Micronaut Data JDBC查询Tournament时关联表未加载问题求助

你的问题核心是Micronaut Data JDBC默认不会自动加载关联实体，默认的CrudRepository方法仅查询主表数据。以下是几种可行的解决方式：

方案1：修改关联关系的FetchType为EAGER #

直接将所有需要立即加载的关联字段的fetch属性设置为FetchType.EAGER，强制查询时加载关联数据：

修改TournamentEntity：

@MappedEntity(value = "tournament")
data class TournamentEntity(
    @field:Id
    @GeneratedValue
    val id: Int? = null,

    @JoinTable(name = "tournament_player",
        joinColumns = [JoinColumn(name = "tournament_id", referencedColumnName = "id")],
        inverseJoinColumns = [JoinColumn(name = "player_id", referencedColumnName = "id")]
    )
    @ManyToMany(cascade = [CascadeType.ALL], fetch = FetchType.EAGER)
    val players: List<PlayerEntity>,

    @OneToMany(mappedBy = "tournament", orphanRemoval = true, cascade = [CascadeType.ALL], fetch = FetchType.EAGER)
    val tournamentRounds: List<TournamentRoundEntity>,
)

修改TournamentRoundEntity的matches字段：

@OneToMany(mappedBy = "tournamentRound", orphanRemoval = true, cascade = [CascadeType.ALL], fetch = FetchType.EAGER)
val matches: List<MatchEntity>,

修改MatchEntity的选手关联字段：

@OneToOne(fetch = FetchType.EAGER)
@JoinColumn(name = "player_a_id", nullable = false, columnDefinition = "int")
val playerA: PlayerEntity,

@OneToOne(fetch = FetchType.EAGER)
@JoinColumn(name = "player_b_id", nullable = false, columnDefinition = "int")
val playerB: PlayerEntity,

注意：此方式会每次查询主实体时都加载所有关联数据，适合数据量较小的场景，避免大数量级下的性能问题。

方案2：自定义仓库查询，使用JOIN FETCH显式加载 #

通过自定义JPQL查询，使用JOIN FETCH一次性加载所有层级的关联数据，避免N+1查询：

@JdbcRepository(dialect = Dialect.MYSQL)
@Transactional(Transactional.TxType.MANDATORY)
interface TournamentRepository : CrudRepository<TournamentEntity, Int>, JpaSpecificationExecutor<TournamentEntity> {

    @Query("SELECT t FROM TournamentEntity t JOIN FETCH t.players JOIN FETCH t.tournamentRounds tr JOIN FETCH tr.matches m JOIN FETCH m.playerA JOIN FETCH m.playerB")
    override fun findAll(): List<TournamentEntity>

    @Query("SELECT t FROM TournamentEntity t JOIN FETCH t.players JOIN FETCH t.tournamentRounds tr JOIN FETCH tr.matches m JOIN FETCH m.playerA JOIN FETCH m.playerB WHERE t.id = :id")
    override fun findById(id: Int): Optional<TournamentEntity>
}

此方式更灵活，可按需指定加载的关联，性能可控性更强。

方案3：使用@EntityGraph指定关联加载路径 #

利用Micronaut Data支持的@EntityGraph注解，通过路径声明要加载的关联，写法更简洁：

@JdbcRepository(dialect = Dialect.MYSQL)
@Transactional(Transactional.TxType.MANDATORY)
interface TournamentRepository : CrudRepository<TournamentEntity, Int>, JpaSpecificationExecutor<TournamentEntity> {

    @EntityGraph(attributePaths = ["players", "tournamentRounds", "tournamentRounds.matches", "tournamentRounds.matches.playerA", "tournamentRounds.matches.playerB"])
    override fun findAll(): List<TournamentEntity>

    @EntityGraph(attributePaths = ["players", "tournamentRounds", "tournamentRounds.matches", "tournamentRounds.matches.playerA", "tournamentRounds.matches.playerB"])
    override fun findById(id: Int): Optional<TournamentEntity>
}

attributePaths通过点分隔层级关联，清晰定义需要加载的所有子级数据。

内容的提问来源于stack exchange，提问作者noko