src/main/java/org/gridsuite/modification/server/entities/equipment/creation/LineCreationEntity.java (1)

44-49: ⚠️ Potential issue | 🟠 Major

Persist lineSegments order.

This is still mapped as an ordered List, but the association does not store any index. After a save/reload cycle, segment order is unspecified and the reconstructed LineCreationInfos can change. Add an @OrderColumn here and the matching column in line_creation_line_segments.

Suggested change

     `@OneToMany`(cascade = CascadeType.ALL, orphanRemoval = true)
+    `@OrderColumn`(name = "segment_index")
     `@JoinTable`(
         joinColumns = `@JoinColumn`(name = "line_id"), foreignKey = `@ForeignKey`(name = "line_id_fk"),
         inverseJoinColumns = `@JoinColumn`(name = "line_segments_id"), inverseForeignKey = `@ForeignKey`(name = "line_segments_id_fk"),
         uniqueConstraints = `@UniqueConstraint`(name = "line_creation_line_segments_uk", columnNames = {"line_segments_id"}))
     private List<LineSegmentEntity> lineSegments;

🤖 Prompt for AI Agents

Verify each finding against the current code and only fix it if needed.

In
`@src/main/java/org/gridsuite/modification/server/entities/equipment/creation/LineCreationEntity.java`
around lines 44 - 49, The lineSegments List in LineCreationEntity is not
persisted with an order index; add `@OrderColumn` to the lineSegments field (e.g.,
`@OrderColumn`(name = "segment_order")) and update the join table definition for
line_creation_line_segments to include the matching index column (segment_order)
so JPA will persist and restore the List element order; ensure the column is
included in the JoinTable DDL/migration and name matches the `@OrderColumn` value.

src/main/resources/db/changelog/changesets/changelog_20260324T101024Z.xml (1)

4-11: Index line_creation_line_segments.line_id.

LineCreationEntity.lineSegments will be read and cleaned up through line_id, but the schema only gets an index on line_segments_id via the unique constraint. Add a parent-side index here to avoid full scans and expensive FK checks once this join table grows.

Suggested change

     <changeSet author="elcheikhbas (generated)" id="1774347058503-15">
         <createTable tableName="line_creation_line_segments">
             <column name="line_id" type="UUID">
                 <constraints nullable="false"/>
             </column>
             <column name="line_segments_id" type="UUID">
                 <constraints nullable="false"/>
             </column>
         </createTable>
+        <createIndex indexName="line_creation_line_segments_line_id_idx"
+                     tableName="line_creation_line_segments">
+            <column name="line_id"/>
+        </createIndex>
     </changeSet>

🤖 Prompt for AI Agents

Verify each finding against the current code and only fix it if needed.

In `@src/main/resources/db/changelog/changesets/changelog_20260324T101024Z.xml`
around lines 4 - 11, Add a non-unique index on the parent column to avoid full
scans: modify the changeset that creates the join table
line_creation_line_segments to add an index for column "line_id" (in addition to
the existing unique constraint on "line_segments_id"); update the XML to include
an <createIndex> (or equivalent index element) targeting
tableName="line_creation_line_segments" and column name="line_id" so lookups
from LineCreationEntity.lineSegments use the indexed parent key.

src/test/java/org/gridsuite/modification/server/modifications/LineCreationInNodeBreakerTest.java (1)

324-325: Consider extracting a shared line-segment test fixture.

The same lineSegments(List.of(...)) literal now appears across multiple integration tests; a shared helper/fixture would reduce drift when the segment DTO evolves.

🤖 Prompt for AI Agents

Verify each finding against the current code and only fix it if needed.

In
`@src/test/java/org/gridsuite/modification/server/modifications/LineCreationInNodeBreakerTest.java`
around lines 324 - 325, The repeated inline List.of(new LineSegmentInfos(...))
literal across tests (seen in LineCreationInNodeBreakerTest) should be extracted
into a shared test fixture helper; add a static factory method (e.g.,
TestFixtures.defaultLineSegments() or a helper in LineCreationInNodeBreakerTest
like createDefaultLineSegments()) that returns the List<LineSegmentInfos> with
the two populated segments, replace all inline occurrences with calls to that
helper, and update imports/usages so tests reference the single source of truth
(class name references: LineCreationInNodeBreakerTest, LineSegmentInfos, and any
other test classes using the literal).

src/test/java/org/gridsuite/modification/server/modifications/LineCreationInBusBreakerTest.java (1)

255-256: Use deterministic IDs in test fixtures

Using UUID.randomUUID() in test payloads makes failures less reproducible and fixture diffs noisier. Prefer stable, explicit identifiers for segmentTypeId in tests.

♻️ Proposed change

-            .lineSegments(List.of(new LineSegmentInfos(UUID.randomUUID().toString(), 1, "1", "50", null),
-                new LineSegmentInfos(UUID.randomUUID().toString(), 1, "1", null, 0.95)))
+            .lineSegments(List.of(new LineSegmentInfos("segmentType-1", 1, "1", "50", null),
+                new LineSegmentInfos("segmentType-2", 1, "1", null, 0.95)))

🤖 Prompt for AI Agents

Verify each finding against the current code and only fix it if needed.

In
`@src/test/java/org/gridsuite/modification/server/modifications/LineCreationInBusBreakerTest.java`
around lines 255 - 256, Tests use UUID.randomUUID() for
LineSegmentInfos.segmentTypeId which makes fixtures non-deterministic; replace
those calls in the test (LineCreationInBusBreakerTest) with stable, explicit IDs
(e.g., "segment-1", "segment-2") when constructing LineSegmentInfos so the
payloads are reproducible and diffs are stable.