countDownLatch = new CountDownLatch(2); + PhaseConfig<?> customPhaseConfig...
Yes, it makes sense. The goal is to ensure the termination time is overridden, but the internal state is not fetchable. I'll try to think of something else.
timefold-solver
github_2023
java
515
TimefoldAI
zepfred
@@ -0,0 +1,41 @@ +package ai.timefold.solver.core.api.solver; + +import java.util.UUID; +import java.util.function.BiConsumer; +import java.util.function.Consumer; + +import ai.timefold.solver.core.api.domain.solution.PlanningSolution; +import ai.timefold.solver.core.config.solver.termination.TerminationConfig; + +/** ...
Let's wait for the design meeting to define the next steps here.
timefold-solver
github_2023
java
515
TimefoldAI
zepfred
@@ -0,0 +1,41 @@ +package ai.timefold.solver.core.api.solver; + +import java.util.UUID; +import java.util.function.BiConsumer; +import java.util.function.Consumer; + +import ai.timefold.solver.core.api.domain.solution.PlanningSolution; +import ai.timefold.solver.core.config.solver.termination.TerminationConfig; + +/** ...
Let's wait for the design meeting to define the next steps here.
timefold-solver
github_2023
java
515
TimefoldAI
zepfred
@@ -0,0 +1,37 @@ +package ai.timefold.solver.core.api.solver; + +import java.util.UUID; +import java.util.function.BiConsumer; +import java.util.function.Consumer; + +import ai.timefold.solver.core.api.domain.solution.PlanningSolution; +import ai.timefold.solver.core.config.solver.termination.TerminationConfig; + +/** ...
Let's wait for the design meeting to define the next steps here.
timefold-solver
github_2023
java
515
TimefoldAI
zepfred
@@ -268,11 +364,46 @@ default SolverJob<Solution_, ProblemId_> solveAndListen(ProblemId_ problemId, * If null it defaults to logging the exception as an error. * @return never null */ + default SolverJob<Solution_, ProblemId_> solveAndListen(ProblemId_ problemId, + Function<? supe...
Let's wait for the design meeting to define the next steps here.
timefold-solver
github_2023
java
515
TimefoldAI
zepfred
@@ -98,6 +99,65 @@ static <Solution_, ProblemId_> SolverManager<Solution_, ProblemId_> create( return new DefaultSolverManager<>(solverFactory, solverManagerConfig); } + // ************************************************************************ + // Builder methods + // ***********************...
Let's wait for the design meeting to define the next steps here.
timefold-solver
github_2023
java
515
TimefoldAI
zepfred
@@ -196,10 +264,35 @@ default SolverJob<Solution_, ProblemId_> solve(ProblemId_ problemId, * If null it defaults to logging the exception as an error. * @return never null */ + default SolverJob<Solution_, ProblemId_> solve(ProblemId_ problemId, + Fu...
Let's wait for the design meeting to define the next steps here.
timefold-solver
github_2023
java
515
TimefoldAI
zepfred
@@ -0,0 +1,38 @@ +package ai.timefold.solver.core.config.solver; + +import ai.timefold.solver.core.api.domain.solution.PlanningSolution; +import ai.timefold.solver.core.config.solver.termination.TerminationConfig; + +/** + * Includes settings to override default {@link ai.timefold.solver.core.api.solver.Solver} configu...
> It is questionable if this class belongs in `config` package. As you can see, it caused it to be needlessly inserted in the XSD. I'd move this to `api/solver`, which is where it is primarily used. > Yes, I was unsure about how to deal with this XSD auto-generation, and it makes sense to move it to `api/solver.` ...
timefold-solver
github_2023
java
515
TimefoldAI
zepfred
@@ -0,0 +1,101 @@ +package ai.timefold.solver.core.config.solver; + +import java.util.UUID; +import java.util.function.BiConsumer; +import java.util.function.Consumer; +import java.util.function.Function; + +import ai.timefold.solver.core.api.domain.solution.PlanningSolution; +import ai.timefold.solver.core.config.solv...
I removed this class as it had become unused after updating the public API contract.
timefold-solver
github_2023
java
515
TimefoldAI
zepfred
@@ -0,0 +1,82 @@ +package ai.timefold.solver.core.impl.solver; + +import java.util.Objects; +import java.util.UUID; +import java.util.function.BiConsumer; +import java.util.function.Consumer; +import java.util.function.Function; + +import ai.timefold.solver.core.api.domain.solution.PlanningSolution; +import ai.timefold...
> If you're implementing `SolverJobBuilder`, typically we call the implementation `DefaultSolverJobBuilder`. Done! > In this case, it is questionable whether we need the interface at all - why not make this the `SolverJobBuilder` directly? Make the constructors package-private and there are no issues. The inte...
timefold-solver
github_2023
others
515
TimefoldAI
zepfred
@@ -1435,6 +1435,18 @@ </xs:complexType> + <xs:complexType name="solverConfigOverride">
Yes, I confirmed that no changes have been made to this file.
timefold-solver
github_2023
java
515
TimefoldAI
zepfred
@@ -115,6 +116,15 @@ static <Solution_> SolverFactory<Solution_> create(SolverConfig solverConfig) { * * @return never null */ - Solver<Solution_> buildSolver(); + default Solver<Solution_> buildSolver() { + return this.buildSolver(null); + } + /** + * As defined by {@link #bu...
I have made changes to the logic to ensure that the configOverride is now required.
timefold-solver
github_2023
java
515
TimefoldAI
zepfred
@@ -0,0 +1,95 @@ +package ai.timefold.solver.core.api.solver; + +import java.util.UUID; +import java.util.function.BiConsumer; +import java.util.function.Consumer; +import java.util.function.Function; + +import ai.timefold.solver.core.api.domain.solution.PlanningSolution; + +/** + * Provides a fluid contract that allow...
The only case I can think of is if we want to reuse a builder and update the configuration, which seems an anti-pattern.
timefold-solver
github_2023
java
515
TimefoldAI
zepfred
@@ -72,7 +72,10 @@ void solve() throws ExecutionException, InterruptedException { Collections.emptyList(), HardSoftLongScore.ZERO); - SolverJob<TestDataKitchenSinkSolution, Long> solverJob = solverManager.solve(1L, problem); + SolverJob<TestDataKitchenSinkSolution, Long...
Agreed. Should we keep using the builder or revert changes for this method?
timefold-solver
github_2023
others
623
TimefoldAI
triceo
@@ -0,0 +1,286 @@ += Define the constraints and calculate the score +:imagesdir: ../.. + +A _score_ represents the quality of a specific solution. +The higher the better. +Timefold Solver looks for the best solution, which is the solution with the highest score found in the available time. +It might be the _optimal_ so...
```suggestion For example: _The sum total of travel time._ ```
timefold-solver
github_2023
others
623
TimefoldAI
triceo
@@ -0,0 +1,286 @@ += Define the constraints and calculate the score +:imagesdir: ../.. + +A _score_ represents the quality of a specific solution. +The higher the better. +Timefold Solver looks for the best solution, which is the solution with the highest score found in the available time. +It might be the _optimal_ so...
```suggestion // Max end-time not met ```
timefold-solver
github_2023
others
623
TimefoldAI
triceo
@@ -0,0 +1,286 @@ += Define the constraints and calculate the score +:imagesdir: ../.. + +A _score_ represents the quality of a specific solution. +The higher the better. +Timefold Solver looks for the best solution, which is the solution with the highest score found in the available time. +It might be the _optimal_ so...
```suggestion // Max end-time not met ```
timefold-solver
github_2023
others
623
TimefoldAI
triceo
@@ -0,0 +1,286 @@ += Define the constraints and calculate the score +:imagesdir: ../.. + +A _score_ represents the quality of a specific solution. +The higher the better. +Timefold Solver looks for the best solution, which is the solution with the highest score found in the available time. +It might be the _optimal_ so...
Let's make this an `int`.
timefold-solver
github_2023
others
623
TimefoldAI
triceo
@@ -0,0 +1,286 @@ += Define the constraints and calculate the score +:imagesdir: ../.. + +A _score_ represents the quality of a specific solution. +The higher the better. +Timefold Solver looks for the best solution, which is the solution with the highest score found in the available time. +It might be the _optimal_ so...
To avoid this.
timefold-solver
github_2023
others
623
TimefoldAI
triceo
@@ -0,0 +1,286 @@ += Define the constraints and calculate the score +:imagesdir: ../.. + +A _score_ represents the quality of a specific solution. +The higher the better. +Timefold Solver looks for the best solution, which is the solution with the highest score found in the available time. +It might be the _optimal_ so...
`int` please.
timefold-solver
github_2023
others
623
TimefoldAI
triceo
@@ -0,0 +1,286 @@ += Define the constraints and calculate the score +:imagesdir: ../.. + +A _score_ represents the quality of a specific solution. +The higher the better. +Timefold Solver looks for the best solution, which is the solution with the highest score found in the available time. +It might be the _optimal_ so...
```suggestion The `ConstraintProvider` scales much better than the `EasyScoreCalculator`: typically __O__(n) instead of __O__(n²). ```
timefold-solver
github_2023
others
623
TimefoldAI
triceo
@@ -0,0 +1,790 @@ += Model the domain objects +:imagesdir: ../.. + +Your goal is to assign each visit to a vehicle. +You will create these classes: + +image::quickstart/vehicle-routing/vehicleRoutingClassDiagramPure.png[] + +== Location + +The `Location` class is used to represent the destination for deliveries or the ...
Maybe reference planning entity doc?
timefold-solver
github_2023
others
623
TimefoldAI
triceo
@@ -0,0 +1,790 @@ += Model the domain objects +:imagesdir: ../.. + +Your goal is to assign each visit to a vehicle. +You will create these classes: + +image::quickstart/vehicle-routing/vehicleRoutingClassDiagramPure.png[] + +== Location + +The `Location` class is used to represent the destination for deliveries or the ...
```suggestion Timefold Solver recognizes it as a sequence of connected variables, the field must have an `@PlanningListVariable` annotation. ```
timefold-solver
github_2023
others
623
TimefoldAI
triceo
@@ -0,0 +1,790 @@ += Model the domain objects +:imagesdir: ../.. + +Your goal is to assign each visit to a vehicle. +You will create these classes: + +image::quickstart/vehicle-routing/vehicleRoutingClassDiagramPure.png[] + +== Location + +The `Location` class is used to represent the destination for deliveries or the ...
Aren't we repeating ourselves here?
timefold-solver
github_2023
others
623
TimefoldAI
triceo
@@ -0,0 +1,790 @@ += Model the domain objects +:imagesdir: ../.. + +Your goal is to assign each visit to a vehicle. +You will create these classes: + +image::quickstart/vehicle-routing/vehicleRoutingClassDiagramPure.png[] + +== Location + +The `Location` class is used to represent the destination for deliveries or the ...
This paragraph contains more information than the first one regarding the list variable, but I still wonder why does it duplicate some of the information.
timefold-solver
github_2023
others
623
TimefoldAI
triceo
@@ -0,0 +1,790 @@ += Model the domain objects +:imagesdir: ../.. + +Your goal is to assign each visit to a vehicle. +You will create these classes: + +image::quickstart/vehicle-routing/vehicleRoutingClassDiagramPure.png[] + +== Location + +The `Location` class is used to represent the destination for deliveries or the ...
Link to section on shadow variables?
timefold-solver
github_2023
others
623
TimefoldAI
triceo
@@ -0,0 +1,790 @@ += Model the domain objects +:imagesdir: ../.. + +Your goal is to assign each visit to a vehicle. +You will create these classes: + +image::quickstart/vehicle-routing/vehicleRoutingClassDiagramPure.png[] + +== Location + +The `Location` class is used to represent the destination for deliveries or the ...
```suggestion `@NextElementShadowVariable`, respectively. The method returns a reference of the previous and next visit of ```
timefold-solver
github_2023
others
623
TimefoldAI
triceo
@@ -0,0 +1,518 @@ += Gather the domain objects in a planning solution +:imagesdir: ../.. + +A `VehicleRoutePlan` wraps all `Vehicle` and `Visit` instances of a single dataset. +Furthermore, because it contains all vehicles and visits, each with a specific planning variable state, +it is a _planning solution_ and it has...
Link to planning solution doc.
timefold-solver
github_2023
others
623
TimefoldAI
triceo
@@ -0,0 +1,518 @@ += Gather the domain objects in a planning solution +:imagesdir: ../.. + +A `VehicleRoutePlan` wraps all `Vehicle` and `Visit` instances of a single dataset. +Furthermore, because it contains all vehicles and visits, each with a specific planning variable state, +it is a _planning solution_ and it has...
It is not `null`. The solver fails when list variable is null. It is an empty list.
timefold-solver
github_2023
others
623
TimefoldAI
triceo
@@ -0,0 +1,518 @@ += Gather the domain objects in a planning solution +:imagesdir: ../.. + +A `VehicleRoutePlan` wraps all `Vehicle` and `Visit` instances of a single dataset. +Furthermore, because it contains all vehicles and visits, each with a specific planning variable state, +it is a _planning solution_ and it has...
```suggestion ** For each `Vehicle`: ```
timefold-solver
github_2023
others
623
TimefoldAI
triceo
@@ -0,0 +1,518 @@ += Gather the domain objects in a planning solution +:imagesdir: ../.. + +A `VehicleRoutePlan` wraps all `Vehicle` and `Visit` instances of a single dataset. +Furthermore, because it contains all vehicles and visits, each with a specific planning variable state, +it is a _planning solution_ and it has...
Shadow vars are not optional - if you specify them, they will be filled. (They are optional in a sense that you don't need to specify them.) However, with a fresh solution, they are null because the listeners did not run yet.
timefold-solver
github_2023
others
623
TimefoldAI
triceo
@@ -0,0 +1,518 @@ += Gather the domain objects in a planning solution +:imagesdir: ../.. + +A `VehicleRoutePlan` wraps all `Vehicle` and `Visit` instances of a single dataset. +Furthermore, because it contains all vehicles and visits, each with a specific planning variable state, +it is a _planning solution_ and it has...
```suggestion Then create an implementation using Haversine method: ```
timefold-solver
github_2023
others
623
TimefoldAI
triceo
@@ -0,0 +1,518 @@ += Gather the domain objects in a planning solution +:imagesdir: ../.. + +A `VehicleRoutePlan` wraps all `Vehicle` and `Visit` instances of a single dataset. +Furthermore, because it contains all vehicles and visits, each with a specific planning variable state, +it is a _planning solution_ and it has...
```suggestion contract for driving time calculation: ```
timefold-solver
github_2023
others
623
TimefoldAI
triceo
@@ -0,0 +1,918 @@ +[#quarkusQuickStart] += Vehicle Routing Quick Start Guide +:doctype: book +:imagesdir: ../.. +:sectnums: +:icons: font +include::../../_attributes.adoc[] + +// Keep this in sync with the quarkus repo's copy +// https://github.com/quarkusio/quarkus/blob/main/docs/src/main/asciidoc/timefold.adoc +// Ke...
Maybe suggest `SDKMAN`?
timefold-solver
github_2023
others
623
TimefoldAI
triceo
@@ -0,0 +1,918 @@ +[#quarkusQuickStart] += Vehicle Routing Quick Start Guide +:doctype: book +:imagesdir: ../.. +:sectnums: +:icons: font +include::../../_attributes.adoc[] + +// Keep this in sync with the quarkus repo's copy +// https://github.com/quarkusio/quarkus/blob/main/docs/src/main/asciidoc/timefold.adoc +// Ke...
```suggestion when given two visits assigned to the same vehicle, penalizes with a match weight of `20` (exceeded capacity). ```
timefold-solver
github_2023
others
734
TimefoldAI
zepfred
@@ -346,17 +346,19 @@ It is not available in the Community Edition. There are several ways of doing multi-threaded solving: -* *Multitenancy*: solve different datasets in parallel -** The `SolverManager` will make it even easier to set this up, in a future version. +* *<<multithreadedIncrementalSolving,Multi-threa...
This sentence is repeated
timefold-solver
github_2023
others
734
TimefoldAI
zepfred
@@ -366,89 +368,109 @@ A xref:using-timefold-solver/running-the-solver.adoc#logging[logging level] of ` and slow down the xref:constraints-and-score/performance.adoc#scoreCalculationSpeed[score calculation speed]. ==== -[#planningId] -==== `@PlanningId` - -For some functionality (such as multi-threaded solving and ...
```suggestion With this feature, the solver can run significantly faster, getting you the right solution earlier. ```