jpa-patterns_skill

Overview

This skill codifies battle-tested JPA/Hibernate patterns for Spring Boot applications, covering entity design, relationships, query optimization, transactions, auditing, indexing, pagination, and connection pooling. It focuses on practical, production-ready guidance to build efficient, maintainable data layers and avoid common pitfalls like N+1 queries and incorrect transaction scopes.

How this skill works

The skill inspects typical data-access concerns and prescribes concrete patterns: annotated entity examples, relationship mappings, repository query styles, and configuration snippets for auditing and HikariCP. It highlights when to use fetch joins, DTO projections, pagination strategies, batching, and second-level caching while advising migration and testing practices. Each recommendation maps to simple code fragments or configuration properties you can apply directly in a Spring Boot project.

When to use it

• Designing JPA entities and table mappings with indexes and auditing
• Defining and tuning relationships (OneToMany, ManyToOne, ManyToMany) to prevent N+1
• Optimizing read paths with projections, JOIN FETCH, and pagination
• Configuring transactions, propagation, and read-only boundaries
• Tuning connection pooling (HikariCP), batching, and second-level cache usage
• Setting up migrations and integration tests with Testcontainers

Best practices

• Default collections to LAZY and avoid EAGER on collections; use JOIN FETCH for specific queries
• Prefer DTO projections or interface projections for read-heavy endpoints to reduce selected columns
• Annotate service methods with @Transactional; use readOnly=true for queries and keep transactions short
• Add indexes for common filters and composite indexes that match query patterns
• Batch writes with saveAll and set hibernate.jdbc.batch_size; monitor SQL via Hibernate logs during testing
• Use Flyway/Liquibase for migrations; avoid Hibernate auto-DDL in production

Example use cases

• Fetch a Market with its Positions using a JOIN FETCH query to prevent N+1
• Expose a lightweight MarketSummary projection for paginated dashboards
• Implement soft deletes and auditing with @CreatedDate/@LastModifiedDate and @EnableJpaAuditing
• Configure HikariCP properties for stable connection pooling in production
• Write integration data tests with @DataJpaTest plus Testcontainers to validate SQL and migrations