jpa-patterns
Spring Boot 的 JPA/Hibernate 寫法，涵蓋 entity、N+1、交易、分頁跟連線池
中文翻譯
---
name: jpa-patterns
description: Spring Boot 中的 JPA/Hibernate 模式，涵蓋 entity 設計、關聯、查詢最佳化、交易、稽核、索引、分頁和連線池。設計 JPA entity 或關聯，或需要修正 Hibernate 查詢、交易或 N+1 問題時使用。
metadata:
origin: ECC
---
# JPA/Hibernate Patterns
用於 Spring Boot 中的資料建模、repository 和效能調校。
## When to Activate
- 設計 JPA entity 與資料表對應
- 定義關聯（@OneToMany、@ManyToOne、@ManyToMany）
- 最佳化查詢（避免 N+1、fetch 策略、projection）
- 設定交易、稽核或軟刪除
- 設定分頁、排序或自訂 repository 方法
- 調校連線池（HikariCP）或二級快取
## Entity Design
```java
@Entity
@Table(name = "markets", indexes = {
@Index(name = "idx_markets_slug", columnList = "slug", unique = true)
})
@EntityListeners(AuditingEntityListener.class)
public class MarketEntity {
@Id @GeneratedValue(strategy = GenerationType.IDENTITY)
private Long id;
@Column(nullable = false, length = 200)
private String name;
@Column(nullable = false, unique = true, length = 120)
private String slug;
@Enumerated(EnumType.STRING)
private MarketStatus status = MarketStatus.ACTIVE;
@CreatedDate private Instant createdAt;
@LastModifiedDate private Instant updatedAt;
}
```
啟用稽核：
```java
@Configuration
@EnableJpaAuditing
class JpaConfig {}
```
## Relationships and N+1 Prevention
```java
@OneToMany(mappedBy = "market", cascade = CascadeType.ALL, orphanRemoval = true)
private List<PositionEntity> positions = new ArrayList<>();
```
- 預設用 lazy loading；需要時在查詢裡用 `JOIN FETCH`
- 集合上避免用 `EAGER`；讀取路徑用 DTO projection
```java
@Query("select m from MarketEntity m left join fetch m.positions where m.id = :id")
Optional<MarketEntity> findWithPositions(@Param("id") Long id);
```
## Repository Patterns
```java
public interface MarketRepository extends JpaRepository<MarketEntity, Long> {
Optional<MarketEntity> findBySlug(String slug);
@Query("select m from MarketEntity m where m.status = :status")
Page<MarketEntity> findByStatus(@Param("status") MarketStatus status, Pageable pageable);
}
```
- 輕量查詢用 projection：
```java
public interface MarketSummary {
Long getId();
String getName();
MarketStatus getStatus();
}
Page<MarketSummary> findAllBy(Pageable pageable);
```
## Transactions
- service 方法加上 `@Transactional`
- 讀取路徑用 `@Transactional(readOnly = true)` 來最佳化
- 謹慎選擇 propagation；避免長時間執行的交易
```java
@Transactional
public Market updateStatus(Long id, MarketStatus status) {
MarketEntity entity = repo.findById(id)
.orElseThrow(() -> new EntityNotFoundException("Market"));
entity.setStatus(status);
return Market.from(entity);
}
```
## Pagination
```java
PageRequest page = PageRequest.of(pageNumber, pageSize, Sort.by("createdAt").descending());
Page<MarketEntity> markets = repo.findByStatus(MarketStatus.ACTIVE, page);
```
類似 cursor 的分頁，要在 JPQL 裡加上 `id > :lastId` 並搭配排序。
## Indexing and Performance
- 為常用的過濾條件加索引（`status`、`slug`、外鍵）
- 使用符合查詢模式的複合索引（`status, created_at`）
- 避免 `select *`；只 project 需要的欄位
- 用 `saveAll` 和 `hibernate.jdbc.batch_size` 做批次寫入
## Connection Pooling (HikariCP)
建議的屬性：
```
spring.datasource.hikari.maximum-pool-size=20
spring.datasource.hikari.minimum-idle=5
spring.datasource.hikari.connection-timeout=30000
spring.datasource.hikari.validation-timeout=5000
```
PostgreSQL 處理 LOB 時，要加上：
```
spring.jpa.properties.hibernate.jdbc.lob.non_contextual_creation=true
```
## Caching
- 一級快取是每個 EntityManager 各自一份；避免讓 entity 跨交易存活
- 讀取量大的 entity，可以謹慎考慮二級快取；要驗證驅逐策略
## Migrations
- 用 Flyway 或 Liquibase；正式環境絕對不要依賴 Hibernate 的自動 DDL
- migration 要保持冪等且只加不減；沒有計畫就不要刪欄位
## Testing Data Access
- 優先用 `@DataJpaTest` 搭配 Testcontainers 來貼近正式環境
- 用 log 斷言 SQL 效率：設定 `logging.level.org.hibernate.SQL=DEBUG` 和 `logging.level.org.hibernate.orm.jdbc.bind=TRACE` 來看參數值
**記住**：entity 要精簡、查詢要有意圖、交易要短。用 fetch 策略和 projection 避免 N+1，並針對你的讀寫路徑建立索引。英文原文
---
name: jpa-patterns
description: JPA/Hibernate patterns for entity design, relationships, query optimization, transactions, auditing, indexing, pagination, and pooling in Spring Boot. Use when designing JPA entities or relationships, or when a Hibernate query, transaction, or N+1 problem needs fixing.
metadata:
origin: ECC
---
# JPA/Hibernate Patterns
Use for data modeling, repositories, and performance tuning in Spring Boot.
## When to Activate
- Designing JPA entities and table mappings
- Defining relationships (@OneToMany, @ManyToOne, @ManyToMany)
- Optimizing queries (N+1 prevention, fetch strategies, projections)
- Configuring transactions, auditing, or soft deletes
- Setting up pagination, sorting, or custom repository methods
- Tuning connection pooling (HikariCP) or second-level caching
## Entity Design
```java
@Entity
@Table(name = "markets", indexes = {
@Index(name = "idx_markets_slug", columnList = "slug", unique = true)
})
@EntityListeners(AuditingEntityListener.class)
public class MarketEntity {
@Id @GeneratedValue(strategy = GenerationType.IDENTITY)
private Long id;
@Column(nullable = false, length = 200)
private String name;
@Column(nullable = false, unique = true, length = 120)
private String slug;
@Enumerated(EnumType.STRING)
private MarketStatus status = MarketStatus.ACTIVE;
@CreatedDate private Instant createdAt;
@LastModifiedDate private Instant updatedAt;
}
```
Enable auditing:
```java
@Configuration
@EnableJpaAuditing
class JpaConfig {}
```
## Relationships and N+1 Prevention
```java
@OneToMany(mappedBy = "market", cascade = CascadeType.ALL, orphanRemoval = true)
private List<PositionEntity> positions = new ArrayList<>();
```
- Default to lazy loading; use `JOIN FETCH` in queries when needed
- Avoid `EAGER` on collections; use DTO projections for read paths
```java
@Query("select m from MarketEntity m left join fetch m.positions where m.id = :id")
Optional<MarketEntity> findWithPositions(@Param("id") Long id);
```
## Repository Patterns
```java
public interface MarketRepository extends JpaRepository<MarketEntity, Long> {
Optional<MarketEntity> findBySlug(String slug);
@Query("select m from MarketEntity m where m.status = :status")
Page<MarketEntity> findByStatus(@Param("status") MarketStatus status, Pageable pageable);
}
```
- Use projections for lightweight queries:
```java
public interface MarketSummary {
Long getId();
String getName();
MarketStatus getStatus();
}
Page<MarketSummary> findAllBy(Pageable pageable);
```
## Transactions
- Annotate service methods with `@Transactional`
- Use `@Transactional(readOnly = true)` for read paths to optimize
- Choose propagation carefully; avoid long-running transactions
```java
@Transactional
public Market updateStatus(Long id, MarketStatus status) {
MarketEntity entity = repo.findById(id)
.orElseThrow(() -> new EntityNotFoundException("Market"));
entity.setStatus(status);
return Market.from(entity);
}
```
## Pagination
```java
PageRequest page = PageRequest.of(pageNumber, pageSize, Sort.by("createdAt").descending());
Page<MarketEntity> markets = repo.findByStatus(MarketStatus.ACTIVE, page);
```
For cursor-like pagination, include `id > :lastId` in JPQL with ordering.
## Indexing and Performance
- Add indexes for common filters (`status`, `slug`, foreign keys)
- Use composite indexes matching query patterns (`status, created_at`)
- Avoid `select *`; project only needed columns
- Batch writes with `saveAll` and `hibernate.jdbc.batch_size`
## Connection Pooling (HikariCP)
Recommended properties:
```
spring.datasource.hikari.maximum-pool-size=20
spring.datasource.hikari.minimum-idle=5
spring.datasource.hikari.connection-timeout=30000
spring.datasource.hikari.validation-timeout=5000
```
For PostgreSQL LOB handling, add:
```
spring.jpa.properties.hibernate.jdbc.lob.non_contextual_creation=true
```
## Caching
- 1st-level cache is per EntityManager; avoid keeping entities across transactions
- For read-heavy entities, consider second-level cache cautiously; validate eviction strategy
## Migrations
- Use Flyway or Liquibase; never rely on Hibernate auto DDL in production
- Keep migrations idempotent and additive; avoid dropping columns without plan
## Testing Data Access
- Prefer `@DataJpaTest` with Testcontainers to mirror production
- Assert SQL efficiency using logs: set `logging.level.org.hibernate.SQL=DEBUG` and `logging.level.org.hibernate.orm.jdbc.bind=TRACE` for parameter values
**Remember**: Keep entities lean, queries intentional, and transactions short. Prevent N+1 with fetch strategies and projections, and index for your read/write paths.