Solution
You tried to delete an object that was allready deleted from the database.
This can happen when cascading deletes where performed somewhere and you did not reattach your object. For deleting you do not nee to reattach your object but this could prevent this problem.
Exception Message
19:47:56,078 ERROR AbstractBatcher:61 - Exception executing batch:
org.hibernate.StaleStateException: Batch update returned unexpected row count from update: 0 actual row count: 0 expected: 1
at org.hibernate.jdbc.BatchingBatcher.checkRowCount(BatchingBatcher.java:93)
at org.hibernate.jdbc.BatchingBatcher.checkRowCounts(BatchingBatcher.java:79)
at org.hibernate.jdbc.BatchingBatcher.doExecuteBatch(BatchingBatcher.java:58)
at org.hibernate.jdbc.AbstractBatcher.executeBatch(AbstractBatcher.java:195)
at org.hibernate.engine.ActionQueue.executeActions(ActionQueue.java:230)
at org.hibernate.engine.ActionQueue.executeActions(ActionQueue.java:145)
at org.hibernate.event.def.AbstractFlushingEventListener.performExecutions(AbstractFlushingEventListener.java:296)
at org.hibernate.event.def.DefaultFlushEventListener.onFlush(DefaultFlushEventListener.java:27)
at org.hibernate.impl.SessionImpl.flush(SessionImpl.java:1009)
at org.hibernate.impl.SessionImpl.managedFlush(SessionImpl.java:356)
]]>
一 hibernate配置
1.下蝲所需?br>下蝲hibernate2.13, hibernate-extension2.13,Middlegen(已装?jin)jdk和ant)
2.创徏数据?例子:student;course,stu_cou)
Student(id,name)
Course(id,name)
Stu_cou(stu_id,cou_id)(分别是student表和course表的外健)
3.使用MiddleGen(见附?
?hibernate映射关系(many-to-many)
1. 分析inverse参数(student.hbm.xml course.hbm.xml)
inverse该参数用来负责映之间的关系,讄为false的一方可以对他们的关p进行维?br>讄inverse = false的实?表示在映中是它是主动关p?由它来维户关pM的数?br>讄inverse = true 的实?表示在映中是它是被动关p?它不l护关系中的数据
1) inverse=false的情?br>a) d记录的情?br>student.hbm.xml是主动方,由它来决定stu_cou表中的记?故inverse=false
当向stu_cou中写入记录时?
/*
* 试目的:在stu_cou中添加一l对应关p?student,course)
* 试l果:当执行完该段代码?br>* 如stu_cou中没?student,course)?自动加入(student,course)
* 如stu_cou中存?student,course)?什么都不作,不会(x)存在d冲突
Student student=(Student)session.load(Student.class,student2.getId())
Course course=(Course)session.load(Course.class,course2.getId());
student.getCourses().add(course);
session.save(student);
执行完后,hibenrate成功昄
Hibernate: select course0_.id as id0_, course0_.name as name0_ from course course0_ where course0_.id=?
Hibernate: select student0_.id as id0_, student0_.name as name0_ from student student0_ where student0_.id=?
Hibernate: select courses0_.cid as cid__, courses0_.sid as sid__ from stu_cou courses0_ where courses0_.sid=?
Hibernate: insert into stu_cou (sid, cid) values (?, ?)
b) 删除记录的情?br>Student student=(Student)session.load(Student.class,student2.getId())
Course course=(Course)session.load(Course.class,course2.getId());
student.getCourses().让remove(course);
session.save(student);
代码执行完后,可以成功执行,hibernate昄:
Hibernate: select course0_.id as id0_, course0_.name as name0_ from course course0_ where course0_.id=?
Hibernate: select student0_.id as id0_, student0_.name as name0_ from student student0_ where student0_.id=?
Hibernate: select courses0_.cid as cid__, courses0_.sid as sid__ from stu_cou courses0_ where courses0_.sid=?
Hibernate: delete from stu_cou where sid=?
2) inverse=true的情?br>student.hbm.xml是被动方,讄它的inverse=true
a) d记录情况
* 试目的:在stu_cou中添加一l对应关p?student,course)
* 试l果:当执行完该段代码?br>* 如stu_cou中没?student,course)?自动加入(student,course)
* 如stu_cou中存?student,course)?什么都不作,不会(x)存在d冲突
Student tudent=(Student)session.load(Student.class,student2.getId());
Course course=(Course)session.load(Course.class,course2.getId());
student.getCourses().add(course);
session.save(student);
代码执行完后,hibernate昄,没有插入记录
Hibernate: select course0_.id as id0_, course0_.name as name0_ from course course0_ where course0_.id=?
Hibernate: select student0_.id as id0_, student0_.name as name0_ from student student0_ where student0_.id=?
Hibernate: select courses0_.cid as cid__, courses0_.sid as sid__ from stu_cou courses0_ where courses0_.sid=?
b) 删除记录情况
当向stu_cou中删除记录时?
Student student=(Student)session.load(Student.class,student2.getId())
Course course= (Course)session.load(Course.class,course2.getId());
student.getCourses().remove(course);
session.save(student);
执行?hibernate昄如下,证明没有删除记录
Hibernate: select course0_.id as id0_, course0_.name as name0_ from course course0_ where course0_.id=?
Hibernate: select student0_.id as id0_, student0_.name as name0_ from student student0_ where student0_.id=?
Hibernate: select courses0_.cid as cid__, courses0_.sid as sid__ from stu_cou courses0_ where courses0_.sid=?
Hibernate: select course0_.id as id0_, course0_.name as name0_ from course course0_ where course0_.id=?
Hibernate: select course0_.id as id0_, course0_.name as name0_ from course course0_ where course0_.id=?
2. 分析cascade参数(student.hbm.xml course.hbm.xml)
只有讄?jin)inverse=false才可以操Ustu_cou的数?br>当inverse=true的时?下列情况1中b)提示错误
原因: stu_cou中存在外健约?你无法直接删除student表中的数?br>1) cascade=save-update的情?br>a)d记录情况
Set set=new HashSet();
set.add(course1);
set.add(course2);
set.add(course3);
student3.setCourses(set);
session.save(student3);
执行后hibernate提示:
Hibernate: insert into student (name) values (?)
Hibernate: insert into course (name) values (?)
Hibernate: insert into course (name) values (?)
Hibernate: insert into course (name) values (?)
Hibernate: insert into stu_cou (sid, cid) values (?, ?)
可见,插入?jin)student,course,和stu_cou三个?/p>
b)删除记录情况
Student student=(Student)session.load(Student.class,new Integer(22));
session.delete(student);
成功执行?hibernate提示:
Hibernate: select student0_.id as id0_, student0_.name as name0_ from student student0_ where student0_.id=?
Hibernate: delete from stu_cou where sid=?
Hibernate: delete from student where
2) cascade=all的情?br>a) d记录情况
Set set=new HashSet();
set.add(course1);
set.add(course2);
set.add(course3);
student3.setCourses(set);
session.save(student3);
执行后hibernate提示:成功dU联关系
Hibernate: insert into student (name) values (?)
Hibernate: insert into course (name) values (?)
Hibernate: insert into course (name) values (?)
Hibernate: insert into course (name) values (?)
Hibernate: insert into stu_cou (sid, cid) values (?, ?)
b) 删除记录情况
Student student=(Student)session.load(Student.class,new Integer(22));
session.delete(student);
试l果(所有和student(22)相关的记录均被删?
删除student表中的student(22)记录
删除stu_cou表中的student(22)的记?br>删除course表中的student(22)对应的course的记?/p>
2) cascade=none情况
a) d记录情况
Set set=new HashSet();
set.add(course1);
set.add(course2);
set.add(course3);
student3.setCourses(set);
session.save(student3);
p|,因ؓ(f)没有U联关系,所以course记录d不进?造成外健p|
b) 删除记录情况
Student student=(Student)session.load(Student.class,new Integer(22));
session.delete(student);
正常,删除?jin)student表中的记录student(22),同时删除?jin)stu_cou中对应的student(22)的记?/p>
附录
1QMiddleGen的?br>1环境变量讄
%Hibernate_Home%/lib/*.jar?MiddleGen%/lib?br>%Hibernate_Home%/hibernate2.jar?MiddleGen%/lib?br>%Hibernate-Extension_Home%/tools/lib/*.jar?MiddleGen%/lib?br>%Hibernate-Extension_Home%/tools/hibernate-tools.jar?MiddleGen%/lib?/p>
2 MiddleGen的配|?br>配置目标数据库参?br>q入MiddleGen 目录下的\config\database 子目录，Ҏ(gu)我们实际采用的数据库打开对应的配|文件。如q里我用的是mysql数据库，对应的就是mysql.xml文g?br><property
name="database.script.file" value="${src.dir}/sql/${name}-mysql.sql"/>
<property
name="database.driver.file" value="${lib.dir}/mysql.jar"/>
<property
name="database.driver.classpath" value="${database.driver.file}"/>
<property
name="database.driver" value="org.gjt.mm.mysql.Driver"/>
<property
name="database.url" value="jdbc:mysql://localhost/sample"/>
<property
name="database.userid" value="user"/>
<property
name="database.password" value="mypass"/>
<property
name="database.schema" value=""/>
<property
name="database.catalog" value=""/>
<property
name="jboss.datasource.mapping" value="mySQL"/>
其中下划U标准的部分是我们进行配|的内容Q分别是数据url以及(qing)数据库用
户名和密码?/p>
1) 修改Build.xml
修改MiddleGen 根目录下的build.xml 文gQ此文g是Middlegen-Hibernate 的Ant构徏配置。Middlegen-Hibernate根据build.xml文g中的具体参数生成数据库表映射文g。可配置的项目包括：(x)
a) 目标数据库配|文件地址
查找关键?”!ENTITY”Q得刎ͼ(x)
<!DOCTYPE project [
<!ENTITY database SYSTEM
"file:./config/database/hsqldb.xml">
]>
默认情况下，MiddleGen 采用的是hsqldb.xmlQ?br>其修改为我们所用的数据库配|文Ӟmysql.xmlQ：(x)
<!DOCTYPE project [
<!ENTITY database SYSTEM
"file:./config/database/mysql.xml">
]>
b) Application name
查找Q?br><property value="airline"/>
“aireline”是MiddleGen原始配置中默认的 Application NameQ将其修改ؓ(f)我们
所希望的名Uͼ?#8220;HibernateSample”Q?br><property value="HibernateSample"/>
c) 输出目录
查找关键?#8220;name="build.gen-src.dir"”Q得刎ͼ(x)
<property
value="${build.dir}/gen-src"/>
修改value="${build.dir}/gen-src"使其指向我们所期望的输出目录，
q里我们修改为：(x)<property
value="C:\sample"/>
d) 对应代码的Package name
查找关键?#8220;destination”Q得刎ͼ(x)
<hibernate
destination="${build.gen-src.dir}"
package="${name}.hibernate"
genXDocletTags="false"
genIntergratedCompositeKeys="false"
javaTypeMapper=
"middlegen.plugins.hibernate.HibernateJavaTypeMapper"
/>
可以看到Qhibernate 节点package 属性的默认讄实际上是由前面的
Application Name Q?{name}Q和“.hibernate”l合而成Q根据我们的需要，
q里q有一个属性genXDocletTagsQ如果设|ؓ(f)trueQ则生成的代码将包含
xdoclet tagQ这Z后在开发过E中借助xdocletq行映射调整提供?jin)帮助。注意，如果使用的数据库为SQLServerQ需要将build.xml部分删除(参见?^的文?Q否则Middlegen?x)报出找不到表的错误?br>x(chng)为止QMiddleGen 已经配置完毕Q在MiddleGen 根目录下q行antQ就出?br>MiddleGen的界面：(x)
2 *.xml?java
在得?xml?使用ant hbm2java 卛_得到相应的java文g
ȝ:
我第一ơ运行成?但是后来则出现如下类似错?(g)查发现在MiddleGen的lib目录下存在多个类似的jar?br>(velocity.jar,velocity1-0.8.jar),删除旧的jar包后,问题解决
[middlegen] java.lang.IncompatibleClassChangeError
[middlegen] at [middlegen] at org.apache.velocity.Template.process(Template.java:136)
2Q?评论)关于Many-to-Many需要注意的问题
比如group/user,user/role,group/role,都是多对多的关系?br>所以有必要搞搞清楚。以group/userZ
(1)group.getUsers() & user.getGroups()双向映射。在q里Q考虑到user.setGroups()是不?x)放出来的。所以在user端加入inverse=trueQ就是说Q让Group端来l护两者的关系表?br>(2)于是Q添加的时候，要把user加入到group的user set中，保存Q才有意义！
(3)如果要删除group_user的关p，而不是group/user本nQ那么，需要在group端设定casade=save-update。否则，group.getUsers().remove(user)不会(x)起作用！
(4)׃有了(jin)FK保护Q你是无法单独删除User的。同Ӟ׃你指定了(jin)inverse=true,所以无法光光的讄user.getGroups().clearp够搞定一切。你必须首先在每个Group.getUsers中删除这个userQ才能最后删除。所以，UserManagerImpl代码可能看v来像q样Q?br>public void delUser(User user) {
if (user == null) {
logger.info("Got a null user object!");
return;
}
Session session = null;
Transaction trans = null;
boolean bSuccess = false;
try {
String username = new String(user.getName());
session = PLUtils.getSession();
trans = session.beginTransaction();
session.update(user);
Set groups = ((UserEntity)user).getGroups();
for(Iterator i = groups.iterator();i.hasNext();){
GroupEntity g = (GroupEntity)i.next();
session.update(g);
g.getUsers().remove(user);
session.update(g);
}
// ((UserEntity)user).getGroups().clear();
session.delete(user);
bSuccess = true;
logger.debug("User " + username + " deleted ok!");
}
catch (ObjectNotFoundException onfe) {
logger.info("user not found!", onfe);
}
catch (Exception e) {
logger.error("failed!", e);
}
finally {
try {
if (bSuccess)
trans.commit();
else
trans.rollback();
session.close();
}
catch (Exception ex) {
}
}
}
]]>
最q在目中?Spring ?Hibernate q行开发，有感?Criteria 比较好用Q在查询Ҏ(gu)设计上可以灵zȝҎ(gu) Criteria 的特Ҏ(gu)方便地进行查询条件的l装。所以现在对 Hibernate ?Criteria 深入研究一下。?Hibernate Reference 》及(qing)|上其它一些资料对 Criteria 已经做了(jin)很多介绍?/p>
最q在目中?/span> Spring ?/span> Hibernate q行开发，有感?/span> Criteria 比较好用Q在查询Ҏ(gu)设计上可以灵zȝҎ(gu) Criteria 的特Ҏ(gu)方便地进行查询条件的l装。所以现在对 Hibernate ?/span> Criteria 深入研究一下。?/span> Hibernate Reference 》及(qing)|上其它一些资料对 Criteria 已经做了(jin)很多介绍。本文主要是?/span> Criteria 的结构入手来q行分析?/span>
(tng) (tng) (tng) (tng) (tng) (tng) 如图 1 ?/span> Hibernate 设计?/span> CriteriaSpecification 作ؓ(f) Criteria 的顶U接口，其下面提供了(jin) Criteria ?/span> DetachedCriteria ?/span>
Criteria ?/span> DetachedCriteria 的主要区别在于创建的形式不一P Criteria 是在U的Q所以它是由 Hibernate Session q行创徏的；?/span> DetachedCriteria 是离U的Q创建时无需 Session Q?/span> DetachedCriteria 提供?/span> 4 个静(rn)态方?/span> forClass(Class) ?/span> forEntityName(Name) q行 DetachedCriteria 实例的创建?/span> Spring 的框架提供了(jin)
getHibernateTemplate().findByCriteria(detachedCriteria) Ҏ(gu)可以很方便地Ҏ(gu)
DetachedCriteria 来返回查询结果?/span>
如图 1 Q?/span> Criteria ?/span> DetachedCriteria 均可使用 Criterion ?/span> Projection 讄查询条g。可以设|?/span> FetchMode( 联合查询抓取的模?/span> ) Q设|排序方式。对?/span> Criteria q可以设|?/span> FlushModel Q冲?/span> Session 的方式）(j)?/span> LockMode Q数据库锁模式）(j)?/span>
下面对 Criterion ?/span> Projection q行详细说明?/span>
(tng) (tng) (tng) (tng) (tng) (tng) 
?/span> 1
(tng) (tng) (tng) (tng) (tng) (tng) Criterion ?/span> Criteria 的查询条件?/span>
Criteria 提供?/span> add(Criterion criterion) Ҏ(gu)来添加查询条件。图 2 ?/span> Criterion 的结构图?/span> Criterion 接口的主要实现包括：(x) Example ?/span> Junction ?/span> SimpleExpression 。?/span> Junction 的实际用是它的两个子类 conjunction ?/span> disjunction Q分别是使用 AND ?/span> OR 操作W进行来联结查询条g集合?/span>
Criterion 的实例可以通过 Restrictions 工具cL创徏Q?/span> Restrictions 提供?jin)大量的静(rn)态方法，?/span> eq Q等于）(j)?/span> ge Q大于等于）(j)?/span> between {来Ҏ(gu)的创?/span> Criterion 查询条g
Q?/span> SimpleExpression 实例Q。除此之外， Restrictions q提供了(jin)Ҏ(gu)来创?/span> conjunction ?/span> disjunction 实例Q通过往该实例的 add(Criteria) Ҏ(gu)来增加查询条件Ş成一个查询条仉合?/span>
至于 Example 的创建有所不同Q?/span> Example 本n提供?jin)一个静(rn)态方?/span> create(Object entity) Q即Ҏ(gu)一个对象（实际使用中一般是映射到数据库的对象）(j)来创建。然后可以设|一些过滤条Ӟ(x)
Example exampleUser =Example.create(u)
.ignoreCase() // 忽略大小?/span>
.enableLike(MatchMode.ANYWHERE);
// ?/span> String cd的属性，无论在那里值在那里都匹配。相当于 %value%
?/span>
2
(tng) (tng) (tng) (tng) (tng) (tng)
Project 主要是让 Criteria 能够q行报表查询Qƈ可以实现分组?/span> Project 主要?/span> SimpleProjection ?/span> ProjectionList ?/span> Property 三个实现。其?/span> SimpleProjection ?/span> ProjectionList 的实例化是由内徏?/span> Projections 来完成，如提供的 avg ?/span> count ?/span> max ?/span> min ?/span> sum 可以让开发者很Ҏ(gu)Ҏ(gu)个字D进行统计查询?/span>
(tng) (tng) (tng) (tng) (tng) (tng) Property 是对某个字段q行查询条g的设|，如通过
(tng) (tng) (tng) (tng) (tng) Porperty.forName(“color?.in(new String[]{“black?”red?”write”}); 则可以创Z?/span> Project 实例。通过 criteria ?/span> add(Project) Ҏ(gu)加入到查询条件中厅R?/span>
?/span>
3
(tng) (tng) (tng) (tng) (tng) (tng)
(tng) (tng) (tng) (tng) (tng) (tng) 使用 Criteria q行查询Q主要要清晰的是 Hibernate 提供?jin)那些类和方法来满开发中查询条g的创建和l装Q其l构层次如何。这样用v来便可得?j)应手?/span>
(tng)]]>
javax.servlet.ServletException: Cannot retrieve mapping for action /companyNews
struts-config.xml中没有写相关companyNews的action.
============================================
在myeclipse增加一个jar包时候，先打开目properties,选择java build path --> libraries -->add external JARs 选择到需要的jar包，加入后，发现Q他q个包加载的是绝对\径，而我们的目需要cvs׃nQ无?br />
commit到cvs服务器。myeclipse有以下提CZ息：(x)
2 build path entries are missing.
解决办法是，关闭myeclipse,用notepad打开目目录下的 .classpath, 手动修改成相对\径，q且(g)查，指定的相对\径中是否真正d?jin)jar包，
启动myeclipse可以上传新d的jar包了(jin)?br />
=============================================
在用junit/StrutsTest时候，报错Q?br />java.lang.UnsupportedClassVersionError: junit/framework/TestListener (Unsupported major.minor version 49.0)
原因是如果是jdk1.4的话Q只能用junit3.8以下版本Q如果是使用jdk1.5的话Q必M用junit4.0以上版本?br />
==========================================================
Cannot find ActionMappings or ActionformBeans collection
原因是：(x)web.xml文g中没有配|struts-config.xml的相关信息?br />
============================================================
org.apache.jasper.JasperException: The absolute uri: http://java.sun.com/jstl/core cannot be resolved in either web.xml or the jar files deployed with this application
~少jstl的相关jar和web.xml配置
jstl.jar
(tng)
(tng)
(tng)
(tng)
(tng)
(tng)
(tng)
(tng)
(tng)
(tng)
(tng)
(tng)
(tng)
(tng)
(tng)
(tng)
==========================================================
Failed to load or instantiate TagLibraryValidator class: org.apache.taglibs.standard.tlv.JstlCoreTLV
standard.jar没有攑֜lib里面
===========================================================
创徏oracle表的时候，使用?jin)，role和comment关键字，不要这两个关键字作名和字段名?br />用toad或plsql developer创徏表，q样软g?x)提C关键字?br />=============================================================
[ERROR] XMLHelper - Error parsing XML: XML InputStream(18) Attribute name "column" associated with an element type "key" must be followed by the ' = ' character.
[ERROR] Configuration - Could not configure datastore from input stream
must be followed by the ' = ' character. Nested exception: Attribute name "column" associated with an element type "key" must be followed by the ' = ' character.>org.dom4j.DocumentException: Error
on line 18 of document : Attribute name "column" associated with an element type "key" must be followed by the ' = ' character. Nested exception: Attribute name "column" associated with an element
type "key" must be followed by the ' = ' character.
xml 语法错误Qkey语法中应该类DL(fng)写法
=============================================================
hibernate3,对象中一对多的one方，必须写private Set pays=new HashSet();Q否则包错java.lang.NullPointerException Q?br />而hibernate2中private Set pays;却不报错Q?br />
=============================================================
[DEBUG] AbstractSaveEventListener - generated identifier: 1, using strategy: org.hibernate.id.IncrementGenerator
org.hibernate.PropertyValueException: not-null property references a null or transient value: com.xxx.yyy.company
at org.hibernate.engine.Nullability.checkNullability(Nullability.java:72)
================================================================
org.hibernate.TransientObjectException: object references an unsaved transient instance - save the transient instance before flushing: com.xxx.yyy.Company
在save的同时也需要save其他的表Q然后再flush()
=========================================================================
Parse Fatal Error at line 12 column 1
struts-config.xml文g被修改，语法错误。检查语法?br />
========================================
org.hibernate.QueryException: could not resolve property: userid of: com.xxx.yyy.Pay
使用到外键userid的时候，必须使用userinfo.useridҎ(gu)才能得到?br />
=========================================
javax.naming.NameNotFoundException: Name hibernate_connection_factory is not bound in this Context
原因Qhibernate的数据库映射.xml文g有配|错误，Dhibernate_connection_factory无法l定数据库?br />例如many-to-one讄?jin)以后，仍然在其中设|相冲突?property>属性?br />
==========================================
[WARN] RequestProcessor - Unhandled Exception thrown: class java.lang.NullPointerException
必须用到的对象newh?br />
============================================
GROUP BY 表达式的查询必须满如下Q?br />select 子句后的每一必需出现在group by 子句中，除非该项使用?jin)聚集函数?br />
===========================================
org.hibernate.TransientObjectException: object references an unsaved transient instance - save the transient instance before flushing: com.xxx.yyy.Company
要联保存，多次session.save()
==============================================
java.lang.IllegalArgumentException: id to load is required for loading
原因Q?br />
session.load(Company.class,payForm.getCompanyId());
load()Ҏ(gu)W二个参数必Lsearlizable,q且必须是和数据库映类的属性值类型一_(d)即强制转换都不行?br />
=============================================
[INFO] DefaultLoadEventListener - Error performing load command
>org.hibernate.ObjectNotFoundException: No row with the given identifier exists: [com.xxx.yyy.Company#0]
表示你现在查询的对象所兌的对象有问题Q一般是因ؓ(f)数据的问题（该对象所兌的对象找不到Q?数据的错误，影响?jin)程序正常执行?br />
========================================================
eclipse 3.1,myeclipse 4错误
Deployment is out of date due to changes in the underlying project contents . You'll need to mannally 'Redeploy' the project to update the deployed archive.
原因
tomcat 中部|的某个文g的拒l访问媄(jing)响了(jin)部v?br />重启后，去除tomcat中部|的文g。重新在eclipse中设|部|Ӏ?br />Z么需要重启？
因ؓ(f)Qgoogle desktop软g正在Ҏ(gu)部v的一?00MB大文件进行烦(ch)引，锁定?jin)这个大文gQ我估计google desktop需要对q个文g索引半个时以上。因此，eclipse无法对过去部|的文gQ做先删除后重新部v的工作?br />
===================================================
ERROR LazyInitializationException:19 - could not initialize proxy - the owning Session was closed
org.hibernate.LazyInitializationException: could not initialize proxy - the owning Session was closed
解决办法?br />cmpy=(CompanyEdit)ss.load(CompanyEdit.class,companyId1);
(tng) Hibernate.initialize(cmpy);//强制初始化cmpy,否则ss.close()?cmpy消?
====================================================
javascript错误
? 56
字符: 45
错误: 未结束的字符串常?br />代码: 0
是编码的问题Q用ANSI~码另存后就好了(jin)?br />
====================================================
[WARN] JDBCExceptionReporter - SQL Error: 904, SQLState: 42000
[ERROR] JDBCExceptionReporter - ORA-00904: 无效列名
[INFO] DefaultLoadEventListener - Error performing load command
>org.hibernate.exception.SQLGrammarException: could not load an entity: [com.xxx.yyy.Sellinfo#1]
***.hbm.xml文g中的某个列名和数据库中的不同?br />
=======================================================
ConnectionManager - unclosed connection, forgot to call close() on your session?
原因Q没有关闭h(hun)ibernate的session的transaction。或者没有关闭session
=======================================================
[WARN] SellCommentDAO - org.hibernate.ObjectDeletedException: deleted object would be re-saved by cascade (remove deleted object from associations): [com.xxx.yyy.SellComment#7]
原因Q父亲对象（one方）(j)讄cascade="save-update" Ӟ直接删除子对象时Q会(x)报错Q?br />处理Ҏ(gu)Qsave信息需要联操作，delete时候也要用相同的原理?br />SellComment sellComment=new SellComment();
sellComment = (SellComment)session.load(SellComment.class,sellCommentId);//获取儿子对象
Long sellInfoId=sellComment.getSellInfo().getSellId();//获取父亲id
SellInfo sellinfo=(SellInfo)session.load(SellInfo.class,sellInfoId);//获取父亲对象
sellInfo.getSellComments().remove(sellComment);//断绝父子关系
sellComment.setSellInfo(null);//断绝子父关系
session.delete(sellComment);//删除儿子
session.flush();
]]>
选自<<_NHibernateQJava对象持久化技术详?gt;> (tng)作者：(x)孙卫琾b来?www.javathinker.org
如果转蝲Q请标明出处Q谢?br />
延迟初始化错误是q用Hibernate开发项目时最常见的错误。如果对一个类或者集合配|了(jin)延迟(g)索策略，那么必须当代理类实例或代理集合处于持久化状态（卛_于Session范围内）(j)Ӟ才能初始化它。如果在游离状态时才初始化它，׃(x)产生延迟初始化错误?br />
下面把Customer.hbm.xml文g?lt;class>元素的lazy属性设为trueQ表CZ用gq检索策略：(x)
<class (tng)name="mypack.Customer" (tng)table="CUSTOMERS" (tng)lazy="true">
当执行Session的load()Ҏ(gu)ӞHibernate不会(x)立即执行查询CUSTOMERS表的select语句Q仅仅返回Customercȝ代理cȝ实例Q这个代理类L(fng)以下特征Q?br />
Q?Q?tng)由Hibernate在运行时动态生成，它扩展了(jin)Customerc，因此它承了(jin)Customercȝ所有属性和Ҏ(gu)Q但它的实现对于应用E序是透明的?br />Q?Q?tng)当Hibernate创徏Customer代理cd例时Q仅仅初始化?jin)它的OID属性，其他属性都为nullQ因此这个代理类实例占用的内存很?br />Q?Q当应用E序W一ơ访问Customer代理cd例时Q例如调用customer.getXXX()或customer.setXXX()Ҏ(gu)Q， (tng)Hibernate?x)初始化代理cd例，在初始化q程中执行select语句Q真正从数据库中加蝲Customer对象的所有数据。但有个例外Q那是当应用程序访问Customer代理cd例的getId()Ҏ(gu)ӞHibernate不会(x)初始化代理类实例Q因为在创徏代理cd例时OID存在了(jin)Q不必到数据库中L询?br />
提示QHibernate采用CGLIB工具来生成持久化cȝ代理cRCGLIB是一个功能强大的Java字节码生成工P它能够在E序q行时动态生成扩展?tng)JavacL者实现Java接口的代理类。关于CGLIB的更多知识，请参考：(x)http://cglib.sourceforge.net/?br />
以下代码先通过Session的load()Ҏ(gu)加蝲Customer对象Q然后访问它的name属性：(x) (tng)
tx (tng)= (tng)session.beginTransaction();
Customer (tng)customer=(Customer)session.load(Customer.class,new (tng)Long(1));
customer.getName();
tx.commit();
在运行session.load()Ҏ(gu)时Hibernate不执行Q何select语句Q仅仅返回Customercȝ代理cȝ实例Q它的OID?Q这是由load()Ҏ(gu)的第二个参数指定的。当应用E序调用customer.getName()Ҏ(gu)ӞHibernate?x)初始化Customer代理cd例，从数据库中加载Customer对象的数据，执行以下select语句Q?br />
select (tng)* (tng)from (tng)CUSTOMERS (tng)where (tng)I(yng)D=1;
select (tng)* (tng)from (tng)ORDERS (tng)where (tng)CUSTOMER_ID=1;
?lt;class>元素的lazy属性ؓ(f)trueQ会(x)影响Session的load()Ҏ(gu)的各U运行时行ؓ(f)Q下面D例说明?br />
1Q如果加载的Customer对象在数据库中不存在QSession的load()Ҏ(gu)不会(x)抛出异常Q只有当q行customer.getName()Ҏ(gu)时才?x)抛Z下异常：(x)
ERROR (tng)LazyInitializer:63 (tng)- (tng)Exception (tng)initializing (tng)proxy
net.sf.hibernate.ObjectNotFoundException: (tng)No (tng)row (tng)with (tng)the (tng)given (tng)identifier (tng)exists: (tng)1, (tng)of (tng)class: (tng)
mypack.Customer
2Q如果在整个Session范围内，应用E序没有讉KqCustomer对象Q那么Customer代理cȝ实例一直不?x)被初始化，Hibernate不会(x)执行Mselect语句。以下代码试囑֜关闭Session后访问Customer游离对象Q?br />
tx (tng)= (tng)session.beginTransaction();
Customer (tng)customer=(Customer)session.load(Customer.class,new (tng)Long(1));
tx.commit();
session.close();
customer.getName();
׃引用变量customer引用的Customer代理cȝ实例在Session范围内始l没有被初始化，因此在执行customer.getName()Ҏ(gu)ӞHibernate?x)抛Z下异常：(x)
ERROR (tng)LazyInitializer:63 (tng)- (tng)Exception (tng)initializing (tng)proxy
net.sf.hibernate.HibernateException: (tng)Could (tng)not (tng)initialize (tng)proxy (tng)- (tng)the (tng)owning (tng)Session (tng)was (tng)closed
由此可见QCustomer代理cȝ实例只有在当前Session范围内才能被初始化?br />
3Qnet.sf.hibernate.Hibernatecȝinitialize()?rn)态方法用于在Session范围内显式初始化代理cd例，isInitialized()Ҏ(gu)用于判断代理cd例是否已l被初始化。例如：(x)
tx (tng)= (tng)session.beginTransaction();
Customer (tng)customer=(Customer)session.load(Customer.class,new (tng)Long(1));
if(!Hibernate.isInitialized(customer)) (tng)
Hibernate.initialize(customer);
tx.commit();
session.close();
customer.getName();
以上代码在Session范围内通过Hibernatecȝinitialize()Ҏ(gu)昑ּ初始化了(jin)Customer代理cd例，因此当Session关闭后，可以正常讉KCustomer游离对象?br />
4Q当应用E序讉K代理cd例的getId()Ҏ(gu)Ӟ不会(x)触发Hibernate初始化代理类实例的行为，例如Q?br />
tx (tng)= (tng)session.beginTransaction();
Customer (tng)customer=(Customer)session.load(Customer.class,new (tng)Long(1));
customer.getId();
tx.commit();
session.close();
customer.getName();
当应用程序访问customer.getId()Ҏ(gu)Ӟ该方法直接返回Customer代理cd例的OID|无需查询数据库。由于引用变量?tng)customer始终引用的是没有被初始化的Customer代理cd例，因此当Session关闭后再执行customer.getName()Ҏ(gu)Q?tng)Hibernate?x)抛Z下异常：(x)
ERROR (tng)LazyInitializer:63 (tng)- (tng)Exception (tng)initializing (tng)proxy
net.sf.hibernate.HibernateException: (tng)Could (tng)not (tng)initialize (tng)proxy (tng)- (tng)the (tng)owning (tng)Session (tng)was (tng)closed
]]>
]]>
Customer customer=(Customer)session.load(Customer.class,new Long(1));
if(!Hibernate.isInitialized(customer))
Hibernate.initialize(customer);
tx.commit();
session.close();
customer.getName();
以上代码在Session范围内通过Hibernatecȝinitialize()Ҏ(gu)昑ּ初始化了(jin)Customer代理cd例，因此当Session关闭后，可以正常讉KCustomer游离对象?/font>
]]>
Distributed Objects With Hibernate
Problem:
You are building a three tier distributed system and you are passing persistent objects between tiers. When you send the objects back to the server, you want to save the only the changes.
Solutions:
1)Hibernate added the select-before-update option in 2.1. Setting this option on your class will cause Hibernate to load the data from the database and compare that to the data in your object to determine what has changed. It makes use of a version or timestamp field to enforce optimistic concurrency. This works well but introduces additional database access that could effect performance.
2)Save the original data for each persistent object and implement a custom EntityPersister to return the original state to Hibernate instead of accessing the state in the database.
Implementing Solution 2:
Step 1: Save the original state for each persistent object
The easiest way to implement this is to have each persistent object remember the original values of any properties that changed. If you make all your persistent objects proper JavaBeans, you can capture this in the base class when the object fires a propertyChanged event:
public abstract class AbstractDomainObject implements Serializable {
private PropertyChangeSupport support = new PropertyChangeSupport(this);
private Map originalValues = new HashMap();
public AbstractDomainObject() {
super();
}
public Map getOriginalState() {
return originalValues;
}
public void addPropertyChangeListener(PropertyChangeListener propertyChangeListener) {
support.addPropertyChangeListener(propertyChangeListener);
}
protected void firePropertyChange(String property, Object oldValue, Object newValue) {
if (!originalValues.containsKey(property)) {
originalValues.put(property, oldValue);
}
support.firePropertyChange(property, oldValue, newValue);
}
}
Step 2: Create your domain objects
A domain object might look like follows. Because it fires propertyChange events, the base class will remember original values of properties that changed. Note that you will want to map the properties as access=field since the setter fires a propertyChange event when called.
public class Customer extends AbstractDomainObject {
private String name;
public String getName() {
return name;
}
public void setName(String name) {
String oldValue = this.name;
this.name = name;
firePropertyChange("name", oldValue, name);
}
}
Step 3: Create a custome entity persister
Create a custom entity persister and override getCurrentPersistentState. Set the persister for each class to this custom persister.
public Object[] getCurrentPersistentState(Serializable id, Object version, SessionImplementor session) throws HibernateException {
AbstractDomainObject entity = (AbstractDomainObject) session.getEntity(new Key(id, this));
Map originalValues = entity.getOriginalState();
Type[] types = getPropertyTypes();
String[] propertyNames = getPropertyNames();
Object[] values = new Object[propertyNames.length];
boolean[] includeProperty = getPropertyUpdateability();
for (int i = 0; i < propertyNames.length; i++) {
if (includeProperty[i]) {
if (originalValues.containsKey(propertyNames[i])) {
values[i] = types[i].disassemble(originalValues.get(propertyNames[i]), session);
} else {
values[i] = types[i].disassemble(getPropertyValue(entity, propertyNames[i]), session);
}
}
}
return values;
}
What about collections you ask? Well, Hibernate keeps track of changes to collections in the collection class wrappers for you.
Please note this solution has worked great in my example programs, but I have not yet implemented it in a production program.
]]>
Hibernate配置文g可以有两U格式，一U是hibernate.propertiesQ另一U是hibernate.cfg.xml。后者稍微方便一些，当增加hbm映射文g的时候，可以直接在hibernate.cfg.xml里面增加Q不必像hibernate.properties必须在初始化代码中加入。但不管怎么_(d)两种的配|项都是一L(fng)Q下面详l介l：(x)
在Hibernate的src目录下有一个hibernate.properties模板Q我们不必自׃头写Q修Ҏ(gu)板就可以?/p>
代码:
hibernate.query.substitutions true 1, false 0, yes 'Y', no 'N'
q个配置意思是当你?Hibernate里面输入true的时候，Hibernate?x){化ؓ(f)0插入数据库，当你在Hibernate里面输入false的时候， Hibernate?x){化ؓ(f)1插入数据库，后面的YQN同理。对于某些数据库Q例如Oracle来说Q没有boolean数据cdQ就是采?代表 trueQ?代表falseQ因此用这个配|在Hibernate里面直接用true/false?x)非常直观?
代码:
hibernate.dialect net.sf.hibernate.dialect.MySQLDialect
hibernate.connection.driver_class org.gjt.mm.mysql.Driver
hibernate.connection.driver_class com.mysql.jdbc.Driver
hibernate.connection.url jdbc:mysql:///test
hibernate.connection.username root
hibernate.connection.password
q是一个连接MySQL数据库的例子Q很直观Q不必解释，不同的数据库的连接参数模板中全部l出?jin)?
代码:
hibernate.connection.pool_size 1
hibernate.statement_cache.size 25
q是Hibernate自带的连接池的配|参敎ͼ在默认情况下采用。意义很直观Q不多解释。只是提醒一点，Hibernateq个q接池是非常原始非常单的q接池，如果你在目中用Hibernate的话Q徏议你首选App Server的连接池Q次选Hibernate带的DBCPq接池。自带的q接池应该做为末选?
如果你采用DBCPq接池，除了(jin)要配|DBCPq接池以外，q需要取消掉下行的注释：(x)
代码:
hibernate.connection.provider_class net.sf.hibernate.connection.DBCPConnectionProvider
其它的连接池同理?
如果采用App Server的连接池Q假设App Serverq接池的DataSource的JNDI名称?mypool"的话Q配|应该如下：(x)
代码:
hibernate.dialect net.sf.hibernate.dialect.MySQLDialect
hibernate.connection.datasource mypool
hibernate.connection.provider_class net.sf.hibernate.connection.DatasourceConnectionProvider
其它参数׃必写?jin)，因?f)已经在App Server配置q接池的时候指定好?jin)?
如果你不是在App Server环境中用HibernateQ例如远E客L(fng)E序Q但是你又想用App Server的数据库q接池，那么你还需要配|JNDI的参敎ͼ例如Hibernateq接q程Weblogic上的数据库连接池Q?
代码:
hibernate.dialect net.sf.hibernate.dialect.MySQLDialect
hibernate.connection.datasource mypool
hibernate.connection.provider_class net.sf.hibernate.connection.DatasourceConnectionProvider
hibernate.jndi.class weblogic.jndi.WLInitialContextFactory
hibernate.jndi.url t3://servername:7001/
最后，如果你需要在EJB或者JTA中用HibernateQ需要取消下行的注释Q?
代码:
hibernate.transaction.factory_class net.sf.hibernate.transaction.JTATransactionFactory
杂项配置Q?
代码:
hibernate.show_sql false
是否Hibernate发送给数据库的sql昄出来Q这是一个非帔R常有用处的功能。当你在调试Hibernate的时候，让Hibernate打印sql语句Q可以帮助你q速解决问题?
代码:
#hibernate.connection.isolation 4
指定数据库的隔离U别Q往往不同的数据库有自己定义的隔离U别Q未必是Hibernate的设|所能更改的Q所以也不必ȝ它了(jin)?
代码:
hibernate.jdbc.fetch_size 50
hibernate.jdbc.batch_size 25
q两个选项非常非常非常重要Q！Q将严重影响Hibernate的CRUD性能!
: C = create, R = read, U = update, D = delete
Fetch Size 是设定JDBC的Statementd数据的时候每ơ从数据库中取出的记录条数。例如一ơ查?万条记录Q对于Oracle的JDBC驱动来说Q是不会(x)1ơ性把1万条取出来的Q而只?x)取出Fetch Size条数Q当U录集遍历完?jin)这些记录以后，再去数据库取Fetch Size条数据。因此大大节省了(jin)无谓的内存消耗。当然Fetch Size讄大Q读数据库的ơ数少Q速度快QFetch Size小Q读数据库的ơ数多Q速度慢。这有点像^时我们写E序写硬盘文件一P讄一个BufferQ每ơ写入BufferQ等Buffer满了(jin)以后Q一ơ写入硬盘，道理相同?
Oracle数据库的JDBC驱动默认的Fetch Size=10Q是一个非怿守的讑֮Q根据我的测试，当Fetch Size=50的时候，性能?x)提?倍之多，当Fetch Size=100Q性能q能l箋(hu)提升20%QFetch Sizel箋(hu)增大Q性能提升的就不显著了(jin)。因此我使用Oracle的一定要?/span>Fetch Size讑ֈ50?
不过q不是所有的数据库都支持Fetch SizeҎ(gu)，例如MySQL׃支持。MySQL像我上面说的那U最坏的情况Q他L一下就?万条记录完全取出来，内存消耗会(x)非常非常惊hQ这个情况就没有什么好办法?/span> (tng)
Batch Size是设定对数据库进行批量删除，扚w更新和批量插入的时候的Ҏ(gu)大小Q有点相当于讄Buffer~冲区大的意思。Batch Size大Q批量操作的向数据库发送sql的次数越，速度p快。我做的一个测试结果是当Batch Size=0的时候，使用Hibernate对Oracle数据库删?万条记录需?5U，Batch Size = 50的时候，删除仅仅需?U！Q！可见有多么大的性能提升Q很多h做Hibernate和JDBC的插入性能试?x)奇怪的发现Hibernate速度臛_是JDBC的两倍，是因ؓ(f)Hibernate使用?jin)Batch InsertQ而他们写的JDBC没有使用Batch的缘故。以我的l验来看Q?span style="COLOR: red">Oracle数据?Batch Size = 30 的时候比较合适，50也不错，性能?x)l提?/span>Q?0以上Q性能提升的非常微弱，反而消耗内存更加多Q就没有必要?jin)?
代码:
#hibernate.jdbc.use_scrollable_resultset true
讑֮是否可以使用JDBC2.0规范的可滚动l果集，q对Hibernate的分|C有一定的作用Q默认就好了(jin)?
代码:
#hibernate.cglib.use_reflection_optimizer false
默认打开Q启用cglib反射优化。cglib是用来在Hibernate中动态生成PO字节码的Q打开优化可以加快字节码构造的速度?
不过Q当你在调试E序q程中，特别是和proxyQlazy loading相关的应用中Q代码出错，但是出错提示信息有语焉不详，那么你可以把cglib优化x(chng)Q这样Hibernate?x)输出比较详l的调试信息Q帮助你debug?/p>
(1)API的变化，它将影响到JavaE序代码?
(2)元数据，它将影响到对?关系映射文g?
(3)HQL查询语句?
值得注意的是Q?Hibernate3.0q不?x)完全取代Hibernate2.1。在同一个应用程序中Q允许Hibernate3.0和Hibernate2.1q存?
1.1 Hibernate API 变化
1.1.1 包名
Hibernate3.0的包的根路径? “org.hibernate?Q而在Hibernate2.1中“net.sf.hibernate”。这一命名变化使得Hibernate2.1和Hibernate3.0能够同时在同一个应用程序中q行?
如果希望把已有的应用升到Hibernate3.0Q那么升U的W一步是把Java源程序中的所有“net.sf.hibernate”替换ؓ(f)“org.hibernate”?
Hibernate2.1中的“net.sf.hibernate.expression”包被改名ؓ(f)“org.hibernate.criterion”。假如应用程序用了(jin)Criteria APIQ那么在升的过E中Q必LJava源程序中的所有“net.sf.hibernate.expression”替换ؓ(f)“org.hibernate.criterion”?
如果应用使用?jin)除Hibernate以外的其他外部YӞ而这个外部Y件又引用?jin)Hibernate的接口，那么在升U时必须十分心(j)。例如EHCache拥有自己的CacheProviderQ?net.sf.ehcache.hibernate.ProviderQ在q个cM引用?jin)Hibernate2.1中的接口Q在升应用Ӟ可以采用以下办法之一来升UEHCache:
(1)手工修改net.sf.ehcache.hibernate.Providerc，使它引用Hibernate3.0中的接口?
(2){到EHCache软g本n升Z用Hibernate3.0后，使用新的EHCache软g?
(3)使用Hibernate3.0中内|的CacheProviderQorg.hibernate.cache.EhCacheProvider?
1.1.2 org.hibernate.classic?
Hibernate3.0把一些被废弃的接口都转移到org.hibernate.classic中?
1.1.3 Hibernate所依赖的第三方软g?
在Hibernate3.0的Y件包的lib目录下的README.txt文g中，描述?jin)Hibernate3.0所依赖的第三方软g包的变化?
1.1.4 异常模型
在Hibernate3.0中，HibernateException异常以及(qing)它的所有子c都l承?jin)java.lang.RuntimeException。因此在~译Ӟ~译器不?x)再(g)查HibernateException?
1.1.5 Session接口
在Hibernate3.0中，原来Hibernate2.1的Session接口中的有些基本Ҏ(gu)也被废弃Q但Z(jin)化升U，q些Ҏ(gu)依然是可用的Q可以通过org.hibernate.classic.Session子接口来讉K它们Q例如：(x)
org.hibernate.classic.Session session=sessionFactory.openSession();
session.delete("delete from Customer ");
在Hibernate3.0中，org.hibernate.classic.Session接口l承?jin)org.hibernate.Session接口Q在org.hibernate.classic.Session接口中包含了(jin)一pd被废弃的Ҏ(gu)Q如find()、interate(){。SessionFactory接口的openSession()Ҏ(gu)q回org.hibernate.classic.Sessioncd的实例。如果希望在E序中完全用Hibernate3.0Q可以采用以下方式创建Session实例Q?
org.hibernate.Session session=sessionFactory.openSession();
如果是对已有的程序进行简单的升Qƈ且希望仍然调用Hibernate2.1中Session的一些接口，可以采用以下方式创徏Session实例Q?
org.hibernate.classic.Session session=sessionFactory.openSession();
在Hibernate3.0中，Session接口中被废弃的方法包括：(x)
* 执行查询的方法：(x)find()、iterate()、filter()和delete(String hqlSelectQuery)
* saveOrUpdateCopy()
Hibernate3.0一律采用createQuery()Ҏ(gu)来执行所有的查询语句Q采用DELETE 查询语句来执行批量删除，采用merge()Ҏ(gu)来替?saveOrUpdateCopy()Ҏ(gu)?
提示Q在Hibernate2.1中，Session的delete()Ҏ(gu)有几U重载Ş式，其中参数为HQL查询语句的delete()Ҏ(gu)在Hibernate3.0中被废弃Q而参Cؓ(f)Ojbectcd的的delete()Ҏ(gu)依然被支持。delete(Object o)Ҏ(gu)用于删除参数指定的对象，该方法支持联删除?
Hibernate2.1没有Ҏ(gu)量更新和扚w删除提供很好的支持，参见<<_NHibernate>>一书的W?3章的13.1.1节（扚w更新和批量删除）(j)Q而Hibernate3.0Ҏ(gu)量更新和扚w删除提供?jin)支持，能够直接执行扚w更新或批量删除语句，无需把被更新或删除的对象先加载到内存?sh)。以下是通过Hibernate3.0执行扚w更新的程序代码：(x)
Session session = sessionFactory.openSession();
Transaction tx = session.beginTransaction();
String hqlUpdate = "update Customer set name = :newName where name =
ldName"; int updatedEntities = s.createQuery( hqlUpdate )
.setString( "newName", newName )
.setString( "oldName", oldName )
.executeUpdate();
tx.commit();
session.close();
以下是通过Hibernate3.0执行扚w删除的程序代码：(x)
Session session = sessionFactory.openSession();
Transaction tx = session.beginTransaction();
String hqlDelete = "delete Customer where name =
ldName"; int deletedEntities = s.createQuery( hqlDelete )
.setString( "oldName", oldName )
.executeUpdate();
tx.commit();
session.close();
1.1.6 createSQLQuery()
在Hibernate3.0中，Session接口的createSQLQuery()Ҏ(gu)被废弃，被移到org.hibernate.classic.Session接口中。Hibernate3.0采用新的SQLQuery接口来完成相同的功能?
1.1.7 Lifecycle ?Validatable 接口
Lifecycle和Validatable 接口被废弃，q且被移到org.hibernate.classic包中?
1.1.8 Interceptor接口
在Interceptor 接口中加入了(jin)两个新的Ҏ(gu)?用户创徏的Interceptor实现cd升的过E中Q需要ؓ(f)q两个新Ҏ(gu)提供Ҏ(gu)体ؓ(f)I的实现。此外，instantiate()Ҏ(gu)的参C?jin)修改，isUnsaved()Ҏ(gu)被改名ؓ(f)isTransient()?
1.1.9 UserType和CompositeUserType接口
在UserType和CompositeUserType接口中都加入?jin)一些新的方法，q两个接口被Udorg.hibernate.usertype包中Q用户定义的UserType和CompositeUserType实现cdd现这些新Ҏ(gu)?
Hibernate3.0提供?jin)ParameterizedType接口Q用于更好的重用用户自定义的cd?
1.1.10 FetchModec?
FetchMode.LAZY ?FetchMode.EAGER被废弃。取而代之的分别为FetchMode.SELECT 和FetchMode.JOIN?
1.1.11 PersistentEnumc?
PersistentEnum被废弃ƈ删除。已l存在的应用应该采用UserType来处理枚丄型?
1.1.12 对Blob 和Clob的支?
Hibernate对Blob和Clob实例q行?jin)包装，使得那些拥有Blob或Clobcd的属性的cȝ实例可以被游R序列化或反序列化，以及(qing)传递到merge()Ҏ(gu)中?
1.1.13 Hibernate中供扩展的API的变?
org.hibernate.criterion?org.hibernate.mapping?org.hibernate.persister和org.hibernate.collection 包的l构和实现发生了(jin)重大的变化。多数基于Hibernate
2.1 的应用不依赖于这些包Q因此不?x)被影响。如果你的应用扩展了(jin)q些包中的类Q那么必非常小?j)的对受影响的程序代码进行升U?
1.2 元数据的变化
1.2.1 (g)索策?
在Hibernate2.1中，lazy属性的默认gؓ(f)“false”，而在Hibernate3.0中，lazy属性的默认gؓ(f)“true”。在升映射文gӞ如果原来的映文件中的有兛_素，?lt;set>?lt;class>{没有显式设|lazy属性，那么必须把它们都昑ּ的设|ؓ(f)lazy=“true”。如果觉得这U升U方式很ȝ(ch)Q可以采取另一单的升方式Q在<hibernate-mapping>元素中设|? default-lazy=“false”?
1.2.2 对象标识W的映射
unsaved-value属性是可选的Q在多数情况下，Hibernate3.0把unsaved-value="0" 作ؓ(f)默认倹{?
在Hibernate3.0中，当用自然主键和游离对象Ӟ不再实现Interceptor.isUnsaved()Ҏ(gu)?如果没有讄q个Ҏ(gu)Q当Hibernate3.0无法区分对象的状态时Q会(x)查询数据库，来判断这个对象到底是临时对象Q还是游d象。不q，昑ּ的用Interceptor.isUnsaved()Ҏ(gu)?x)获得更好的性能Q因可以减少Hibernate直接讉K数据库的ơ数?
1.2.3 集合映射
<index>元素在某些情况下?lt;list-index>?lt;map-key>元素替代。此外，Hibernate3.0?lt;map-key-many-to-many> 元素来替代原来的<key-many-to-many>.元素Q用<composite-map-key>元素来替代原来的<composite-index>元素?
1.2.4 DTD
对象-关系映射文g中的DTD文档Q由原来的：(x)
http://hibernate.sourceforge.net/hi...mapping-2.0.dtd
改ؓ(f)Q?
http://hibernate.sourceforge.net/hi...mapping-3.0.dtd
1.3 查询语句的变?
Hibernate3.0 采用新的ZANTLR的HQL/SQL查询译器，不过QHibernate2.1的查询翻译器也依然存在。在Hibernate的配|文件中Qhibernate.query.factory_class属性用来选择查询译器。例如：(x)
Q?Q选择Hibernate3.0的查询翻译器Q?
hibernate.query.factory_class= org.hibernate.hql.ast.ASTQueryTranslatorFactory
Q?Q选择Hibernate2.1的查询翻译器
hibernate.query.factory_class= org.hibernate.hql.classic.ClassicQueryTranslatorFactory
提示QANTLR是用UJava语言~写出来的一个编译工P它可生成Java语言或者是C++的词法和语法分析器，q可产生语法分析?wi)ƈ对该树(wi)进行遍历。ANTLR׃是纯Java的，因此可以安装在Q意^CQ但是需要JDK的支持?
Hibernate开发小l尽力保证Hibernate3.0的查询翻译器能够支持Hibernate2.1的所有查询语句。不q，对于许多已经存在的应用，在升U过E中Q也不妨仍然使用Hibernate2.1的查询翻译器?
值得注意的是Q?Hibernate3.0的查询翻译器存在一个BugQ不支持某些theta-styleq结查询方言Q如Oracle8i的OracleDialect方言、Sybase11Dialect。解册一问题的办法有两种Q（1Q改Z用支持ANSI-styleq结查询的方aQ如 Oracle9Dialect,Q?Q如果升U的时候遇到这一问题Q那么还是改Z用Hibernate2.1的查询翻译器?
1.3.1 indices()和elements()函数
在HQL的select子句中废弃了(jin)indices()和elements()函数Q因两个函数的语法很让用戯解，可以用显式的q接查询语句来替?select elements(...) 。而在HQL的where子句中，仍然可以使用elements()函数
hibernate2 升为hibernate3的需要注意的?br />
1.首先hibernate2.jar替换为hibernate3.jar(hibernate-3.0.5)
hibernate-tools.jar也替换成新的(从hibernate-tools-3.0.0.alpha4a扑և来的)
2.所有程序中的net.sf.hibernate替换为org.hibernate.
3.但是有例?br />net.sf.hibernate.expression.Expression换ؓ(f)org.hibernate.criterion.Expression
如果用eclipse,用ctrl+shift+o快捷键可以加快速度
4.在用hql查询时将
createSQLQuery(hql,"c",EZCampaignDTO.class);改ؓ(f)createSQLQuery(hql).addEntity("c",EZCampaignDTO.class);
5.在批量插入时
原来的int size = ((SessionFactoryImpl)(session.getSessionFactory())).getJdbcBatchSize()
改ؓ(f)int size = ((SessionFactoryImpl)(session.getSessionFactory())).getSettings().getJdbcBatchSize();
6.在计count?br />原来的int size = ((Integer) session.iterate(hql).next()).intValue();
改ؓ(f)int size = ((Integer) session.createQuery(hql).iterate().next()).intValue();
其中h(hun)ql="select count(*) from " + DAOVar.contactClass;
7.q有是?hbm中的hibernate-mapping-2.0.dtd替换为hibernate-mapping-3.0.dtd
Hibernate Mapping DTD 2.0替换为Hibernate Mapping DTD 3.0
8.hibernate.cfg.xml?br />Hibernate Mapping DTD 2.0替换为Hibernate Mapping DTD 3.0
<property name="hibernate.dialect">org.hibernate.dialect.SQLServerDialect</property>
9.hibernate.properties中类?br />
10.cache-config.xml?br /><provider className="net.sf.hibernate.cache.OSCacheProvider"/>替换?br /><provider className="org.hibernate.cache.OSCacheProvider"/>
11.classeshibernate.properties?br />hibernate.cache.provider_class=org.hibernate.cache.EhCacheProvider
hibernate.dialect=org.hibernate.dialect.SQLServerDialect
晕s?怎么q里q有
q是用编辑器暴力替换一下吧q脆
然后部v,集成试,希望一切ok
l果咣铛,q是报错
12.在自动外部模块部分有一个功能是Ҏ(gu)模版自动生成.hbm文g在load,l果出来?hbm中有问题:
生成?<composite-id unsaved-value="any" mapped="false">其中mapped="false" 出错.
找了(jin)半天才发现在|上的hibernate-mapping-3.0.dtd文g有支持mapped="false"q个属?而本地的hebernate3.0.5中的
hibernate-mapping-3.0.dtd文g没有q个属??hibernate也太不负责了(jin)? 解决办法把hibernate-mapping-3.0.dtd
copy到jboss\bin目录下然?在template文g?br /><!DOCTYPE hibernate-mapping PUBLIC "-//Hibernate/Hibernate Mapping DTD//EN" "hibernate-mapping-3.0.dtd">
然后他会(x)在jboss\bin目录下读取该文g
13.重新试,q是咣铛
发现子类ȝcL据时抛出异常
"org.hibernate.LazyInitializationException: could not initialize proxy"
延迟抓取出的?hb3对many-to-one的默认处理是lazy = "proxy"
没有搞懂到底怎么回事,把所有many-to-one,one-to-one都加上lazy="false"
再测试终于大功告?/span>
]]>
public class AllTablePb implements Serializable {
(tng)// Fields
(tng)private AllTablePbId id;
(tng)// Property accessors
(tng)public AllTablePbId getId() {
(tng) return this.id;
(tng)}
(tng)public void setId(AllTablePbId id) {
(tng) this.id = id;
(tng)}
}
public class AllTablePbId implements java.io.Serializable {
(tng)// Fields
(tng)private String owner;
(tng)private String tableName;
(tng)private String columnName;
(tng)private String dataType;
(tng)private String pbcCnam;
(tng)private String pbcCmnt;
(tng)// Property accessors
(tng)public String getOwner() {
(tng) return this.owner;
(tng)}
(tng)public void setOwner(String owner) {
(tng) this.owner = owner;
(tng)}
(tng)public String getTableName() {
(tng) return this.tableName;
(tng)}
(tng)public void setTableName(String tableName) {
(tng) this.tableName = tableName;
(tng)}
(tng)public String getColumnName() {
(tng) return this.columnName;
(tng)}
(tng)public void setColumnName(String columnName) {
(tng) this.columnName = columnName;
(tng)}
(tng)public String getDataType() {
(tng) return this.dataType;
(tng)}
(tng)public void setDataType(String dataType) {
(tng) this.dataType = dataType;
(tng)}
(tng)public String getPbcCnam() {
(tng) return this.pbcCnam;
(tng)}
(tng)public void setPbcCnam(String pbcCnam) {
(tng) this.pbcCnam = pbcCnam;
(tng)}
(tng)public String getPbcCmnt() {
(tng) return this.pbcCmnt;
(tng)}
(tng)public void setPbcCmnt(String pbcCmnt) {
(tng) this.pbcCmnt = pbcCmnt;
(tng)}
}
配置文gQ?br /><hibernate-mapping>
(tng)<class name="com.hhkj.workflow.bean.AllTablePb" table="V_ALLTAB_PB" schema="CANP">
(tng) <composite-id name="id" class="com.hhkj.workflow.bean.AllTablePbId">
(tng) (tng) <key-property name="owner" type="string">
(tng) (tng) (tng) <column name="OWNER" length="30" />
(tng) (tng) </key-property>
(tng) (tng) <key-property name="tableName" type="string">
(tng) (tng) (tng) <column name="TABLE_NAME" length="30" />
(tng) (tng) </key-property>
(tng) (tng) <key-property name="columnName" type="string">
(tng) (tng) (tng) <column name="COLUMN_NAME" length="30" />
(tng) (tng) </key-property>
(tng) (tng) <key-property name="dataType" type="string">
(tng) (tng) (tng) <column name="DATA_TYPE" length="106" />
(tng) (tng) </key-property>
(tng) (tng) <key-property name="pbcCnam" type="string">
(tng) (tng) (tng) <column name="PBC_CNAM" length="30" />
(tng) (tng) </key-property>
(tng) (tng) <key-property name="pbcCmnt" type="string">
(tng) (tng) (tng) <column name="PBC_CMNT" length="254" />
(tng) (tng) </key-property>
(tng) </composite-id>
(tng)</class>
</hibernate-mapping>
q样可以通过AllTablePb.getId()取得相关的信息?br /> (tng) (tng) (tng) (tng) (tng) 感觉myeclipse的确不错Q特别是对于使用hibernateq(sh)是非常熟l的人来_(d)通过它可以帮解决不少问题?br /> (tng) (tng) (tng) (tng) (tng) 通过q样对视囄操作Q同样也可以用到对于那些没有定义主键的表Q操作方法是一L(fng)?br /> (tng) (tng) (tng) (tng) (tng) 呵呵Q个人的一点体?x)，Ƣ迎大家多提意见?/p>