三、ORM(Object/Relational Mapper)---Hibernate对象映射

一. ORM(Object/Relational Mapper)

1. OR映射---Hibernate基本数据类型

Hibernate基本类型	Java类型	标准SQL字段类型
boolean	boolean, java.lang.Boolean	BIT
yes_no	boolean, java.lang.Boolean	CHAR(1) ( 'Y'/'N')
true_false	boolean, java.lang.Boolean	CHAR(1) ( 'Y'/'N')
byte	byte, java.lang.Byte	TINYINT
short	short, java.lang.Short	SMALLINT
integer	int, java.lang.Integer	INTEGER
long	long, java.lang.Long	BIGINT
float	float, java.lang.Float	FLOAT
double	double, java.lang.Double	DOUBLE
big_decimal	java.math.BigDecimal	NUMBERIC
character	java.lang.String	CHAR(1)
string	java.lang.String	VARCHAR
date	java.util.Date, java.sql.Date	DATE
time	java.util.Date, java.sql.Time	TIME
timestamp	java.util.Date, java.sql.TimeStamp	TIMESTAMP
calendar	java.util.Calendar	TIMESTAMP
calendar_date	java.util.Calendar	Date
clob	java.sql.Clob	CLOB
blob	java.sql.Blob	BLOB
binary	byte[]	VARBINARY, BLOB
text	java.lang.String	CLOB
seralizable	java.io.Serializable	VARBINARY, BLOB
class	java.lang.Class	VARCHAR
local	java.util.Locale	VARCHAR
timezone	java.util.TimeZone	VARCHAR
currency	java.util.Currency	VARCHAR

2. OR映射---实体映射示例

*.hbm.xml, 主体内容包含表/类映射, id映射, 属性字段映射三个部分.示例文件:
<?xml version="1.0" encoding="UTF-8"?>

<!DOCTYPE hibernate-mapping PUBLIC

"-//Hibernate/Hibernate Mapping DTD 3.0//EN"

"http://hibernate.sourceforge.net/hibernate-mapping-3.0.dtd">

<hibernate-mapping>

<class name="cn.thinkmore.hibernate.pojo.Tuser" table="T_user">

<id column="id" name="id" type="java.lang.Integer">

<generator class="increment" />

</id>

<property column="name" name="name" type="java.lang.String"/>

</class>

</hibernate-mapping>

XML文件头定义了文件的编码方式, DTD与Hibernate的版本有关, 上面是使用hibernate 3.x使用的DTD.

3. OR映射---表/类映射

<class name="cn.thinkmore.hibernate.pojo.Tuser" table="T_user">

name: 指定了Java数据封装的POJO类类型.
table: 指定了数据库的表名

4. OR映射---id映射

<id name="id" type="java.lang.Integer" column="id">

<generator class="increment" />

</id>

name: 指定了映射POJO类的属性名,
type: POJO类中属性的数据类型;
column: 数据库表的主键字段;
generator子元素:由其class属性指定主键生成方式:

assigned: 主键由应用程序产生, 不需要hibernate干预

identity: 使用数据库的主键生成机制, 如MySQL, DB2、SQL Server的自增主键.

sequence: 使用数据库的sequence机制, 如Oracle的sequence

uuid.hex: 由Hibernate基于128位唯一值产生算法, 根据ip, 时间, jvm启动时间, 内部自增量生成十六进制的数值, 编码后成为一个32位长的字符串. 该方法提供了最好的数据库插入性能和数据库平台适应性.

uuid.string: 与uuid.hex类似, 只是生成的主键没有进行编码, 只有16位长度. 在某些数据库可能出错.

hilo: 通过hilo算法实现主键生成, 需要额外的数据库表保存主键生成历史状态.

seqhilo: 与hilo类似, 只是主键历史状态保存在sequence中, 适用于支持sequence的数据库.

increment: 主键按数值递增, 但如果多个实例同时访问同一个数据库, 各自生成主键,则容易造成主键重复.

native: 由hibernate根据数据库适配器中的定义, 自动采用identity, hilo, sequence中的一种方式.

foreign: 外部表的字段作主键.

select: hibernate 3 中新增的.需要提供一个唯一的标识字段进行二次读取, 以获取触发器生成的主键值, 通过param子元素进行定义, 比如:

<generator class="select">

<param name="key">key_field</param>

</generator>

该方法主要针对遗留系统的改造工程, 一些早期的系统主键依赖于触发器生成. 当数据库insert时, 触发器捕获这一操作, 并为主键赋值, 在插入数据库后, 再次读取某一识别字段读取已经插入的数据, 获取其主键值.

5. OR映射---复合主键映射(composite-id)---实体属性组成主键

复合主键使用<composite-id>将取代id元素, 并具有property属性列表.
<composite-id>
<key-property column="userid" name="userid" type="java.lang.String" />
<key-property column="when" name="when" type="java.sql.Date" />
</composite-id>
复合主键的POJO类需要实现equals和hashcode方法, 可以使用apache commons lang包中的工具类实现(commons-lang.jar), 比如:

import org.apache.commons.lang.builder.EqualsBuilder;
import org.apache.commons.lang.builder.HashCodeBuilder;
import org.apache.commons.lang.builder.ToStringBuilder;

...

public String toString(){
return new ToStringBuilder(this)
.append("userid ", getUserid ())
.append("when", getWhen ())
.toString();
}

public boolean equals(Object other){
if(!(other instanceof MyPoJoClass)){
return false;
}
MyPoJoClass castOther=( MyPoJoClass)other;
return new EqualsBuilder()
.appendSuper(super.equals(other))
.append(this.getUserid (),castOther.getUserid() )
.append(this.getWhen (),castOther.getWhen() )
.isEquals();
}

public int hashCode(){
return new HashCodeBuilder()
.appendSuper(super.hashCode() )
.append(getUserid ())
.append(getWhen ())
.toHashCode();
}
装载复合主键的记录时, 考虑把类对应的对象的主键值填充好后作为load的第二个参数.
MyPoJoClass obj = new MyPoJoClass();
obj.setUserid(...);
obj.setWhen(...);
obj=(MyPoJoClass)session.load(MyPoJoClass.class, obj);

6. OR映射---复合主键映射(composite-id)---基于主键类

可以用一个独立的类来描述主键, 示例:
<composite-id name="keyClassProperty" class="MyPoJoClass">
<key-property column="userid" name="userid" type="java.lang.String" />
<key-property column="when" name="when" type="java.sql.Date" />
</composite-id>
此时, 需要定义一个新的类KeyClass来作为主键类, KeyClass实现equals和hashcode方法, 而在POJO中, 使用属性名keyClassProperty来表示主键, 其类类型为KeyClass.

7. OR映射---composite-id其他子节点

composite-id除了key-property子节点外, 还具有可选<key-many-to-one>子节点.完整的节点内容:
<composite-id name="propertyName"
class="ClassName"
unsaved-value="any|none" >

<key-property column="colum_name"
name=" propertyName "
type="type_name" />
...
<key-many-to-one column=" colum_name " calss=" ClassName "
name=" propertyName " />
...
</composite-id>

8. OR映射---属性/字段映射

<property name="name" type="java.lang.String" column="name" />
name: POJO类的属性名,
type: POJO类中属性的数据类型; 如果是字符串,可以省略
column: 数据库中的字段名. 如果和属性同名, 可以省略.

9. OR映射---自定义数据类型 UserType, CompositeUserType接口

hibernate3 中的为: org.hibernate.usertype(hibernate2中为net.sf包)下面的两个接口: UserType 和 CompositeUserType . 它们提供自定义类型的规范, 这里以UserType为例.
package org.hibernate.usertype;

import java.io.Serializable;
import java.sql.PreparedStatement;
import java.sql.ResultSet;
import java.sql.SQLException;

import org.hibernate.HibernateException;

public interface UserType {

/**
* 返回本类型所映射的SQL类型, 返回的是int[]数组. 可以取java.sql.Types中的int静态常量.
* 比如有: public final static int DOUBLE = 8;
*/
public int[] sqlTypes();

/**
* 定义本类型的nullSafeGet() 方法返回的数据的类型
*/
public Class returnedClass();

/**
* 定义两个数据的比较方法, 返回true表示相等, false表示不等.
*/
public boolean equals(Object x, Object y) throws HibernateException;

/**
* Get a hashcode for the instance, consistent with persistence "equality"
*/
public int hashCode(Object x) throws HibernateException;

/**
* 定义从ResultSet中读取数据并转换成自定义类型的方法, 对数据库null应该考虑
* 参数names 包含了当前自定义类型的映射字段名称
*/
public Object nullSafeGet(ResultSet rs, String[] names, Object owner)
throws HibernateException, SQLException;

/**
* Hibernate进行数据保存时被调用的方法
* @param st a JDBC prepared statement
* @param value the object to write, 类型应该是returnedClass()方法返回的Class指定的类型
* @param index statement parameter index
* @throws HibernateException
* @throws SQLException
*/
public void nullSafeSet(PreparedStatement st, Object value, int index)
throws HibernateException, SQLException;
/**
* 自定义的对象复制方法, 用作构造返回对象.
* nullSafeGet方法被调用后, Hibernate获得了自定义数据对象,
* 这个对象成为数据库读出的原始对象, 通过deepCopy方法, Hibernate
* 返回一个复本给用户. 脏数据检查时, 如果两个对象不等(equals方法返回false),
* 就会执行数据库持久化操作.
* @param value the object to be cloned, which may be null
* @return Object a copy
*/
public Object deepCopy(Object value) throws HibernateException;

/**
* Are objects of this type mutable?
* 本类型实例是否可变
* @return boolean
*/
public boolean isMutable();

/**
* 把对象转化为二级缓存中保存. 经常把当前对象对其他对象的引用, 保存为其id值.
* Transform the object into its cacheable representation. At the very least this
* method should perform a deep copy if the type is mutable. That may not be enough
* for some implementations, however; for example, associations must be cached as
* identifier values. (optional operation)
*
* @param value the object to be cached
* @return a cachable representation of the object
* @throws HibernateException
*/
public Serializable disassemble(Object value) throws HibernateException;

/**
* 把二级缓存中获取的数据转换为自定义的对象数据
* Reconstruct an object from the cacheable representation. At the very least this
* method should perform a deep copy if the type is mutable. (optional operation)
*
* @param cached the object to be cached
* @param owner the owner of the cached object
* @return a reconstructed object from the cachable representation
* @throws HibernateException
*/
public Object assemble(Serializable cached, Object owner) throws HibernateException;

/**
* During merge, replace the existing (target) value in the entity we are merging to
* with a new (original) value from the detached entity we are merging. For immutable
* objects, or null values, it is safe to simply return the first parameter. For
* mutable objects, it is safe to return a copy of the first parameter. For objects
* with component values, it might make sense to recursively replace component values.
*
* @param original the value from the detached entity being merged
* @param target the value in the managed entity
* @return the value to be merged
*/
public Object replace(Object original, Object target, Object owner)
throws HibernateException;
}

如果一个用户具有多个email, 可以分别定义多个字段保存, 也可以定义一个子表专门保存, 但都有点大动干戈. 可以考虑用一个字符串字段保存, 相邻的email用一个;隔开. 比如:
package cn.thinkmore.hibernate.pojo;

import java.io.Serializable;
import java.sql.PreparedStatement;
import java.sql.ResultSet;
import java.sql.SQLException;
import java.sql.Types;
import java.util.*;

import org.hibernate.Hibernate;
import org.hibernate.HibernateException;
import org.hibernate.usertype.UserType;

public class EmailList implements UserType {

private static final String SPLITTER = ";";

private static final int[] TYPES = new int[] { Types.VARCHAR };

public Object assemble(Serializable cached, Object owner)
throws HibernateException {
return null;
}

public Object deepCopy(Object value) throws HibernateException {
if (!(value instanceof List)) {
return null;
}
List src = (List) value;
List tar = new ArrayList();
tar.addAll(src);
return tar;
}

public Serializable disassemble(Object value) throws HibernateException {
return null;
}

public boolean equals(Object x, Object y) throws HibernateException {
if (x == y)
return true;
if (x != null && y != null) {
List xList = (List) x;
List yList = (List) y;
if (xList.size() != yList.size())
return false;
for (int i = 0; i < xList.size(); i++) {
String strX = xList.get(i).toString();
String strY = yList.get(i).toString();
if (!strX.equals(strY))
return false;
}
}
return false;
}

public boolean isMutable() {
return false;
}

public Object nullSafeGet(ResultSet rs, String[] names, Object owner)
throws HibernateException, SQLException {
String value = Hibernate.STRING.nullSafeGet(rs, names[0]).toString();
if(null!=value)
return parse(value);
return null;
}

private List parse(String value) {
String[] strs = org.apache.commons.lang.StringUtils.split(value,SPLITTER);
List<String> emails = new ArrayList<String>();
for(int i=0; i<strs.length;i++){
emails.add(strs[i]);
}
return emails;
}

private String assemble(List list) {
StringBuilder sb = new StringBuilder();
int maxIndex = list.size()-1;
for(int i=0; i<maxIndex;i++){
sb.append(list.get(i)).append(SPLITTER);
}
if(0<=maxIndex){
sb.append(list.get(maxIndex));
}
return sb.toString();
}

public void nullSafeSet(PreparedStatement st, Object value, int index)
throws HibernateException, SQLException {
if(null!=value&&(value instanceof List)){
String str = assemble((List)value);
Hibernate.STRING.nullSafeSet(st,str,index);
}else if(null!=value){
Hibernate.STRING.nullSafeSet(st,value.toString(), index);
}else{
Hibernate.STRING.nullSafeSet(st,null, index);
}
}

public Object replace(Object original, Object target, Object owner)
throws HibernateException {
return null;
}

public Class returnedClass() {
return List.class;
}

public int[] sqlTypes() {
return TYPES;
}

public int hashCode(Object x) throws HibernateException {
return 0;
}
}

数据关联关系

数据关联关系在实体的子类里面不会被自动继承.

10. 一对一关联 ----- 主键关联 one-to-one

一对一的主键关联, 指两个表的记录是一对一的关系, 且一个表的主键取为另一个表对应的记录的主键.
<hibernate-mapping>

<class name="cn.thinkmore.hibernate.pojo.Citizen" table="citizen" lazy="true">

<id name="id" column="id">

<generator class="uuid.hex" />

</id>

<property name="name" />

<property name="sex"/>

<property name="idcard" />

<property name="address"/>

<one-to-one name="gene" cascade="all" outer-join="true"

class="cn.thinkmore.hibernate.pojo.Gene" />

</class>

</hibernate-mapping>

name: 属性名
class: java全路径类名
cascade: 是否级连操作, all , none
outer-join: 是否外链接, true/false
<hibernate-mapping>

<class name="cn.thinkmore.hibernate.pojo.Gene" table="gene" lazy="true">

<id name="id" column="id" >

<generator class="foreign">

<param name="property">citizen</param>

</generator>

</id>

<property name="dna" />

<property name="chromosome"/>

<property name="species"/>

<one-to-one name="citizen" constrained="true"

class="cn.thinkmore.hibernate.pojo.Citizen" />

</class>

</hibernate-mapping>

Constrained: 次关联中的约束方.
.....
Session s = HibernateSession3.getSession();

Citizen one = new Citizen();

one.setIdcard("123456789012345678");

one.setName("Mike");

one.setSex("M");

Gene gene = new Gene();

gene.setBloodType("AB");

gene.setDna("Ajodjjd[[joa");

gene.setSpecial("White");

one.setGene(gene);

gene.setCitizen(one);

Transaction trans = s.beginTransaction();

s.saveOrUpdate(one);

trans.commit();

// 两条数据一起被保存, 查询时也会关联查出来.

Query query = s.createQuery("From Citizen");

List list = query.list();

for (int i = 0; i < list.size(); i++) {

one = (Citizen) list.get(i);

if (null == one)

continue;

gene = (Gene) one.getGene();

if (null != gene){

System.out.println(gene.getId()+":"+gene.getSpecial()+":"+gene.getDna());

}

}

11. 单向一对多关联 ----- 唯一外键关联 one-to-many

一个用户可以多个email:

<set name="email" table="emails" cascade="all">
<key column="userid" />
<one-to-many calss="....Email" />
</set>

12. 双向一对多关联

除了在主控方配置一对多, 还要在被控方配置对应的多对一关系:

User:
<set name="email" table="emails" cascade="all" inverse="true">
<key column="userid" />
<one-to-many calss="....Email" />
</set>

Email:
<many-to-one name="user" class=".....User" cascade="none" column="userid" />

再有例子:一个职员(是人的子类)前后可能有多条工作(每次工作是一条记录):
<hibernate-mapping >

<class name="webapp.hibernate.pojo.People" table="people" lazy="false">

<comment>Users may bid for or sell auction items.</comment>

<id name="id">

<generator class="uuid.hex"/>

</id>

<property name="name" />

<property name="idcard" />

<property name="sex"/>

<property name="birth" type="date"/>

<one-to-one name="gene"

cascade="all"

class="webapp.hibernate.pojo.Gene"/>

</class>

</hibernate-mapping>

<hibernate-mapping >

<class name="webapp.hibernate.pojo.Employee" table="people" lazy="false">

<id name="id">

<generator class="uuid.hex"/>

</id>

<property name="name" />

<property name="idcard" />

<property name="sex"/>

<property name="birth" type="date"/>

<one-to-one name="gene"

cascade="all"

class="webapp.hibernate.pojo.Gene"/>

<set name="jobs" table="joblist" cascade="all" inverse="true">

<key column="empid" />

<one-to-many class="webapp.hibernate.pojo.JobList" />

</set>

</class>

</hibernate-mapping>

<hibernate-mapping >

<class name="webapp.hibernate.pojo.JobList" table="joblist" lazy="false">

<id name="id">

<generator class="uuid.hex"/>

</id>

<!-- <property name="userid" column="empid"/> -->

<property name="company" />

<property name="begin" type="date"/>

<property name="end" type="date"/>

<many-to-one name="employee" class="webapp.hibernate.pojo.Employee" cascade="none" column="empid" />

</class>

</hibernate-mapping>
public static void testOne2Many() {

Session s = HibernateSession3.getSession();

Employee e = new Employee();

e.setName("一对多");

e.setIdcard("123456789012345678");

e.setBirth(new java.sql.Date(System.currentTimeMillis()-24L*3600*1000*365*25));

JobList job1 = new JobList();

job1.setEmployee(e);

job1.setCompany("A gongsi");

job1.setBegin(new java.sql.Date(System.currentTimeMillis()-24L*3600*1000*365*5));

job1.setEnd(new java.sql.Date(System.currentTimeMillis()-24L*3600*1000*365*4));

JobList job2 = new JobList();

job2.setEmployee(e);

job2.setCompany("B gongsi");

job2.setBegin(new java.sql.Date(System.currentTimeMillis()-24L*3600*1000*365*4));

job2.setEnd(new java.sql.Date(System.currentTimeMillis()-24L*3600*1000*365*2));

Set<JobList> jobs = new HashSet<JobList>();

jobs.add(job1);

jobs.add(job2);

e.setJobs(jobs);

Transaction trans = s.beginTransaction();

// 两个表的数据一起被保存, 查询时也会关联查出来.

s.saveOrUpdate(e);

trans.commit();

Query query = s.createQuery("From Employee");

List list = query.list();

for (int i = 0; i < list.size(); i++) {

Employee one = (Employee) list.get(i);

if (null == one)

continue;

System.out.println(one.getName()+":"+one.getBirth()+":");

Set joblist = one.getJobs();

for(Object o:joblist){

JobList j = (JobList)o;

System.out.println("\t工作:"+j.getCompany()+":"+j.getBegin()+":"+j.getEnd());

}

System.out.println("---------------------------------------");

}

}

13. 双向多对多关联

一个用户可以在银行有多个帐号
User:
<class name="webapp.hibernate.pojo.MoneyUser" table="people" lazy="false">

<id name="id">

<generator class="uuid.hex"/>

</id>

<property name="name" />

<property name="idcard" />

<property name="sex"/>

<property name="birth" type="date"/>

<one-to-one name="gene"

cascade="all"

class="webapp.hibernate.pojo.Gene"/>

<set name="banks" table="bankuser" cascade="none" >

<key column="userid" />

<many-to-many class="webapp.hibernate.pojo.Bank" column="bankid" />

</set>

</class>

Bank:
<class name="webapp.hibernate.pojo.Bank" table="bank" lazy="false">

<comment>Users may bid for or sell auction items.</comment>

<id name="id">

<generator class="uuid.hex"/>

</id>

<property name="bankname" />

<property name="type"/>

<property name="createday" type="date"/>

<set name="cutomers" table="bankuser" cascade="none" >

<key column="bankid" />

<many-to-many class="webapp.hibernate.pojo.MoneyUser" column="userid" />

</set>

</class>

public static void testMany2Many() {

Session s = HibernateSession3.getSession();

Iterator<MoneyUser> itor =(Iterator<MoneyUser>)s.createQuery("FROM MoneyUser").iterate();

while(itor.hasNext()){

MoneyUser moneyUser = itor.next();

Set banks = moneyUser.getBanks();

Iterator<Bank> bankItor = banks.iterator();

while(bankItor.hasNext()){

Bank bank = bankItor.next();

System.out.println(moneyUser.getName()+" "+bank.getBankname());

}

}

}

public static void testMany2Many2() {

Session s = HibernateSession3.getSession();

Iterator<Bank> itor = s.createQuery("FROM Bank").iterate();

while(itor.hasNext()){

Bank bank = itor.next();

Set customers = bank.getCutomers();

Iterator<MoneyUser> customerItor = customers.iterator();

while(customerItor.hasNext()){

MoneyUser moneyUser = customerItor.next();

System.out.println(bank.getBankname()+" "+moneyUser.getName());

}

}

}