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 | 415 | TimefoldAI | rsynek | @@ -98,6 +98,10 @@ public double getAngle(Location location) {
return Math.atan2(latitudeDifference, longitudeDifference);
}
+ protected static long adjust(double distance) {
+ return (long) (distance + 0.5); // +0.5 to avoid floating point rounding errors | Is it any different from `Math.round()` ? |
timefold-solver | github_2023 | java | 411 | TimefoldAI | rsynek | @@ -179,7 +179,8 @@ private ConstructionHeuristicDecider<Solution_> buildDecider(HeuristicConfigPoli
if (moveThreadCount == null) {
decider = new ConstructionHeuristicDecider<>(configPolicy.getLogIndentation(), termination, forager);
} else {
- decider = MultithreadedSolvingEnt... | (soft suggestion) These pairs of a feature and a workaround could be an `enum`. |
timefold-solver | github_2023 | others | 404 | TimefoldAI | rsynek | @@ -0,0 +1,707 @@
+= Enterprise Edition
+:doctype: book
+:sectnums:
+:icons: font
+
+Timefold Solver Enterprise Edition is a commercial product that offers additional features,
+such as <<nearbySelection,nearby selection>> and <<multithreadedSolving,multi-threaded solving>>.
+These features are essential to scale out t... | Should we explicitly say that Enterprise Edition can still be used for an evaluation/development purposes? |
timefold-solver | github_2023 | java | 401 | TimefoldAI | triceo | @@ -7,17 +7,138 @@
import ai.timefold.solver.core.api.score.Score;
import ai.timefold.solver.core.api.score.stream.ConstraintStream;
import ai.timefold.solver.core.config.solver.EnvironmentMode;
+import ai.timefold.solver.core.impl.util.Pair;
+import ai.timefold.solver.core.impl.util.Quadruple;
+import ai.timefold.s... | This confused me at first.
- `CollectorA_` et al. are not `...ConstraintCollector` subtypes. (Because that would cause further explosion in the number of interfaces you'd have to create for this.)
- `NoKeysThreeCollectorsGroupByBuilder` also doesn't reference constraint collectors.
- But then the builder itself is... |
timefold-solver | github_2023 | java | 401 | TimefoldAI | triceo | @@ -7,17 +7,138 @@
import ai.timefold.solver.core.api.score.Score;
import ai.timefold.solver.core.api.score.stream.ConstraintStream;
import ai.timefold.solver.core.config.solver.EnvironmentMode;
+import ai.timefold.solver.core.impl.util.Pair;
+import ai.timefold.solver.core.impl.util.Quadruple;
+import ai.timefold.s... | It appears we no longer need the tuple here.
(Which also means we can now replace all declarations of `GroupNodeConstructor` variables with `var`.)
Actually, I checked the history and this was the case on `main` already.
Please make the changes anyway - it's a good thing to leave the code in a better shape than th... |
timefold-solver | github_2023 | java | 401 | TimefoldAI | triceo | @@ -36,6 +157,128 @@ AbstractNode apply(int groupStoreIndex, TupleLifecycle<Tuple_> nextNodesTupleLif
}
+ @FunctionalInterface
+ interface NoKeysOneCollectorGroupByBuilder<CollectorA_, Tuple_ extends AbstractTuple> {
+ AbstractNode build(int groupStoreIndex, int undoStoreIndex,
+ Co... | Let's stay consistent with the node naming and use the `GroupByXMappingYCollectorNodeBuilder` pattern in these names. |
timefold-solver | github_2023 | java | 401 | TimefoldAI | triceo | @@ -36,6 +180,128 @@ AbstractNode apply(int groupStoreIndex, TupleLifecycle<Tuple_> nextNodesTupleLif
}
+ @FunctionalInterface
+ interface GroupBy0Mapping1CollectorNodeBuilder<CollectorA_, Tuple_ extends AbstractTuple> {
+ AbstractNode build(int groupStoreIndex, int undoStoreIndex,
+ ... | One thing just crossed my mind... if we don't need `Collector_` to extend `...ConstraintCollector`, do we need `Tuple_` to extend `AbstractTuple`?
If not, even more boilerplate to remove right here. |
timefold-solver | github_2023 | java | 392 | TimefoldAI | triceo | @@ -80,90 +167,56 @@ public final class ConstraintCollectors {
* @return never null
*/
public static <A> UniConstraintCollector<A, ?, Integer> count() {
- return new DefaultUniConstraintCollector<>(
- MutableInt::new,
- (resultContainer, a) -> innerCount(resultConta... | We didn't do this before, but now that I see it like this, why not make these argless collectors a singleton? Works nicely with the factory idea below. |
timefold-solver | github_2023 | java | 392 | TimefoldAI | triceo | @@ -47,23 +40,117 @@
import ai.timefold.solver.core.api.score.stream.tri.TriConstraintCollector;
import ai.timefold.solver.core.api.score.stream.uni.UniConstraintCollector;
import ai.timefold.solver.core.api.score.stream.uni.UniConstraintStream;
-import ai.timefold.solver.core.impl.util.MutableInt;
-import ai.timefo... | General comment: please make it so that the collector implementations aren't public. I do not want them to show up in IDEs. If there needs to be a public factory for them, so be it. Let's call it `InnerConstraintCollectors`, to follow the `Inner...` convention we already use elsewhere.
You may think it's boilerplate... |
timefold-solver | github_2023 | java | 392 | TimefoldAI | triceo | @@ -0,0 +1,9 @@
+package ai.timefold.solver.core.impl.score.stream;
+
+public interface IntCalculator<Output_> { | Let's start getting used to `sealed` types. Let's seal every interface in the collector impl hierarchy. |
timefold-solver | github_2023 | java | 392 | TimefoldAI | triceo | @@ -0,0 +1,33 @@
+package ai.timefold.solver.core.impl.score.stream;
+
+import java.util.ArrayList;
+import java.util.Collection;
+import java.util.List;
+import java.util.function.IntFunction;
+
+public final class CustomCollectionUndoableActionable<Mapped, Result extends Collection<Mapped>> | Let's follow the convention that generic type arguments end with `_`, so that we can easily distinguish them from non-generic ones.
I know we didn't do it in `ConstraintCollectors`, because the class already started like that and we kept it internally consistent throughout the years. But now that this is new code, l... |
timefold-solver | github_2023 | java | 392 | TimefoldAI | triceo | @@ -1547,14 +1121,14 @@ private static <A, B, C, D, Mapped> QuadConstraintCollector<A, B, C, D, SortedMa
* @return never null
*/
public static <A extends Comparable<A>> UniConstraintCollector<A, ?, SortedSet<A>> toSortedSet() {
- return toSortedSet(a -> a);
+ return toSortedSet(ConstantLa... | No deprecation after all? |
timefold-solver | github_2023 | java | 392 | TimefoldAI | triceo | @@ -80,90 +167,56 @@ public final class ConstraintCollectors {
* @return never null
*/
public static <A> UniConstraintCollector<A, ?, Integer> count() {
- return new DefaultUniConstraintCollector<>(
- MutableInt::new,
- (resultContainer, a) -> innerCount(resultConta... | And because these are no longer used and they are not public, we can delete the `Default...` implementations. Nice! |
timefold-solver | github_2023 | java | 392 | TimefoldAI | triceo | @@ -0,0 +1,17 @@
+package ai.timefold.solver.core.impl.score.stream;
+
+public final class IntCounter {
+ private int count;
+
+ public void increment() {
+ count++;
+ }
+
+ public void decrement() {
+ count--;
+ }
+
+ public Integer result() { | Since there are no generics here, can we return `int`? Let's return what we have, and deal with boxing elsewhere. |
timefold-solver | github_2023 | java | 392 | TimefoldAI | triceo | @@ -0,0 +1,17 @@
+package ai.timefold.solver.core.impl.score.stream;
+
+public final class LongCounter {
+ private long count;
+
+ public void increment() {
+ count++;
+ }
+
+ public void decrement() {
+ count--;
+ }
+
+ public Long result() { | `long` instead of `Long`? |
timefold-solver | github_2023 | java | 392 | TimefoldAI | triceo | @@ -0,0 +1,47 @@
+package ai.timefold.solver.core.impl.score.stream;
+
+import java.util.LinkedHashMap;
+import java.util.Map;
+import java.util.function.BinaryOperator;
+import java.util.stream.Stream;
+
+public final class ToMapPerKeyCounter<Value_> {
+
+ private final Map<Value_, Long> counts = new LinkedHashMap<... | ```suggestion
.orElseThrow(() -> new IllegalStateException("Impossible state: Should have had at least one value."));
```
"Impossible state" seems to be the convention for this. |
timefold-solver | github_2023 | java | 392 | TimefoldAI | triceo | @@ -0,0 +1,114 @@
+package ai.timefold.solver.core.impl.score.stream.bi;
+
+import java.util.Objects;
+import java.util.function.Function;
+import java.util.function.Supplier;
+
+import ai.timefold.solver.core.api.function.QuadFunction;
+import ai.timefold.solver.core.api.function.TriFunction;
+import ai.timefold.solve... | I'd argue we can avoid this craziness with `var`.
The type is already on the right-hand side of the assignment, no need to duplicate it.
Let's try to apply it consistently. |
timefold-solver | github_2023 | java | 392 | TimefoldAI | triceo | @@ -0,0 +1,42 @@
+package ai.timefold.solver.core.impl.score.stream.bi;
+
+import java.util.function.Function;
+import java.util.function.Supplier;
+
+import ai.timefold.solver.core.api.function.TriFunction;
+import ai.timefold.solver.core.api.score.stream.bi.BiConstraintCollector;
+import ai.timefold.solver.core.impl.... | If you make argless `count...()` a singleton, you get to avoid this, and also needless instances.
(Not that it'd ever be a problem.) |
timefold-solver | github_2023 | java | 336 | TimefoldAI | rsynek | @@ -0,0 +1,84 @@
+package ai.timefold.solver.core.api.score.constraint;
+
+import java.util.Objects;
+
+import ai.timefold.solver.core.api.domain.constraintweight.ConstraintConfiguration;
+import ai.timefold.solver.core.api.domain.constraintweight.ConstraintWeight;
+
+/**
+ * Represents a unique identifier of a constra... | nitpick: `parseConstraintId` as there is already the method `composeConstraintId`. By using the same name we make sure the user understands it's the same concept. |
timefold-solver | github_2023 | java | 336 | TimefoldAI | rsynek | @@ -174,22 +174,40 @@ public boolean isConstraintMatchEnabled() {
public final Map<String, ConstraintMatchTotal<Score_>> getConstraintMatchTotalMap() {
if (constraintMatchTotalMap == null) {
- rebuildConstraintMatchTotalsAndIndictments();
+ rebuildConstraintMatchTotals();
... | nitpick: `constraintIdToConstraintMatchTotalMap` gives an idea what the key is. |
timefold-solver | github_2023 | java | 336 | TimefoldAI | rsynek | @@ -2,13 +2,16 @@
import java.util.List;
-public class TimefoldDevUIProperties {
+import ai.timefold.solver.core.api.score.constraint.ConstraintRef;
+
+public class TimefoldDevUIProperties { // TODO make record? | Is it a leftover or a suggestion for some next PR (just checking)? |
timefold-solver | github_2023 | java | 336 | TimefoldAI | rsynek | @@ -0,0 +1,134 @@
+package ai.timefold.solver.core.api.score.analysis;
+
+import java.util.List;
+import java.util.Map;
+import java.util.Objects;
+import java.util.stream.Collectors;
+import java.util.stream.Stream;
+
+import ai.timefold.solver.core.api.score.Score;
+import ai.timefold.solver.core.api.score.constraint... | (relevant to all records in this package): Not sure if we should take it that far, but instead of this note, we could expose just interfaces with getters and move these records implementing them to the `impl` package. That way, we make instantiation impossible. |
timefold-solver | github_2023 | java | 336 | TimefoldAI | rsynek | @@ -1,19 +1,47 @@
package ai.timefold.solver.core.api.score.stream;
+import java.util.UUID;
+
+import ai.timefold.solver.core.api.score.Score;
+import ai.timefold.solver.core.api.score.analysis.MatchAnalysis;
+import ai.timefold.solver.core.api.score.analysis.ScoreAnalysis;
import ai.timefold.solver.core.api.score.... | nitpicking: "random" does not necessarily imply the ID is unique; I would rather use "unique ID" |
timefold-solver | github_2023 | java | 336 | TimefoldAI | rsynek | @@ -86,4 +78,85 @@ private <Result_> Result_ callScoreDirector(Solution_ solution,
return function.apply(scoreDirector);
}
}
+
+ @Override
+ public ScoreExplanation<Solution_, Score_> explain(Solution_ solution, SolutionUpdatePolicy solutionUpdatePolicy) {
+ var currentScore = (S... | Is there a reason for having this method public on the implementation class? |
timefold-solver | github_2023 | java | 336 | TimefoldAI | rsynek | @@ -41,6 +57,126 @@ void polymorphicScore() {
.isEqualTo(BendableScore.of(new int[] { -1, -20 }, new int[] { -300, -4000, -50000 }));
}
+ @Test
+ void scoreAnalysisWithoutMatches() throws JsonProcessingException {
+ ObjectMapper objectMapper = new ObjectMapper();
+ objectMapp... | Can the justification be just any object and it's a list just in this example, or is it always a list? |
timefold-solver | github_2023 | java | 374 | TimefoldAI | triceo | @@ -252,4 +252,12 @@ default void expandToQuad() {
void impactNegativeBigDecimalCustomJustifications();
+ // ************************************************************************ | Let's start a new `NodeSharingTest` for this. |
timefold-solver | github_2023 | java | 374 | TimefoldAI | triceo | @@ -3000,4 +3008,190 @@ public void joinerEqualsAndSameness() {
assertMatch(entity3, entity2));
}
+ // ************************************************************************
+ // Node Sharing
+ // ************************************************************************
+
+ @Overrid... | This is needlessly complicated. Since you never need to call those constraint streams, you don't need `buildScoreDirector` or `failureList`.
Let's change the test to look like this:
Assertions.assertThat(base.join(...))
.as(...)
.isSameAs(base.join(...)
Maybe you'll need to figure out how... |
timefold-solver | github_2023 | java | 374 | TimefoldAI | triceo | @@ -0,0 +1,216 @@
+package ai.timefold.solver.core.impl.util;
+
+import java.util.Objects;
+import java.util.function.BiFunction;
+import java.util.function.BiPredicate;
+import java.util.function.Function;
+import java.util.function.ToIntBiFunction;
+import java.util.function.ToIntFunction;
+
+import ai.timefold.solve... | Let's add a private constructor to prevent instances. |
timefold-solver | github_2023 | java | 374 | TimefoldAI | triceo | @@ -0,0 +1,216 @@
+package ai.timefold.solver.core.impl.util;
+
+import java.util.Objects;
+import java.util.function.BiFunction;
+import java.util.function.BiPredicate;
+import java.util.function.Function;
+import java.util.function.ToIntBiFunction;
+import java.util.function.ToIntFunction;
+
+import ai.timefold.solve... | ```suggestion
* A class that holds common lambdas that are guaranteed to be the same across method calls. In most JDK's, stateless lambdas are
``` |
timefold-solver | github_2023 | java | 374 | TimefoldAI | triceo | @@ -0,0 +1,216 @@
+package ai.timefold.solver.core.impl.util;
+
+import java.util.Objects;
+import java.util.function.BiFunction;
+import java.util.function.BiPredicate;
+import java.util.function.Function;
+import java.util.function.ToIntBiFunction;
+import java.util.function.ToIntFunction;
+
+import ai.timefold.solve... | Let's have consistent whitespace. Either use newlines, or don't.
If you want to separate blocks of lambdas, do so, but do it consistently. |
timefold-solver | github_2023 | others | 379 | TimefoldAI | triceo | @@ -66,6 +66,23 @@ $ mvn exec:java
This is an open source project, and you are more than welcome to contribute!
For more, see link:CONTRIBUTING.adoc[Contributing].
+== Editions
+
+There are 2 editions of Timefold Solver:
+
+- Timefold Solver Community Edition (CE) (this repo).
+- Timefold Solver Enterprise Edition ... | ```suggestion
=== Key Features of Enterprise Edition (EE)
``` |
timefold-solver | github_2023 | others | 379 | TimefoldAI | triceo | @@ -66,6 +66,23 @@ $ mvn exec:java
This is an open source project, and you are more than welcome to contribute!
For more, see link:CONTRIBUTING.adoc[Contributing].
+== Editions
+
+There are 2 editions of Timefold Solver:
+
+- Timefold Solver Community Edition (CE) (this repo).
+- Timefold Solver Enterprise Edition ... | ```suggestion
=== Licensing and Usage
``` |
timefold-solver | github_2023 | others | 379 | TimefoldAI | triceo | @@ -66,6 +66,23 @@ $ mvn exec:java
This is an open source project, and you are more than welcome to contribute!
For more, see link:CONTRIBUTING.adoc[Contributing].
+== Editions
+
+There are 2 editions of Timefold Solver:
+
+- Timefold Solver Community Edition (CE) (this repo).
+- Timefold Solver Enterprise Edition ... | ```suggestion
- **Nearby Selection:** Get better solutions quicker, especially with spatial problems.
``` |
timefold-solver | github_2023 | others | 379 | TimefoldAI | triceo | @@ -66,6 +66,23 @@ $ mvn exec:java
This is an open source project, and you are more than welcome to contribute!
For more, see link:CONTRIBUTING.adoc[Contributing].
+== Editions
+
+There are 2 editions of Timefold Solver:
+
+- Timefold Solver Community Edition (CE) (this repo).
+- Timefold Solver Enterprise Edition ... | ```suggestion
=== Key Features of Timefold Solver Enterprise Edition (EE)
``` |
timefold-solver | github_2023 | java | 369 | TimefoldAI | triceo | @@ -0,0 +1,59 @@
+package ai.timefold.solver.core.impl.util;
+
+import java.util.Objects;
+import java.util.function.BiFunction;
+import java.util.function.BiPredicate;
+import java.util.function.Function;
+import java.util.function.ToIntBiFunction;
+import java.util.function.ToIntFunction;
+
+import ai.timefold.solver... | If these functions are returned via static methods, you can get the casting done inside the methods, as opposed to plaguing the entire codebase with it.
Example:
public static final <A, B> BiPredicate<A, B> notEquals() {
return (BiPredicate<A, B>) NOT_EQUALS;
} |
timefold-solver | github_2023 | others | 81 | TimefoldAI | rsynek | @@ -0,0 +1,54 @@
+name: Downstream - Timefold Quickstarts
+
+on:
+ pull_request:
+ branches: [main]
+ paths-ignore:
+ - 'LICENSE*'
+ - '.gitignore'
+ - '**.md'
+ - '**.adoc'
+ - '*.txt'
+
+jobs:
+ build:
+ runs-on: ubuntu-latest
+ concurrency:
+ group: downstream-quickstarts-... | Can we set the executable flag for the script as a part of this PR instead of doing it each time in the GHA? |
timefold-solver | github_2023 | others | 81 | TimefoldAI | rsynek | @@ -0,0 +1,54 @@
+name: Downstream - Timefold Quickstarts
+
+on:
+ pull_request:
+ branches: [main] | We will eventually have also release branches. |
timefold-solver | github_2023 | others | 81 | TimefoldAI | rsynek | @@ -0,0 +1,54 @@
+name: Downstream - Timefold Quickstarts
+
+on:
+ pull_request: | That's a question. Cloning the PR author repositories without merging with the target repositories does not protected from stale PR, which might, in theory, result in issues only after merging the PR.
If we want to cover this corner case, we have to either A) merge the forks with origin repos before the build or B) ... |
timefold-solver | github_2023 | java | 340 | TimefoldAI | Christopher-Chianelli | @@ -65,7 +65,7 @@ private static RuntimeException createJustificationException(Constraint constrai
private static String factsToString(Object... facts) {
return Arrays.stream(facts)
- .map(Object::toString)
+ .map(s -> s == null ? null : s.toString()) | Maybe `Objects::toString`? https://docs.oracle.com/javase/8/docs/api/java/util/Objects.html#toString-java.lang.Object- |
timefold-solver | github_2023 | others | 340 | TimefoldAI | Christopher-Chianelli | @@ -1514,10 +1514,14 @@ See <<constraintStreamsDealingWithDuplicateTuplesUsingDistinct,`distinct()`>> fo
[[constraintStreamsConcat]]
==== Concat
-The `concat` building block allows you to create a constraint stream containing tuples of two constraint streams of the same <<constraintStreamsCardinality,cardinality>>.... | Inaccurate, since now multiple cardinalities are allowed |
timefold-solver | github_2023 | java | 346 | TimefoldAI | triceo | @@ -52,8 +52,11 @@ public class GizmoSolutionClonerImplementor {
Object.class);
private static final MethodDescriptor PUT_METHOD = MethodDescriptor.ofMethod(Map.class, "put", Object.class,
Object.class, Object.class);
+ private static final String FALLBACK_CLONER = "fallbackCloner";
... | Let's figure out a different way to pass this information. Static mutable fields are fundamentally wrong, and we've seen it numerous times already.
Do we need a Quarkus-specific subtype? So be it.
Quarkus-specific constructor? Sure.
Just not this, please.
In my opinion, the fact that there are ifs for particula... |
timefold-solver | github_2023 | java | 346 | TimefoldAI | triceo | @@ -28,7 +28,8 @@ public static <T> SolutionCloner<T> build(SolutionDescriptor<T> solutionDescript
") the classpath or modulepath must contain io.quarkus.gizmo:gizmo.\n" +
"Maybe add a dependency to io.quarkus.gizmo:gizmo.");
}
- return GizmoSolutionClonerImplem... | Why not:
```suggestion
return GizmoSolutionClonerImplementor.createClonerFor(solutionDescriptor,
```
Seems like the instance creation is just boilerplate, it can happen internally. People need not know what's happening under the covers. |
timefold-solver | github_2023 | java | 346 | TimefoldAI | triceo | @@ -100,14 +101,20 @@ public static Comparator<Class<?>> getInstanceOfComparator(Set<Class<?>> deepClo
.thenComparing(Class::getName).reversed();
}
+ protected void createFields(ClassCreator classCreator) {
+ classCreator.getFieldCreator(FALLBACK_CLONER, FieldAccessingSolutionCloner.cl... | Could have stayed static. |
timefold-solver | github_2023 | java | 346 | TimefoldAI | triceo | @@ -417,6 +419,76 @@ public String generateGizmoBeanFactory(ClassOutput classOutput, Set<Class<?>> be
return generatedClassName;
}
+ private static class QuarkusGizmoSolutionClonerImplementor extends GizmoSolutionClonerImplementor { | The class is IMO a big too big to be an inner class. Consider making it a top-level class; it's already static, so you don't need the reference to the outer class. |
timefold-solver | github_2023 | java | 346 | TimefoldAI | triceo | @@ -69,7 +69,7 @@ protected <Solution_> SolutionCloner<Solution_> createSolutionCloner(SolutionDes
generateGizmoSolutionOrEntityDescriptor(solutionDescriptor, clazz));
});
- GizmoSolutionClonerImplementor.defineClonerFor(classCreator, solutionDescriptor,
+ n... | Same comment as above; I think it could've been made static and hidden the instance. |
timefold-solver | github_2023 | java | 338 | TimefoldAI | triceo | @@ -125,11 +129,26 @@ private Optional<ListVariableDescriptor<?>> findValidListVariableDescriptor(
+ listVariableDescriptors + ").");
}
+ ListVariableDescriptor<Solution_> listVariableDescriptor = listVariableDescriptors.get(0);
+ failIfBasicAndListVariablesAreCombinedOnSin... | I'd argue this needs to happen earlier, before the CH, when the solver config is being processed. |
timefold-solver | github_2023 | java | 338 | TimefoldAI | triceo | @@ -125,11 +129,26 @@ private Optional<ListVariableDescriptor<?>> findValidListVariableDescriptor(
+ listVariableDescriptors + ").");
}
+ ListVariableDescriptor<Solution_> listVariableDescriptor = listVariableDescriptors.get(0);
+ failIfBasicAndListVariablesAreCombinedOnSin... | ```suggestion
throw new UnsupportedOperationException("""
Combining basic variables (%s) with list variables (%) on a single planning entity (%s) is currently not supported.
""".formatted(basicVariableDescriptors, listVariableDescriptor, listVariableDescriptor.get... |
timefold-solver | github_2023 | java | 338 | TimefoldAI | triceo | @@ -221,6 +224,20 @@ void variableWithPlanningIdIsARecord() {
Assertions.assertThatNoException().isThrownBy(() -> solver.solve(solution));
}
+ @Test
+ void entityWithMixedBasicAndListPlanningVariables() {
+ var solverConfig = new SolverConfig()
+ .withSolutionClass(DummySolut... | This last line shouldn't be necessary. AFAIK phases are auto-configured if not present. |
timefold-solver | github_2023 | java | 338 | TimefoldAI | triceo | @@ -549,6 +551,31 @@ private void determineGlobalShadowOrder() {
}
}
+ private void validateListVariableDescriptors() {
+ if (listVariableDescriptors.isEmpty()) {
+ return;
+ }
+
+ if (listVariableDescriptors.size() > 1) {
+ throw new UnsupportedOperationExc... | ```suggestion
"Combining basic variables (%s) with list variables (%s) on a single planning entity (%s)" +
" is currently not supported."
``` |
timefold-solver | github_2023 | java | 338 | TimefoldAI | triceo | @@ -549,6 +551,31 @@ private void determineGlobalShadowOrder() {
}
}
+ private void validateListVariableDescriptors() {
+ if (listVariableDescriptors.isEmpty()) {
+ return;
+ }
+
+ if (listVariableDescriptors.size() > 1) {
+ throw new UnsupportedOperationExc... | ```suggestion
listVariableDescriptor.getEntityDescriptor().getEntityClass().getCanonicalName()));
``` |
timefold-solver | github_2023 | others | 337 | TimefoldAI | triceo | @@ -1030,9 +1030,14 @@ public class StartTimeUpdatingVariableListener implements VariableListener<TaskA
==== Planning value
A planning value is a possible value for a genuine planning variable.
-Usually, a planning value is a problem fact, but it can also be any object, for example a ``double``.
+Usually, a plannin... | ```suggestion
Usually, a planning value is a problem fact, but it can also be any object, for example an ``Integer``.
``` |
timefold-solver | github_2023 | others | 337 | TimefoldAI | triceo | @@ -1030,9 +1030,14 @@ public class StartTimeUpdatingVariableListener implements VariableListener<TaskA
==== Planning value
A planning value is a possible value for a genuine planning variable.
-Usually, a planning value is a problem fact, but it can also be any object, for example a ``double``.
+Usually, a plannin... | ```suggestion
Primitive types (such as ``int``) are not allowed.
``` |
timefold-solver | github_2023 | java | 297 | TimefoldAI | triceo | @@ -140,6 +141,13 @@ private long determineLayerIndex(AbstractNode node, NodeBuildHelper<Score_> buil
var leftParentNode = buildHelper.findParentNode(leftParent);
var rightParentNode = buildHelper.findParentNode(rightParent);
return Math.max(leftParentNode.getLayerIndex(), rightPa... | Why would concat implement join? The two operations have barely anything in common. |
timefold-solver | github_2023 | java | 297 | TimefoldAI | triceo | @@ -359,6 +359,19 @@ public BiConstraintStream<A, B> distinct() {
}
}
+ @Override
+ public BiConstraintStream<A, B> concat(BiConstraintStream<A, B> otherStream) {
+ var other = (BavetAbstractBiConstraintStream<Solution_, A, B>) otherStream;
+ var leftBridge = new BavetForeBridgeBiCon... | This is not a join; let's not call it like that. |
timefold-solver | github_2023 | java | 297 | TimefoldAI | triceo | @@ -0,0 +1,19 @@
+package ai.timefold.solver.constraint.streams.bavet.bi;
+
+import ai.timefold.solver.constraint.streams.bavet.common.AbstractConcatNode;
+import ai.timefold.solver.constraint.streams.bavet.common.tuple.BiTuple;
+import ai.timefold.solver.constraint.streams.bavet.common.tuple.TupleLifecycle;
+
+public ... | Since concat does not modify the tuples, it doesn't need to create new ones.
I'm betting it will be much more efficient if it's rewritten as a pass-through node, without node creation. |
timefold-solver | github_2023 | java | 297 | TimefoldAI | triceo | @@ -0,0 +1,103 @@
+package ai.timefold.solver.constraint.streams.bavet.common;
+
+import static ai.timefold.solver.constraint.streams.bavet.common.tuple.TupleState.ABORTING;
+import static ai.timefold.solver.constraint.streams.bavet.common.tuple.TupleState.CREATING;
+import static ai.timefold.solver.constraint.streams.... | There aren't any indexes. |
timefold-solver | github_2023 | java | 297 | TimefoldAI | triceo | @@ -0,0 +1,103 @@
+package ai.timefold.solver.constraint.streams.bavet.common;
+
+import static ai.timefold.solver.constraint.streams.bavet.common.tuple.TupleState.ABORTING;
+import static ai.timefold.solver.constraint.streams.bavet.common.tuple.TupleState.CREATING;
+import static ai.timefold.solver.constraint.streams.... | These vars need better naming, as there are no indexes and no out tuple list. |
timefold-solver | github_2023 | java | 297 | TimefoldAI | triceo | @@ -0,0 +1,103 @@
+package ai.timefold.solver.constraint.streams.bavet.common;
+
+import static ai.timefold.solver.constraint.streams.bavet.common.tuple.TupleState.ABORTING;
+import static ai.timefold.solver.constraint.streams.bavet.common.tuple.TupleState.CREATING;
+import static ai.timefold.solver.constraint.streams.... | If you store the node-specific tuple state in the store instead of the outtuple, you won't need to create the outtuple just to store the state. |
timefold-solver | github_2023 | java | 297 | TimefoldAI | triceo | @@ -0,0 +1,104 @@
+package ai.timefold.solver.constraint.streams.bavet.uni;
+
+import java.util.Objects;
+import java.util.Set;
+
+import ai.timefold.solver.constraint.streams.bavet.BavetConstraintFactory;
+import ai.timefold.solver.constraint.streams.bavet.common.BavetAbstractConstraintStream;
+import ai.timefold.solv... | Not a join node.
(Same comment also likely in Bi, Tri and Quad.) |
timefold-solver | github_2023 | java | 297 | TimefoldAI | triceo | @@ -140,6 +142,13 @@ private long determineLayerIndex(AbstractNode node, NodeBuildHelper<Score_> buil
var leftParentNode = buildHelper.findParentNode(leftParent);
var rightParentNode = buildHelper.findParentNode(rightParent);
return Math.max(leftParentNode.getLayerIndex(), rightPa... | This code seems to repeat a lot now. Maybe we extract it? |
timefold-solver | github_2023 | others | 297 | TimefoldAI | triceo | @@ -1511,6 +1511,47 @@ the tuple `(SomePerson, USER)` is sent downstream twice.
See <<constraintStreamsDealingWithDuplicateTuplesUsingDistinct,`distinct()`>> for how to deal with duplicate tuples.
====
+[[constraintStreamsConcat]]
+==== Concat
+
+Concat allows you to create a constraint stream containing tuples of ... | ```suggestion
Unlike union of sets, concatenation of lists repeats duplicated elements.
``` |
timefold-solver | github_2023 | others | 297 | TimefoldAI | triceo | @@ -1511,6 +1511,47 @@ the tuple `(SomePerson, USER)` is sent downstream twice.
See <<constraintStreamsDealingWithDuplicateTuplesUsingDistinct,`distinct()`>> for how to deal with duplicate tuples.
====
+[[constraintStreamsConcat]]
+==== Concat
+
+Concat allows you to create a constraint stream containing tuples of ... | ```suggestion
If this is undesired, use the <<constraintStreamsDealingWithDuplicateTuplesUsingDistinct,`distinct` building block>>.
``` |
timefold-solver | github_2023 | others | 297 | TimefoldAI | triceo | @@ -1511,6 +1511,47 @@ the tuple `(SomePerson, USER)` is sent downstream twice.
See <<constraintStreamsDealingWithDuplicateTuplesUsingDistinct,`distinct()`>> for how to deal with duplicate tuples.
====
+[[constraintStreamsConcat]]
+==== Concat
+
+Concat allows you to create a constraint stream containing tuples of ... | I am not sure this is a good example. The first question once a smart person sees this is: why not just use a much simple `.filter(hasCat || hasDog)`?
Even though you explain that the single filter approach would actually do something different, I have to ask - why? Why would I want the employee there twice? (It was... |
timefold-solver | github_2023 | others | 297 | TimefoldAI | triceo | @@ -1511,6 +1511,47 @@ the tuple `(SomePerson, USER)` is sent downstream twice.
See <<constraintStreamsDealingWithDuplicateTuplesUsingDistinct,`distinct()`>> for how to deal with duplicate tuples.
====
+[[constraintStreamsConcat]]
+==== Concat
+
+Concat allows you to create a constraint stream containing tuples of ... | ```suggestion
This correctly counts cats and dogs when an Employee has both a cat and a dog.
``` |
timefold-solver | github_2023 | others | 297 | TimefoldAI | triceo | @@ -1511,6 +1511,57 @@ the tuple `(SomePerson, USER)` is sent downstream twice.
See <<constraintStreamsDealingWithDuplicateTuplesUsingDistinct,`distinct()`>> for how to deal with duplicate tuples.
====
+[[constraintStreamsConcat]]
+==== Concat
+
+Concat allows you to create a constraint stream containing tuples of ... | ```suggestion
The `concat` building block allows you to create a constraint stream containing tuples of two constraint streams of the same <<constraintStreamsCardinality,cardinality>>.
``` |
timefold-solver | github_2023 | others | 297 | TimefoldAI | triceo | @@ -1511,6 +1511,57 @@ the tuple `(SomePerson, USER)` is sent downstream twice.
See <<constraintStreamsDealingWithDuplicateTuplesUsingDistinct,`distinct()`>> for how to deal with duplicate tuples.
====
+[[constraintStreamsConcat]]
+==== Concat
+
+Concat allows you to create a constraint stream containing tuples of ... | ```suggestion
If <<constraintStreamsJoin,join>> acts like a cartesian product of two lists, `concat` acts like a concatenation of two lists.
``` |
timefold-solver | github_2023 | others | 297 | TimefoldAI | triceo | @@ -1511,6 +1511,57 @@ the tuple `(SomePerson, USER)` is sent downstream twice.
See <<constraintStreamsDealingWithDuplicateTuplesUsingDistinct,`distinct()`>> for how to deal with duplicate tuples.
====
+[[constraintStreamsConcat]]
+==== Concat
+
+Concat allows you to create a constraint stream containing tuples of ... | ```suggestion
If the two constraint stream parents share tuples, which happens eg. when the streams being concatenated come from the same source of data, the tuples will be repeated downstream.
``` |
timefold-solver | github_2023 | others | 297 | TimefoldAI | triceo | @@ -1511,6 +1511,57 @@ the tuple `(SomePerson, USER)` is sent downstream twice.
See <<constraintStreamsDealingWithDuplicateTuplesUsingDistinct,`distinct()`>> for how to deal with duplicate tuples.
====
+[[constraintStreamsConcat]]
+==== Concat
+
+Concat allows you to create a constraint stream containing tuples of ... | ```suggestion
This correctly counts the number of shifts each Employee has, *even when the Employee has no shifts*.
``` |
timefold-solver | github_2023 | others | 297 | TimefoldAI | triceo | @@ -1511,6 +1511,57 @@ the tuple `(SomePerson, USER)` is sent downstream twice.
See <<constraintStreamsDealingWithDuplicateTuplesUsingDistinct,`distinct()`>> for how to deal with duplicate tuples.
====
+[[constraintStreamsConcat]]
+==== Concat
+
+Concat allows you to create a constraint stream containing tuples of ... | ```suggestion
If it was implemented without `concat` like this:
``` |
timefold-solver | github_2023 | others | 297 | TimefoldAI | triceo | @@ -1511,6 +1511,57 @@ the tuple `(SomePerson, USER)` is sent downstream twice.
See <<constraintStreamsDealingWithDuplicateTuplesUsingDistinct,`distinct()`>> for how to deal with duplicate tuples.
====
+[[constraintStreamsConcat]]
+==== Concat
+
+Concat allows you to create a constraint stream containing tuples of ... | ```suggestion
An employee with no assigned shifts _will not be penalized because no tuples were passed to the `groupBy` building block_.
``` |
timefold-solver | github_2023 | java | 297 | TimefoldAI | triceo | @@ -16,4 +16,10 @@ public final class BavetBiConcatNode<A, B> extends AbstractConcatNode<BiTuple<A,
protected BiTuple<A, B> getOutTuple(BiTuple<A, B> inTuple) {
return new BiTuple<>(inTuple.factA, inTuple.factB, outputStoreSize);
}
+
+ @Override
+ protected void updateOutTuple(BiTuple<A, B> inT... | Tuples have `updateInDifferent(...)` just for this case.
Turns out it is a common operation, so it was extracted already. |
timefold-solver | github_2023 | java | 308 | TimefoldAI | rsynek | @@ -487,10 +487,28 @@ private static <C> Member getSingleMember(Class<C> clazz, Class<? extends Annota
if (memberList.isEmpty()) {
return null;
}
- if (memberList.size() > 1) {
- throw new IllegalArgumentException("The class (" + clazz
- + ") has " + m... | Is it guaranteed that the second member in the list is the getter and not the field? |
timefold-solver | github_2023 | others | 284 | TimefoldAI | triceo | @@ -18,7 +18,7 @@
<!-- Code of integration tests should not be a part of test coverage reports. -->
<sonar.coverage.exclusions>**/*</sonar.coverage.exclusions>
<!-- Quarkus 2 and 3 use different context roots. -->
- <dev.ui.root>/q/dev-v1</dev.ui.root>
+ <dev.ui.root>/q/dev-ui</dev.ui.root> | Since `main` only uses Quarkus 3, I think we can go back to hardcoding this and removing the comment. |
timefold-solver | github_2023 | others | 284 | TimefoldAI | triceo | @@ -651,7 +651,8 @@
<profile>
<id>sonarcloud-analysis</id>
<properties>
- <sonar.coverage.jacoco.xmlReportPaths>${project.reporting.outputDirectory}/jacoco/jacoco.xml</sonar.coverage.jacoco.xmlReportPaths>
+ <sonar.coverage.jacoco.xmlReportPaths>${project.reporting.outputDirectory}/jaco... | Unrelated change? |
timefold-solver | github_2023 | java | 284 | TimefoldAI | triceo | @@ -148,20 +152,48 @@ DetermineIfNativeBuildItem ifNativeBuild() {
}
@BuildStep(onlyIf = IsDevelopment.class)
- public DevConsoleRuntimeTemplateInfoBuildItem getSolverConfig(SolverConfigBuildItem solverConfigBuildItem,
- CurateOutcomeBuildItem curateOutcomeBuildItem) {
+ @Record(STATIC_INIT... | Shall we make it "Timefold Solver"? That way, it'd be consistent with what the project is actually called. |
timefold-solver | github_2023 | javascript | 284 | TimefoldAI | triceo | @@ -0,0 +1,61 @@
+import {css, html, LitElement} from 'lit';
+import {JsonRpc} from 'jsonrpc';
+import '@vaadin/icon';
+import '@vaadin/button';
+import {until} from 'lit/directives/until.js';
+import '@vaadin/grid';
+import '@vaadin/grid/vaadin-grid-sort-column.js';
+
+export class ConfigComponent extends LitElement {... | Timefold Solver? |
timefold-solver | github_2023 | javascript | 284 | TimefoldAI | triceo | @@ -0,0 +1,76 @@
+import {css, html, LitElement} from 'lit';
+import {JsonRpc} from 'jsonrpc';
+import '@vaadin/icon';
+import '@vaadin/button';
+import {until} from 'lit/directives/until.js';
+import '@vaadin/grid';
+import {columnBodyRenderer} from '@vaadin/grid/lit.js';
+import '@vaadin/grid/vaadin-grid-sort-column.... | Dtto. |
timefold-solver | github_2023 | javascript | 284 | TimefoldAI | triceo | @@ -0,0 +1,104 @@
+import {css, html, LitElement} from 'lit';
+import {JsonRpc} from 'jsonrpc';
+import '@vaadin/icon';
+import '@vaadin/button';
+import {until} from 'lit/directives/until.js';
+import '@vaadin/grid';
+import {columnBodyRenderer} from '@vaadin/grid/lit.js';
+import '@vaadin/grid/vaadin-grid-sort-column... | Dtto. |
timefold-solver | github_2023 | java | 284 | TimefoldAI | triceo | @@ -57,51 +46,28 @@ void testSolverConfigPage() throws ParserConfigurationException, SAXException, I
}
@Test
- void testModelPage() throws ParserConfigurationException, SAXException, IOException {
- String body = RestAssured.get(getPage("model"))
- .then()
- .extract(... | I'd use the `class.getCanonicalName()` throughout this test, to make this safe for refactoring. |
timefold-solver | github_2023 | others | 258 | TimefoldAI | rsynek | @@ -86,25 +91,13 @@ jobs:
out/jreleaser/trace.log
out/jreleaser/output.properties
+ # Pull Request will be created with the changes and a summary of next steps.
- name: Set micro snapshot version on the release branch | ```suggestion
- name: Put back the 999-SNAPSHOT version on the release branch
``` |
timefold-solver | github_2023 | others | 258 | TimefoldAI | rsynek | @@ -86,25 +91,13 @@ jobs:
out/jreleaser/trace.log
out/jreleaser/output.properties
+ # Pull Request will be created with the changes and a summary of next steps.
- name: Set micro snapshot version on the release branch
run: |
git checkout -B ${{ github.event.in... | ```suggestion
git checkout -B ${{ github.event.inputs.releaseBranch }}-put-back-999-snapshot
``` |
timefold-solver | github_2023 | others | 258 | TimefoldAI | rsynek | @@ -86,25 +91,13 @@ jobs:
out/jreleaser/trace.log
out/jreleaser/output.properties
+ # Pull Request will be created with the changes and a summary of next steps.
- name: Set micro snapshot version on the release branch
run: |
git checkout -B ${{ github.event.in... | ```suggestion
git push origin ${{ github.event.inputs.releaseBranch }}-put-back-999-snapshot
``` |
timefold-solver | github_2023 | others | 258 | TimefoldAI | rsynek | @@ -86,25 +91,13 @@ jobs:
out/jreleaser/trace.log
out/jreleaser/output.properties
+ # Pull Request will be created with the changes and a summary of next steps.
- name: Set micro snapshot version on the release branch
run: |
git checkout -B ${{ github.event.in... | ```suggestion
gh pr create --reviewer triceo --base ${{ github.event.inputs.releaseBranch }} --head ${{ github.event.inputs.releaseBranch }}-put-back-999-snapshot --title "build: move back to 999-SNAPSHOT" --body-file .github/workflows/release-pr-body.md
``` |
timefold-solver | github_2023 | java | 257 | TimefoldAI | rsynek | @@ -15,15 +16,27 @@ public final class ReflectionBeanPropertyMemberAccessor extends AbstractMemberAc
private final Class<?> propertyType;
private final String propertyName;
private final Method getterMethod;
+ private final MethodHandle getherMethodHandle;
private final Method setterMethod;
+ ... | There is also `unreflectGetter` and `unreflectSetter`, but these operate on `Field` instances, while here we already have the method. |
timefold-solver | github_2023 | java | 254 | TimefoldAI | triceo | @@ -42,6 +43,7 @@ public MultipleDelegateList(List<T>... delegates) {
}
this.totalSize = sizeSoFar;
+ this.cachedRebaseResult = null; | Maybe add a comment here as to why it's safe to store this here? |
timefold-solver | github_2023 | java | 243 | TimefoldAI | rsynek | @@ -525,15 +529,56 @@ private GeneratedGizmoClasses generateDomainAccessors(SolverConfig solverConfig,
Set<String> generatedMemberAccessorsClassNameSet = new HashSet<>();
Set<String> gizmoSolutionClonerClassNameSet = new HashSet<>();
+ /*
+ * TODO consistently change the name "entity"... | Nice; sometimes I forget we are already past Java 15. |
timefold-solver | github_2023 | java | 243 | TimefoldAI | rsynek | @@ -525,15 +529,56 @@ private GeneratedGizmoClasses generateDomainAccessors(SolverConfig solverConfig,
Set<String> generatedMemberAccessorsClassNameSet = new HashSet<>();
Set<String> gizmoSolutionClonerClassNameSet = new HashSet<>();
+ /*
+ * TODO consistently change the name "entity"... | What exactly does it enhance? Only things that may change during solving (solution, genuine and shadow entities) or even facts?
Trying to figure out what the name should be. |
timefold-solver | github_2023 | others | 240 | TimefoldAI | rsynek | @@ -329,17 +334,6 @@
<newArtifacts>
<artifact>${project.groupId}:${project.artifactId}:${project.version}</artifact>
</newArtifacts>
- <!-- By default revapi will check the oldArtifact against the currently executed build -->
- <analysisConfigurationFiles>
... | Is this configuration no longer relevant? |
timefold-solver | github_2023 | java | 240 | TimefoldAI | rsynek | @@ -178,7 +178,6 @@ private boolean isScoreCalculationDefined(SolverConfig solverConfig) {
}
return scoreDirectorFactoryConfig.getEasyScoreCalculatorClass() != null ||
scoreDirectorFactoryConfig.getIncrementalScoreCalculatorClass() != null ||
- scoreDirectorFactoryConfi... | Good catch |
timefold-solver | github_2023 | others | 240 | TimefoldAI | rsynek | @@ -72,6 +72,12 @@
<plugin>
<groupId>org.revapi</groupId>
<artifactId>revapi-maven-plugin</artifactId>
+ <configuration>
+ <analysisConfigurationFiles>
+ <file>${project.basedir}/src/build/revapi-filter.json</file>
+ <file>${project.basedir}/src/build/revap... | Why not define this config in the build parent pom.xml? |
timefold-solver | github_2023 | javascript | 216 | TimefoldAI | rsynek | @@ -26,30 +26,69 @@ function replaceTimefoldAutoHeaderFooter() {
<div class="me-auto"><a class="text-white" href="https://timefold.ai/product/support/">Support</a></div>
</div>
</div>
+ <div id="applicationInfo" class="container text-center"></div>
</footer>... | Maybe remove the prefix, see https://github.com/TimefoldAI/timefold-platform/pull/61#discussion_r1294405188 |
timefold-solver | github_2023 | others | 206 | TimefoldAI | rsynek | @@ -81,4 +81,42 @@
</plugin>
</plugins>
</build>
+
+ <profiles>
+ <profile>
+ <!--
+ Benchmarker is not supported in native.
+ However, Quarkus 3.2.4+ seems to have an issue where the native profile,
+ if present on another unrelated module, is magically and wrongly inherite... | Good as a temporary workaround, but please at least file an issue to investigate this further. Ideally, this should be reported to Quarkus.
If this module is not supposed to work in native, we should not lose time by native compilation in every CI run. |
timefold-solver | github_2023 | others | 193 | TimefoldAI | rsynek | @@ -23,7 +23,7 @@
<!--</appender>-->
<!-- To override the debug log level from the command line, use the VM option "-Dlogback.level.ai.timefold.solver=trace" -->
- <logger name="ai.timefold.solver" level="${logback.level.ai.timefold.solver:-debug}"/>
+ <logger name="ai.timefold.solver" level="${logback.level.... | Intentional? |
timefold-solver | github_2023 | java | 193 | TimefoldAI | rsynek | @@ -96,7 +96,22 @@ public T next() {
@Override
public String toString() {
- return "size = " + size;
+ switch (size) {
+ case 0 -> {
+ return "[]";
+ }
+ case 1 -> {
+ return "[" + first.getElement() + "]";
+ }
+ ... | What is the typical and maximal length of this list? |
timefold-solver | github_2023 | others | 193 | TimefoldAI | rsynek | @@ -458,6 +458,55 @@ stream), use <<constraintStreamsConditionalPropagation,ifExists>> instead.
It does not create cartesian products and therefore generally performs better.
====
+==== Evaluation of multiple joiners
+
+When using multiple joiners, there are some important considerations to keep in mind:
+
+[source... | ```suggestion
However, both of the key mapping functions are calculated independently of the other,
``` |
timefold-solver | github_2023 | others | 193 | TimefoldAI | rsynek | @@ -458,6 +458,55 @@ stream), use <<constraintStreamsConditionalPropagation,ifExists>> instead.
It does not create cartesian products and therefore generally performs better.
====
+==== Evaluation of multiple joiners
+
+When using multiple joiners, there are some important considerations to keep in mind:
+
+[source... | Consistency:
```suggestion
When indexing joiners (such as `equal()` and `lessThan()`) check their indexes,
``` |
timefold-solver | github_2023 | java | 193 | TimefoldAI | rsynek | @@ -0,0 +1,216 @@
+package ai.timefold.solver.constraint.streams.bavet.common;
+
+import java.util.ArrayList;
+import java.util.BitSet;
+import java.util.List;
+import java.util.function.Consumer;
+
+import ai.timefold.solver.constraint.streams.bavet.common.AbstractPropagationMetadataCarrier.PropagationType;
+import ai... | I think this class (and the Static one, too) is complex enough to deserve unit tests. |
timefold-solver | github_2023 | java | 193 | TimefoldAI | rsynek | @@ -0,0 +1,93 @@
+package ai.timefold.solver.constraint.streams.bavet.common;
+
+import ai.timefold.solver.constraint.streams.bavet.common.tuple.AbstractTuple;
+import ai.timefold.solver.constraint.streams.bavet.common.tuple.TupleState;
+
+final class Group<OutTuple_ extends AbstractTuple, ResultContainer_>
+ ex... | Perhaps `GroupDataWithKeyAndAccumulate` is more descriptive? |
timefold-solver | github_2023 | java | 193 | TimefoldAI | rsynek | @@ -0,0 +1,361 @@
+package ai.timefold.solver.constraint.streams.bavet;
+
+import static ai.timefold.solver.core.api.score.stream.Joiners.filtering;
+
+import ai.timefold.solver.constraint.streams.common.AbstractConstraintStreamTest;
+import ai.timefold.solver.core.api.score.buildin.simple.SimpleScore;
+import ai.timef... | Does this note still apply? |
timefold-solver | github_2023 | java | 193 | TimefoldAI | rsynek | @@ -181,82 +184,74 @@ public void update(InTuple_ tuple) {
undoAccumulator.run();
}
- GroupKey_ oldUserSuppliedGroupKey = extractUserSuppliedKey(oldGroup.groupKey);
- GroupKey_ newUserSuppliedGroupKey = hasMultipleGroups ? groupKeyFunction.apply(tuple) : null;
+ var oldUserS... | Don't we need to set the tuple state to UPDATING anymore? |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.