jbpm3.2.2部署流程定义文件的错误

http://blog.csdn.net/geekwang/article/details/2517801




今天按照jbpm开发指南里的数据库持久存储方式存储流程定义文件,一直出现插入字段错误,原来是jbpm3.2.2在mysql中字段的类型的错误.

错误如下:

[java] view plaincopy
警告: SQL Error: 1406, SQLState: 22001 
2008-6-6 18:20:58 org.hibernate.util.JDBCExceptionReporter logExceptions 
严重: Data too long for column 'ISTERMINATIONIMPLICIT_' at row 1 
2008-6-6 18:20:58 org.hibernate.util.JDBCExceptionReporter logExceptions 
警告: SQL Error: 1406, SQLState: 22001 
2008-6-6 18:20:58 org.hibernate.util.JDBCExceptionReporter logExceptions 
严重: Data too long for column 'ISASYNC_' at row 1 
2008-6-6 18:20:58 org.jbpm.persistence.db.DbPersistenceService commit 
严重: hibernate commit failed 
org.hibernate.exception.DataException: could not insert: [org.jbpm.graph.node.StartState] 
    at org.hibernate.exception.SQLStateConverter.convert(SQLStateConverter.java:77) 
    at org.hibernate.exception.JDBCExceptionHelper.convert(JDBCExceptionHelper.java:43) 
    at org.hibernate.id.insert.AbstractReturningDelegate.performInsert(AbstractReturningDelegate.java:40) 
    at org.hibernate.persister.entity.AbstractEntityPersister.insert(AbstractEntityPersister.java:2158) 
    at org.hibernate.persister.entity.AbstractEntityPersister.insert(AbstractEntityPersister.java:2638) 
    at org.hibernate.action.EntityIdentityInsertAction.execute(EntityIdentityInsertAction.java:48) 
    at org.hibernate.engine.ActionQueue.execute(ActionQueue.java:248) 
    at org.hibernate.event.def.AbstractSaveEventListener.performSaveOrReplicate(AbstractSaveEventListener.java:298) 
    at org.hibernate.event.def.AbstractSaveEventListener.performSave(AbstractSaveEventListener.java:181) 
    at org.hibernate.event.def.AbstractSaveEventListener.saveWithGeneratedId(AbstractSaveEventListener.java:107) 
    at org.hibernate.event.def.DefaultSaveOrUpdateEventListener.saveWithGeneratedOrRequestedId(DefaultSaveOrUpdateEventListener.java:187) 
    at org.hibernate.event.def.DefaultSaveOrUpdateEventListener.entityIsTransient(DefaultSaveOrUpdateEventListener.java:172) 
    at org.hibernate.event.def.DefaultSaveOrUpdateEventListener.performSaveOrUpdate(DefaultSaveOrUpdateEventListener.java:94) 
    at org.hibernate.event.def.DefaultSaveOrUpdateEventListener.onSaveOrUpdate(DefaultSaveOrUpdateEventListener.java:70) 
    at org.hibernate.impl.SessionImpl.fireSaveOrUpdate(SessionImpl.java:507) 
    at org.hibernate.impl.SessionImpl.saveOrUpdate(SessionImpl.java:499) 
    at org.hibernate.engine.CascadingAction$1.cascade(CascadingAction.java:218) 
       接下去我想直接在mysql里修改字段类型,但是竟然不能保存,到现在也不知道为什么.用sql语句就可以.

修改的sql语句:

[java] view plaincopy
ALTER TABLE JBPM_PROCESSDEFINITION MODIFY ISTERMINATIONIMPLICIT_ TINYINT(1);  
ALTER TABLE JBPM_ACTION MODIFY ISASYNC_ TINYINT(1);  
ALTER TABLE JBPM_ACTION MODIFY ISPROPAGATIONALLOWED_ TINYINT(1);  
ALTER TABLE JBPM_NODE MODIFY ISASYNC_ TINYINT(1);  
ALTER TABLE JBPM_NODE MODIFY ISASYNCEXCL_ TINYINT(1);  
ALTER TABLE JBPM_NODE MODIFY CREATETASKS_ TINYINT(1);  
ALTER TABLE JBPM_NODE MODIFY ENDTASKS_ TINYINT(1);  
ALTER TABLE JBPM_PROCESSINSTANCE MODIFY ISSUSPENDED_ TINYINT(1);  
ALTER TABLE JBPM_TASK MODIFY ISBLOCKING_ TINYINT(1);  
ALTER TABLE JBPM_TASK MODIFY ISSIGNALLING_ TINYINT(1);  
ALTER TABLE JBPM_TASKINSTANCE MODIFY ISCANCELLED_ TINYINT(1);  
ALTER TABLE JBPM_TASKINSTANCE MODIFY ISSUSPENDED_ TINYINT(1);  
ALTER TABLE JBPM_TASKINSTANCE MODIFY ISOPEN_ TINYINT(1);  
ALTER TABLE JBPM_TASKINSTANCE MODIFY ISSIGNALLING_ TINYINT(1);  
ALTER TABLE JBPM_TASKINSTANCE MODIFY ISBLOCKING_ TINYINT(1);  
ALTER TABLE JBPM_TOKEN MODIFY ISABLETOREACTIVATEPARENT_ TINYINT(1);  
ALTER TABLE JBPM_TOKEN MODIFY ISTERMINATIONIMPLICIT_ TINYINT(1);  
ALTER TABLE JBPM_TOKEN MODIFY ISSUSPENDED_ TINYINT(1); 
把上面这些字段改了之后就插入正常了.

附:编程方式部署流程定义文件代码如下:

[java] view plaincopy
public class HelloWorldDbTest extends TestCase { 
 
  static JbpmConfiguration jbpmConfiguration = null;  
 
  static { 
    // An example configuration file such as this can be found in  
    // 'src/config.files'.  Typically the configuration information is in the  
    // resource file 'jbpm.cfg.xml', but here we pass in the configuration  
    // information as an XML string. 
     
    // First we create a JbpmConfiguration statically.  One JbpmConfiguration 
    // can be used for all threads in the system, that is why we can safely  
    // make it static. 
 
    jbpmConfiguration = JbpmConfiguration.parseXmlString( 
      "<jbpm-configuration></jbpm-configuration>" + 
       
      // A jbpm-context mechanism separates the jbpm core  
      // engine from the services that jbpm uses from  
      // the environment.   
       
      "  <jbpm-context></jbpm-context>" + 
      "    <service name="persistence" factory="org.jbpm.persistence.db.DbPersistenceServiceFactory"></service>" +  
      "  " + 
       
      // Also all the resource files that are used by jbpm are  
      // referenced from the jbpm.cfg.xml 
       
      "  <string name="resource.hibernate.cfg.xml" value="hibernate.cfg.xml"></string>" + 
      "  <string name="resource.business.calendar" value="org/jbpm/calendar/jbpm.business.calendar.properties"></string>" + 
      "  <string name="resource.default.modules" value="org/jbpm/graph/def/jbpm.default.modules.properties"></string>" + 
      "  <string name="resource.converter" value="org/jbpm/db/hibernate/jbpm.converter.properties"></string>" + 
      "  <string name="resource.action.types" value="org/jbpm/graph/action/action.types.xml"></string>" + 
      "  <string name="resource.node.types" value="org/jbpm/graph/node/node.types.xml"></string>" + 
      "  <string name="resource.varmapping" value="org/jbpm/context/exe/jbpm.varmapping.xml"></string>" + 
      "" 
    ); 
  } 
   
  public void setUp() { 
    jbpmConfiguration.createSchema(); 
  } 
   
  public void tearDown() { 
    jbpmConfiguration.dropSchema(); 
  } 
 
  public void testSimplePersistence() { 
    // Between the 3 method calls below, all data is passed via the  
    // database.  Here, in this unit test, these 3 methods are executed 
    // right after each other because we want to test a complete process 
    // scenario.  But in reality, these methods represent different  
    // requests to a server. 
     
    // Since we start with a clean, empty in-memory database, we have to  
    // deploy the process first.  In reality, this is done once by the  
    // process developer. 
    deployProcessDefinition(); 
 
    // Suppose we want to start a process instance (=process execution) 
    // when a user submits a form in a web application... 
    processInstanceIsCreatedWhenUserSubmitsWebappForm(); 
 
    // Then, later, upon the arrival of an asynchronous message the  
    // execution must continue. 
    theProcessInstanceContinuesWhenAnAsyncMessageIsReceived(); 
  } 
 
  public void deployProcessDefinition() { 
    // This test shows a process definition and one execution  
    // of the process definition.  The process definition has  
    // 3 nodes: an unnamed start-state, a state 's' and an  
    // end-state named 'end'. 
    ProcessDefinition processDefinition = ProcessDefinition.parseXmlString( 
      "<process-definition name="hello world"></process-definition>" + 
      "  <start-state name="start"></start-state>" + 
      "    <transition to="s"></transition>" + 
      "  " + 
      "  <state name="s"></state>" + 
      "    <transition to="end"></transition>" + 
      "  " + 
      "  <end-state name="end"></end-state>" + 
      "" 
    ); 
 
    // Lookup the pojo persistence context-builder that is configured above 
    JbpmContext jbpmContext = jbpmConfiguration.createJbpmContext(); 
    try { 
      // Deploy the process definition in the database  
      jbpmContext.deployProcessDefinition(processDefinition); 
 
    } finally { 
      // Tear down the pojo persistence context. 
      // This includes flush the SQL for inserting the process definition   
      // to the database. 
      jbpmContext.close(); 
    } 
  } 
 
  public void processInstanceIsCreatedWhenUserSubmitsWebappForm() { 
    // The code in this method could be inside a struts-action  
    // or a JSF managed bean.  
 
    // Lookup the pojo persistence context-builder that is configured above 
    JbpmContext jbpmContext = jbpmConfiguration.createJbpmContext(); 
    try { 
 
      GraphSession graphSession = jbpmContext.getGraphSession(); 
       
      ProcessDefinition processDefinition =  
          graphSession.findLatestProcessDefinition("hello world"); 
     
      // With the processDefinition that we retrieved from the database, we  
      // can create an execution of the process definition just like in the  
      // hello world example (which was without persistence). 
      ProcessInstance processInstance =  
          new ProcessInstance(processDefinition); 
       
      Token token = processInstance.getRootToken();  
      assertEquals("start", token.getNode().getName()); 
      // Let's start the process execution 
      token.signal(); 
      // Now the process is in the state 's'. 
      assertEquals("s", token.getNode().getName()); 
       
      // Now the processInstance is saved in the database.  So the  
      // current state of the execution of the process is stored in the  
      // database.   
      jbpmContext.save(processInstance); 
      // The method below will get the process instance back out  
      // of the database and resume execution by providing another  
      // external signal. 
 
    } finally { 
      // Tear down the pojo persistence context. 
      jbpmContext.close(); 
    } 
  } 
 
  public void theProcessInstanceContinuesWhenAnAsyncMessageIsReceived() { 
    // The code in this method could be the content of a message driven bean. 
 
    // Lookup the pojo persistence context-builder that is configured above 
    JbpmContext jbpmContext = jbpmConfiguration.createJbpmContext(); 
    try { 
 
      GraphSession graphSession = jbpmContext.getGraphSession(); 
      // First, we need to get the process instance back out of the database. 
      // There are several options to know what process instance we are dealing  
      // with here.  The easiest in this simple test case is just to look for  
      // the full list of process instances.  That should give us only one  
      // result.  So let's look up the process definition. 
       
      ProcessDefinition processDefinition =  
          graphSession.findLatestProcessDefinition("hello world"); 
 
      // Now, we search for all process instances of this process definition. 
      List processInstances =  
          graphSession.findProcessInstances(processDefinition.getId()); 
       
      // Because we know that in the context of this unit test, there is  
      // only one execution.  In real life, the processInstanceId can be  
      // extracted from the content of the message that arrived or from  
      // the user making a choice. 
      ProcessInstance processInstance =  
          (ProcessInstance) processInstances.get(0); 
       
      // Now we can continue the execution.  Note that the processInstance 
      // delegates signals to the main path of execution (=the root token). 
      processInstance.signal(); 
 
      // After this signal, we know the process execution should have  
      // arrived in the end-state. 
      assertTrue(processInstance.hasEnded()); 
       
      // Now we can update the state of the execution in the database 
      jbpmContext.save(processInstance); 
 
    } finally { 
      // Tear down the pojo persistence context. 
      jbpmContext.close(); 
    } 
  } 
} 
[java] view plaincopy
3.2版比3.1版的配置文件多了一个配置项目，加上这个配置项目之后就OK了  
 
<service name="tx" factory="org.jbpm.tx.TxServiceFactory">  
 
因为是按照开发指南的例子写了一个jbpm.cfg.xml，缺少这一条，所以又造成了一个错误。 
[java] view plaincopy
一般来说，jbpm.cfg.xml其实可以不用写的，jbpm会自动使用org.jbpm包下的default.jbpm.cfg.xml作为配置文件来启动整个jbpm服务。