persistence-graph-extraction-jpa

Demonstrates how to assemble an object graph from a relational database using JPA entities and Blaze Persistence Entity Views without ORM magic or pagination.

What it shows

• Assembling a three-level object graph (Owner → Pet → Visit) using Blaze Persistence Entity Views
• Using package-private classes as an encapsulation boundary inside the infrastructure layer
• Two query strategies:
  • full graph — Owner with Pets and Visits (OwnerView)
  • flat list — Owner with pet names only (OwnerListView)

Stack

• Java 25
• Spring Boot
• JPA (EntityManager)
• Blaze Persistence (EntityViewManager, CriteriaBuilderFactory)
• H2 (in-memory database)

Structure

application/
    OwnerReadRepository       # port (interface)
    view/                     # read models: OwnerView, OwnerListView, PetView, VisitView

infrastructure/
    BlazeOwnerReadRepository  # Blaze Persistence implementation (package-private)
    OwnerEntity               # @Entity (package-private)
    PetEntity                 # @Entity (package-private)
    VisitEntity               # @Entity (package-private)
    OwnerEntityView           # @EntityView — full graph (package-private)
    PetEntityView             # @EntityView (package-private)
    VisitEntityView           # @EntityView (package-private)
    OwnerListEntityView       # @EntityView — flat list (package-private)
    ViewMapper                # EntityView → View records (package-private)
    BlazeConfig               # CriteriaBuilderFactory + EntityViewManager beans (package-private)

Key idea

Blaze Persistence Entity Views fetch only the data needed — no N+1, no manual JOIN, no deduplication:

Owner(jack)
  └── Pet(buddy1)
        ├── Visit(2026-01-10)
        └── Visit(2026-02-15)
  └── Pet(buddy2)
        └── Visit(2026-03-01)

@EntityView interfaces define the exact shape of the result. Blaze generates optimized queries automatically and maps the result into view instances. ViewMapper converts them into application-level view records.

Compare with the JDBC and jOOQ approaches:

	JDBC	jOOQ	JPA + Blaze
SQL result	flat rows (cartesian JOIN)	nested collections	optimized per-view query
Deduplication	LinkedHashMap in projections	not needed	not needed
Intermediate classes	OwnerProjection, PetProjection, etc.	none	@EntityView interfaces
Mapping	ViewMapper	Records.mapping()	ViewMapper
N+1 problem	no (single JOIN query)	no (single query)	no (Blaze optimizes automatically)

BlazeOwnerReadRepository is package-private by design; only OwnerReadRepository (the interface) is exposed outside the infrastructure package.

Tests

Integration tests in src/test/java cover graph assembly, flat list extraction, and edge cases.

Related

• persistence-graph-extraction-jdbc — same approach with plain JDBC
• persistence-graph-extraction-jooq — same approach with jOOQ MULTISET
• persistence-graph-pagination-jdbc — same approach with pagination support

Run

./mvnw spring-boot:run