Dynamosa Algorithm by francastagna · Pull Request #1399 · WebFuzzing/EvoMaster

francastagna · 2025-12-02T19:06:04Z

PR Summary

We implemented the DynaMOSA many‑objective search algorithm, wired it into EvoMaster core, and added a graph‑based instrumentation pipeline that builds control dependence graphs (CDGs) and exposes them to the algorithm via a lightweight DTO and remote protocol.

Pseudocode:

Source: https://www.researchgate.net/publication/313164878_Automated_Test_Case_Generation_as_a_Many-Objective_Optimisation_Problem_with_Dynamic_Selection_of_the_Targets

Main Changes

New DynaMOSA algorithm (DynaMosaAlgorithm)
- Adds a many‑objective search algorithm that uses a dynamic focus set of targets per generation instead of optimizing all objectives at once.
- Periodically pulls new CDGs from the driver and updates its focus set based on uncovered targets.
Branch dependency graph and multi‑criteria manager
- BranchDependencyGraph builds a lightweight DAG of branch objectives from ControlDependenceGraphDto (roots, children, and all objectives).
- MulticriteriaManager keeps track of covered/uncovered branch targets and computes the current focus set by combining CDG roots with uncovered children; falls back to all uncovered targets when needed.
Instrumentation: CFG/CDG construction and export
- Extends the Java agent with CFG/CDG support (EvoMasterGraph, CFG and CDG classes, GraphPool) that computes per‑method CDGs from bytecode.
- Introduces ControlDependenceGraphDto (objectives, root ids, edges) as the minimal, serialized view of a CDG sent from the agent to core.
- Adds DynamosaConfig flags and controller hooks (InstrumentationController, AgentController) to enable/disable graph construction and to stream CDG snapshots incrementally.
Core / driver wiring
- Updates EMConfig and Main so Algorithm.DYNAMOSA is available and correctly instantiated for REST, GraphQL, RPC, and web‑frontend problems.
- Extends the Java driver controllers and remote API to:
  - Start a new search (resetting coverage and CDG export state on the agent).
  - Push DynaMOSA‑specific configuration (graph flags) and pull batches of ControlDependenceGraphDto objects used to maintain the branch dependency graph in core.

arcuri82 · 2025-12-10T08:55:59Z

@francastagna unfortunately due to sickness, had to skip some work this week. and now rest of the week i ll be fully in meetings. i doubt i ll have any time to look at this PR this week.
@jgaleotti have you already given feedback to @francastagna on the work for this PR?

arcuri82

hi @francastagna @jgaleotti thanks for this PR. it is more than 8k LOCs. unfortunately, instrumentation is a critical component of EM, and I would have to look into this PR in every details. i started, but this clearly would take me at least a day dedicated to it. not sure when i ll have a time for that. you can start to address the comments I gave so far, meanwhile

arcuri82 · 2025-12-15T09:39:30Z

...ava/controller-api/src/main/java/org/evomaster/client/java/controller/api/dto/SutRunDto.java

+    /**
+     * Whether to enable Dynamosa graphs generation in the Java agent
+     */
+    public Boolean enableDynamosaGraphs;


instrumentation is independent from search algorithm. in other words, there should be no reference to Dynamosa here.

arcuri82 · 2025-12-15T09:40:24Z

...ava/controller-api/src/main/java/org/evomaster/client/java/controller/api/dto/SutRunDto.java

+    /**
+     * Whether to write generated graphs (DOT/PNGs) to disk on the agent side
+     */
+    public Boolean writeCfg;


is this only for debugging purposes? if so, clarify it in the comment. also, i guess this would not work when running expeirments in parallel, as all will overwrite same file?

Yes, only for debugging porpuses.
Probably won't work when running experiments in parallel you are right.

arcuri82 · 2025-12-15T09:41:23Z

...ontroller-api/src/main/java/org/evomaster/client/java/controller/api/dto/TestResultsDto.java

+    /**
+     * Incremental DynaMOSA control-dependence graphs discovered since the last handshake.
+     */
+    public List<ControlDependenceGraphDto> dynamosaCdgs = new ArrayList<>();


no ref to Dynamosa here

Do you mean I should rename the variable to something like 'cdgs' to avoid talking about the algorithm running in the core?

controlDependenceGraph. and yes, there should be no reference to dynamosa under client-java

acrnoyms like cfg are fine inside short-lived code blocks, but not for class fields

arcuri82 · 2025-12-15T09:47:09Z

client-java/controller-api/pom.xml

+        <dependency>
+            <groupId>org.evomaster</groupId>
+            <artifactId>evomaster-client-java-instrumentation-shared</artifactId>
+        </dependency>


remove. api module with DTOs should not have this kind of dependencies

arcuri82 · 2025-12-15T09:48:26Z

...ain/java/org/evomaster/client/java/instrumentation/shared/dto/ControlDependenceGraphDto.java

+ * Contains only the data needed by EvoMaster core: branch objective identifiers,
+ * the subset that are roots, and the parent→child relationships between them.
+ */
+public class ControlDependenceGraphDto implements Serializable {


if it is under instrumentation-shared, then it is not a DTOs. also, here there is plenty of logic that is part of a DTO. a DTO is a POJO, with just public fields.

arcuri82 · 2025-12-15T11:20:30Z

...tation/src/main/java/org/evomaster/client/java/instrumentation/graphs/cdg/DominatorTree.java

+/*
+ * Adapted from the EvoSuite project (https://github.com/EvoSuite/evosuite)
+ * and modified for use in EvoMaster's Dynamosa module.
+ */


in this case, docs/reused_code.md needs to be updated

arcuri82 · 2025-12-15T11:22:38Z

...c/main/java/org/evomaster/client/java/instrumentation/dynamosa/graphs/cdg/DominatorTree.java

+    public DominatorTree(ControlFlowGraph<V> cfg) {
+        super(DefaultEdge.class);
+
+        SimpleLogger.debug("Computing DominatorTree for " + cfg.getName());


no concatentation in debug logs. this applies to all instances in this PR. i ll not add more comments about it

arcuri82 · 2025-12-15T11:23:01Z

...c/main/java/org/evomaster/client/java/instrumentation/dynamosa/graphs/cdg/DominatorTree.java

+
+        computeDominatorFrontiers(rootNode);
+
+        //		toDot();


My bad, I'll remove this

arcuri82 · 2025-12-15T11:23:13Z

...c/main/java/org/evomaster/client/java/instrumentation/dynamosa/graphs/cdg/DominatorTree.java

+
+        createDominatorNodes();
+
+        V root = cfg.determineEntryPoint(); // TODO change to getEntryPoint()


is this TODO still relevant?

It's not, i'll remove it.

arcuri82 · 2025-12-15T11:24:51Z

...tation/src/main/java/org/evomaster/client/java/instrumentation/graphs/cdg/DominatorTree.java

+
+    private void computeSemiDominators() {
+
+        for (int i = nodeCount; i >= 2; i--) {


why i >= 2? add some explanation

We skip node 1 (i >= 2) because node 1 is the root/entry point.
I'll add a comment here.

francastagna · 2025-12-15T13:26:35Z

hi @francastagna @jgaleotti thanks for this PR. it is more than 8k LOCs. unfortunately, instrumentation is a critical component of EM, and I would have to look into this PR in every details. i started, but this clearly would take me at least a day dedicated to it. not sure when i ll have a time for that. you can start to address the comments I gave so far, meanwhile

Hi @arcuri82, thanks for taking a look at the PR and for the initial feedback. I’ll start addressing the comments you’ve provided so far.

jgaleotti

Please @francastagna address these changes plus those asked by @arcuri82

jgaleotti · 2026-01-14T17:25:43Z

...on/src/main/java/org/evomaster/client/java/instrumentation/graphs/cfg/branch/BranchPool.java

+            pool.branchCounter = 0;
+        }
+    }
+    // fill the pool


In this code incomplete?

No, it was an old comment, I'll remove it!

jgaleotti · 2026-01-14T17:27:15Z

...main/java/org/evomaster/client/java/instrumentation/graphs/ControlDependenceGraphConfig.java

+
+
+    public static boolean isGraphsEnabled() {
+        System.out.println("enableGraphs: " + enableGraphs);


why System.out is used? Shouldn't you be using the logging infraestructure for this?

jgaleotti · 2026-01-14T17:28:03Z

...mentation/src/main/java/org/evomaster/client/java/instrumentation/graphs/EvoMasterGraph.java

+            }
+        } catch (IOException e) {
+            // TODO Auto-generated catch block
+            e.printStackTrace();


This doesn't look correct, missing the exception here.

jgaleotti · 2026-01-14T17:31:34Z

client-java/instrumentation/pom.xml

        </dependency>
+        <dependency>
+            <groupId>org.evomaster</groupId>
+            <artifactId>evomaster-client-java-controller-api</artifactId>


is this new dependency necessary? why?

Yes! The instrumentation module uses DTOs from controller-api to exchange control dependence graph information with the controller.

Specifically:
GraphPool.java (lines 17-19) imports ControlDependenceGraphDto, BranchObjectiveDto, and DependencyEdgeDto

These DTOs are used in GraphPool.buildControlDependenceDto() (lines 344-418) to build the graph representation that gets exported to the controller

jgaleotti · 2026-01-14T17:36:46Z

core/src/main/kotlin/org/evomaster/core/EMConfig.kt

+     * This controls only the persistence of graphs to disk; graph creation is controlled separately.
+     */
+    @Cfg("Enable writing CFG/CDG graphs to disk on the agent side")
+    var writeCfg: Boolean = true


This feature should be experimental and false by default

jgaleotti · 2026-01-14T17:37:53Z

docs/options.md

 |`useResponseDataPool`| __Boolean__. Enable the collection of response data, to feed new individuals based on field names matching. *Default value*: `true`.|
 |`useTimeInFeedbackSampling`| __Boolean__. Whether to use timestamp info on the execution time of the tests for sampling (e.g., to reward the quickest ones). *Default value*: `true`.|
 |`weightBasedMutationRate`| __Boolean__. Whether to enable a weight-based mutation rate. *Default value*: `true`.|
+|`writeCfg`| __Boolean__. Enable writing CFG/CDG graphs to disk on the agent side. *Default value*: `true`.|


default value should be false as this feature should only be turned on when using DYNAMOSA

…ntrolDependenceGraphConfig

jgaleotti

No further changes from my side. @arcuri82 can proceed with his own review

francastagna added 16 commits November 20, 2025 08:45

wip

684a8f5

wip: testing graph creation

26aa643

added algoritm logic

18b4f1e

Merge branch 'master' into feature/dynamosa-3

580b45a

moved annotated methods

550f4ba

updated docs/options.md

1a14ed9

updated client-java/pom.xml

1e7634c

fixes

9477a69

refactoring: remove unused class

0782b66

refactoring: cleaning stuff

d29d515

refactoring: cleaning stuff

755bd48

refactoring: cleaning stuff

83149d3

add tests

300ba05

Merge branch 'master' into feature/dynamosa-3

e74328f

enablegraphs and writecfg is false by default

9f184af

fixing stuff

7f63038

francastagna changed the title ~~Feature/dynamosa 3~~ Dynamosa Algorithm Dec 5, 2025

francastagna marked this pull request as ready for review December 5, 2025 15:14

update docs/options.md

38b68f8

francastagna requested a review from arcuri82 December 5, 2025 16:25

arcuri82 requested changes Dec 15, 2025

View reviewed changes

Address Andrea's initial feedback on PR #1399

a44bc04

francastagna force-pushed the feature/dynamosa-3 branch from ab6b218 to a44bc04 Compare December 15, 2025 15:27

francastagna added 2 commits December 15, 2025 12:28

Merge branch 'master' into feature/dynamosa-3

31f7fb8

updating comment

2484d00

francastagna requested a review from arcuri82 December 15, 2025 16:01

jgaleotti requested changes Jan 14, 2026

View reviewed changes

Merge branch 'master' into feature/dynamosa-3

788af33

francastagna added 5 commits January 15, 2026 21:07

change writeCfg config default value to false

d509a3d

improve error handling of IOException in EvoMasterGraph

f725bf9

remove old comment in BranchPool

b539721

replaced all System.out.println() calls with SimpleLogger calls in Co…

6f8ebeb

…ntrolDependenceGraphConfig

Merge branch 'master' into feature/dynamosa-3

cbb9af3

francastagna requested a review from jgaleotti January 16, 2026 16:04

jgaleotti approved these changes Jan 16, 2026

View reviewed changes


		createDominatorNodes();

		V root = cfg.determineEntryPoint(); // TODO change to getEntryPoint()


		private void computeSemiDominators() {

		for (int i = nodeCount; i >= 2; i--) {



		public static boolean isGraphsEnabled() {
		System.out.println("enableGraphs: " + enableGraphs);

Conversation

francastagna commented Dec 2, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

PR Summary

Pseudocode:

Main Changes

Uh oh!

arcuri82 commented Dec 10, 2025

Uh oh!

arcuri82 left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

francastagna commented Dec 15, 2025

Uh oh!

jgaleotti left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

jgaleotti left a comment

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

francastagna commented Dec 2, 2025 •

edited

Loading