Java 为什么hibernate在持久化由joined_子类继承类型映射的实体之前选择实体？

我有超类Person和两个子类Parent和Child。它们映射为联接的表继承类型

父对象和子对象具有双向一对多关系

Person.hbm.xml包含所有配置：

<hibernate-mapping>
    <class name="com.masterhibernate.SimpleHibernateDemo.Person"
        table="Person">
        <cache usage="read-write" />

        <id name="id" column="id">
            <generator class="assigned" />
        </id>

        <property name="name">
            <column name="name" length="16" not-null="true" />
        </property>
        <property name="surname">
            <column name="surname" length="36"></column>
        </property>
        <property name="address">
            <column name="address" length="22"></column>
        </property>
        <joined-subclass name="com.masterhibernate.SimpleHibernateDemo.Parent"
            table="Parent">
            <key column="person_id" foreign-key="parent_person" />
            <property name="job" column="WorkPlace" length="22" type="string" />
            <set name="children" inverse="true" cascade="save-update" lazy="true">
                <cache usage="read-write" />
                <!-- specifies foreign key column of child table -->
                <key column="ParentPK" />
                <one-to-many class="com.masterhibernate.SimpleHibernateDemo.Child" />
            </set>
        </joined-subclass>
        <joined-subclass name="com.masterhibernate.SimpleHibernateDemo.Child"
            table="Child">
            <key column="person_id" foreign-key="child_person" />
            <property name="toy" column="toy" length="55" type="string" />
            <many-to-one name="parent"
                class="com.masterhibernate.SimpleHibernateDemo.Parent" column="ParentPK"
                foreign-key="child_parent" />
        </joined-subclass>
    </class>
</hibernate-mapping>

当我持久化父实体和关联子实体时，会在插入之前发出select语句。这些select语句从子表和人员表中查询子实体的数据

每次运行时都会重新创建数据库，并且没有可选择的内容

以下是从持久化父实体和关联子实体的主方法中提取的内容：

Parent parent = new Parent("Volodia", "Levytskyi", "Mykolaiv");
        parent.setId((long) 1);
        Set<Child> children = new HashSet<>();
        children.add(new Child((long) 2, "Ivan", "McGregor", "Odessa"));
        children.add(new Child((long) 3, "Borys", "McRobin", "Donetsk"));
        children.add(new Child((long) 4, "Nazar", "McCrespo", "Teernopil"));
        parent.setChildren(children);

        Session session = sessionFactory.openSession();
        Transaction transaction = session.beginTransaction();

        session.save(parent);
        // Criteria criteria = session.createCriteria(Person.class);
        // List<Person> list = criteria.list();
        // System.out.println("list1=" + list);
        transaction.commit();
        session.close();

<generator class="assigned" />

Parent Parent=new Parent（“Volodia”、“Levytskyi”、“Mykolaiv”）；
父集合ID（（长）1）；
Set children=newhashset（）；
添加（新儿童（（长）2，“伊万”、“麦格雷戈”、“敖德萨”）；
添加（新儿童（（长）3，“Borys”、“McRobin”、“Donetsk”）；
添加（新儿童（（长）4，“纳扎尔”、“麦克雷波”、“蒂尔诺皮尔”）；
父母、子女（子女）；
Session Session=sessionFactory.openSession（）；
事务=会话。beginTransaction（）；
session.save（父级）；
//条件=session.createCriteria（Person.class）；
//List=criteria.List（）；
//System.out.println（“list1=“+list”）；
commit（）；
session.close（）；

---

为什么这些select语句在insert语句之前？

问题在于父实体和子实体的id生成器策略：

Parent parent = new Parent("Volodia", "Levytskyi", "Mykolaiv");
        parent.setId((long) 1);
        Set<Child> children = new HashSet<>();
        children.add(new Child((long) 2, "Ivan", "McGregor", "Odessa"));
        children.add(new Child((long) 3, "Borys", "McRobin", "Donetsk"));
        children.add(new Child((long) 4, "Nazar", "McCrespo", "Teernopil"));
        parent.setChildren(children);

        Session session = sessionFactory.openSession();
        Transaction transaction = session.beginTransaction();

        session.save(parent);
        // Criteria criteria = session.createCriteria(Person.class);
        // List<Person> list = criteria.list();
        // System.out.println("list1=" + list);
        transaction.commit();
        session.close();

<generator class="assigned" />