Fix @OneToOne composite @JoinColumn handling for @EmbeddedId in JDBC - #3883
Open
radovanradic wants to merge 27 commits into
Open
Fix @OneToOne composite @JoinColumn handling for @EmbeddedId in JDBC#3883radovanradic wants to merge 27 commits into
radovanradic wants to merge 27 commits into
Conversation
…uence identities, including Postgres R2DBC placeholder numbering.
…Ds and add record-based JDBC regression coverage
…duplicate embedded-column regression tests.
Contributor
There was a problem hiding this comment.
Pull request overview
This PR addresses incorrect SQL/DDL generation for owning @OneToOne relationships that use composite @JoinColumns pointing at an associated entity with an @EmbeddedId (issue #3415). The core change ensures explicit join-column names are honored (instead of derived metadata_id_* names) and that shared-identity PK/FK one-to-one mappings don’t cause duplicate/conflicting column mappings in INSERT/UPDATE/DDL.
Changes:
- Resolve explicit
@JoinColumn(name=..., referencedColumnName=...)names during SQL/DDL column name mapping, including for embedded/composite IDs. - Add shared-identity detection to avoid writing shared identity columns via relation paths (UPDATE) and to avoid duplicate/ambiguous physical columns (DDL/INSERT), while still supporting shared-sequence identity placeholder ordering.
- Add JDBC/R2DBC integration tests and processor-level SQL builder tests covering the fixed scenarios and conflict detection.
Reviewed changes
Copilot reviewed 18 out of 18 changed files in this pull request and generated no comments.
Show a summary per file
| File | Description |
|---|---|
| data-r2dbc/src/test/groovy/io/micronaut/data/r2dbc/postgres/one2one/PostgresOneToOneSharedSequenceIdentitySpec.groovy | R2DBC regression test for shared PK one-to-one with sequence identity and correct physical id reuse. |
| data-r2dbc/src/test/groovy/io/micronaut/data/r2dbc/postgres/one2one/PostgresOneToOneEmbeddedIdJoinColumnSpec.groovy | R2DBC regression test for composite join columns targeting an @EmbeddedId. |
| data-processor/src/test/groovy/io/micronaut/data/processor/sql/BuildUpdateSpec.groovy | Processor test ensuring shared-identity join columns aren’t updated via relation path. |
| data-processor/src/test/groovy/io/micronaut/data/processor/sql/BuildTableSpec.groovy | Processor test ensuring DDL doesn’t duplicate embedded-id columns for shared identity joins; plus conflict-case coverage. |
| data-processor/src/test/groovy/io/micronaut/data/processor/sql/BuildQuerySpec.groovy | Processor test for honoring explicit join column names differing from derived names in joins. |
| data-processor/src/test/groovy/io/micronaut/data/processor/sql/BuildInsertSpec.groovy | Processor tests for explicit join column usage in insert/update, shared-identity insert behavior, placeholder contiguity, and fail-fast on conflicting mappings. |
| data-model/src/main/java/io/micronaut/data/model/query/builder/sql/SqlSchemaUtils.java | DDL column emission now detects/handles allowed shared-identity column reuse and fails fast on true conflicts. |
| data-model/src/main/java/io/micronaut/data/model/query/builder/sql/SqlQueryBuilderUtils.java | Adds shared-identity and identity-column detection helpers and a shared path utility. |
| data-model/src/main/java/io/micronaut/data/model/query/builder/sql/SqlQueryBuilder.java | Refactors INSERT generation to handle shared-identity duplicates, sequence literals, and correct positional parameter numbering. |
| data-model/src/main/java/io/micronaut/data/model/query/builder/sql/AbstractSqlLikeQueryBuilder.java | Skips UPDATE SET entries for shared-identity join columns encountered via relation traversal. |
| data-model/src/main/java/io/micronaut/data/model/naming/NamingStrategy.java | Honors explicit join-column name when referencedColumnName matches the mapped/persisted property. |
| data-jdbc/src/test/java/io/micronaut/data/jdbc/h2/one2one/RecordAssetRepository.java | JDBC record-based repository used for embedded-id join-column fetch coverage. |
| data-jdbc/src/test/java/io/micronaut/data/jdbc/h2/one2one/RecordAssetMetadata.java | JDBC record entity for associated side with @EmbeddedId. |
| data-jdbc/src/test/java/io/micronaut/data/jdbc/h2/one2one/RecordAssetId.java | JDBC embeddable record id used in record-based regression test. |
| data-jdbc/src/test/java/io/micronaut/data/jdbc/h2/one2one/RecordAsset.java | JDBC record entity with composite @JoinColumns to associated @EmbeddedId. |
| data-jdbc/src/test/groovy/io/micronaut/data/jdbc/postgres/one2one/PostgresOneToOneSharedSequenceIdentitySpec.groovy | JDBC regression test for shared PK one-to-one with sequence identity on Postgres. |
| data-jdbc/src/test/groovy/io/micronaut/data/jdbc/h2/one2one/OneToOneEmbeddedIdJoinColumnSpec.groovy | JDBC regression test for save/update/fetch join with composite join columns + embedded id. |
| data-jdbc/src/test/groovy/io/micronaut/data/jdbc/h2/one2one/OneToOneEmbeddedIdJoinColumnRecordSpec.groovy | JDBC regression test for fetch join materialization with records + embedded id. |
|
|
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Should fix issue
Adds a focused JDBC regression test for issue #3415 covering an owning ONE_TO_ONE association with two
@JoinColumnstargeting an@EmbeddedId. The test reproduces both failing paths from the report: insert/save and joined fetch.Fixes SQL name resolution so explicit composite
@JoinColumnnames are honored when expanding association backed embedded identifier properties, instead of synthesizing owner-side column names like metadata_container_id / metadata_asset_id. It also adjusts insert SQL generation to de-duplicate physical columns so shared-key one-to-one mappings do not emit the same owner PK columns twice.Hopefully should not introduce regression. Explicit
@JoinColumn/@JoinColumnsnames are now consistently honored for shared/composite identity mappings. If an existing schema relies on previously inferred column names, users can pin those names explicitly with@JoinColumnand@MappedPropertyinstead of changing the database schema.