[JBoss JIRA] Created: (JBPM-1707) pageflow parsing is slow
by Alejandro Guizar (JIRA)
pageflow parsing is slow
------------------------
Key: JBPM-1707
URL: https://jira.jboss.org/jira/browse/JBPM-1707
Project: JBoss jBPM
Issue Type: Bug
Security Level: Public (Everyone can see)
Components: Core Engine
Affects Versions: jBPM 3.2.3
Environment: Red Hat Enterprise Linux AS release 4 (Nahant Update 7)
enterprise-installer-4.3.0.GA-1.ep1.8.jar
jdk-1.5.0_16-fcs
Reporter: Alejandro Guizar
Assignee: Alejandro Guizar
Fix For: jBPM 3.3.0 CR1
On a RHEL machine, while deploying a seam based ear, which contains 13 pageflow definitions, the parsing of these 13 pageflow files takes more than 34 minutes (!).
The culprit seems to be the usage of the original systemId (http://jboss.com/products/seam/pageflow-2.0.xsd) in the org.jbpm.jdpl.xml.JpdlParser$JpdlEntityResolver class, as this
apparently results in slow internet lookups:
log.debug("original systemId as input source");
inputSource = new InputSource(systemId);
Using a null InputSource:
inputSource = new InputSource((InputStream)null);
brings the parsing time down to less than 1 second.
Of course it would be better to actually pass in a reference to the local pageflow-2.0.xsd file, which is inside the jboss-seam.jar.
--
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators: https://jira.jboss.org/jira/secure/Administrators.jspa
-
For more information on JIRA, see: http://www.atlassian.com/software/jira
14 years, 11 months
[JBoss JIRA] Created: (JBPM-1771) Fix JobExecutorDbTest on postgresql
by Thomas Diesler (JIRA)
Fix JobExecutorDbTest on postgresql
-----------------------------------
Key: JBPM-1771
URL: https://jira.jboss.org/jira/browse/JBPM-1771
Project: JBoss jBPM
Issue Type: Task
Security Level: Public (Everyone can see)
Reporter: Thomas Diesler
Fix For: jBPM 3.3.1 GA
expected:<[10X, 10Y, 10Z, 10a, 10b, 10c, 10d, 10e, 11X, 11Y, 11Z, 11a, 11b, 11c, 11d, 11e, 12X, 12Y, 12Z, 12a, 12b, 12c, 12d, 12e, 13X, 13Y, 13Z, 13a, 13b, 13c, 13d, 13e, 14X, 14Y, 14Z, 14a, 14b, 14c, 14d, 14e, 15X, 15Y, 15Z, 15a, 15b, 15c, 15d, 15e, 16X, 16Y, 16Z, 16a, 16b, 16c, 16d, 16e, 17X, 17Y, 17Z, 17a, 17b, 17c, 17d, 17e, 18X, 18Y, 18Z, 18a, 18b, 18c, 18d, 18e, 19X, 19Y, 19Z, 19a, 19b, 19c, 19d, 19e, 1X, 1Y, 1Z, 1a, 1b, 1c, 1d, 1e, 20X, 20Y, 20Z, 20a, 20b, 20c, 20d, 20e, 2X, 2Y, 2Z, 2a, 2b, 2c, 2d, 2e, 3X, 3Y, 3Z, 3a, 3b, 3c, 3d, 3e, 4X, 4Y, 4Z, 4a, 4b, 4c, 4d, 4e, 5X, 5Y, 5Z, 5a, 5b, 5c, 5d, 5e, 6X, 6Y, 6Z, 6a, 6b, 6c, 6d, 6e, 7X, 7Y, 7Z, 7a, 7b, 7c, 7d, 7e, 8X, 8Y, 8Z, 8a, 8b, 8c, 8d, 8e, 9X, 9Y, 9Z, 9a, 9b, 9c, 9d, 9e]>
but was:<[100X, 100Y, 100Z, 100a, 100b, 100c, 100d, 100e, 108X, 108Y, 108Z, 108a, 108b, 108c, 108d, 108e, 116X, 116Y, 116Z, 116a, 116b, 116c, 116d, 116e, 124X, 124Y, 124Z, 124a, 124b, 124c, 124d, 124e, 132X, 132Y, 132Z, 132a, 132b, 132c, 132d, 132e, 140X, 140Y, 140Z, 140a, 140b, 140c, 140d, 140e, 148X, 148Y, 148Z, 148a, 148b, 148c, 148d, 148e, 156X, 156Y, 156Z, 156a, 156b, 156c, 156d, 156e, 164X, 164Y, 164Z, 164a, 164b, 164c, 164d, 164e, 172X, 172Y, 172Z, 172a, 172b, 172c, 172d, 172e, 180X, 180Y, 180Z, 180a, 180b, 180c, 180d, 180e, 188X, 188Y, 188Z, 188a, 188b, 188c, 188d, 188e, 36X, 36Y, 36Z, 36a, 36b, 36c, 36d, 36e, 44X, 44Y, 44Z, 44a, 44b, 44c, 44d, 44e, 52X, 52Y, 52Z, 52a, 52b, 52c, 52d, 52e, 60X, 60Y, 60Z, 60a, 60b, 60c, 60d, 60e, 68X, 68Y, 68Z, 68a, 68b, 68c, 68d, 68e, 76X, 76Y, 76Z, 76a, 76b, 76c, 76d, 76e, 84X, 84Y, 84Z, 84a, 84b, 84c, 84d, 84e, 92X, 92Y, 92Z, 92a, 92b, 92c, 92d, 92e]>
--
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators: https://jira.jboss.org/jira/secure/Administrators.jspa
-
For more information on JIRA, see: http://www.atlassian.com/software/jira
14 years, 11 months
[JBoss JIRA] Created: (JBPM-1178) Script conditions in transitions are evaluated as expressions by the decision node
by Romain Lamarche (JIRA)
Script conditions in transitions are evaluated as expressions by the decision node
----------------------------------------------------------------------------------
Key: JBPM-1178
URL: http://jira.jboss.com/jira/browse/JBPM-1178
Project: JBoss jBPM
Issue Type: Bug
Components: Core Engine
Affects Versions: jBPM jPDL 3.2.2
Environment: Windows XP, JBoss AS 4.2.2GA, MS SQL EXPRESS 2005, Jdk5
Reporter: Romain Lamarche
Assigned To: Tom Baeyens
The decision node does not evaluate scripts in conditional transitions.
Indeed, the execute method of node type decision does not check what type of condition it is.
Listing of the wrong code :
(file Decision.java lines 120-135 in Jbpm Jpdl 3.2.2)
} else {
// new mode based on conditions in the transition itself
Iterator iter = leavingTransitions.iterator();
while (iter.hasNext() && (transition==null)) {
Transition candidate = (Transition) iter.next();
String conditionExpression = candidate.getCondition();
if (conditionExpression!=null) {
Object result = JbpmExpressionEvaluator.evaluate(conditionExpression, executionContext);
if (Boolean.TRUE.equals(result)) {
transition = candidate;
}
}
}
}
The problem is the "JbpmExpressionEvaluator".
--
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators: http://jira.jboss.com/jira/secure/Administrators.jspa
-
For more information on JIRA, see: http://www.atlassian.com/software/jira
14 years, 11 months
[JBoss JIRA] Created: (JBPM-1914) Problem in retrieving variables of Serializable objects
by Mauro Molinari (JIRA)
Problem in retrieving variables of Serializable objects
-------------------------------------------------------
Key: JBPM-1914
URL: https://jira.jboss.org/jira/browse/JBPM-1914
Project: JBoss jBPM
Issue Type: Bug
Security Level: Public (Everyone can see)
Components: Core Engine
Affects Versions: jBPM 3.3.0 GA
Reporter: Mauro Molinari
Priority: Critical
I have encountered a serious problem working with variables whose value is a serializable object.
The problem is that sometimes (not always!) an error like the following is given when retrieving a variable:
java.lang.ClassCastException: org.jbpm.bytes.ByteArray$$EnhancerByCGLIB$$cc67d06e
The code is like:
MySerializableClass obj = (MySerializableClass) contextInstance.getVariable("myVariable");
(of course, "myVariable" is a variable that I previously have set with a MySerializableClass object)
Apart from the CGLIB awful problem with proxies, even if I extract the value from the proxy with the following:
public static Object getCGLIBProxiedObject(final Object proxy)
{
if(proxy instanceof HibernateProxy)
{
final LazyInitializer initializer =
((HibernateProxy) proxy).getHibernateLazyInitializer();
return initializer.getImplementation();
}
return proxy;
}
the problem is that the variable value is retrieved as a org.jbpm.bytes.ByteArray instead of a MyClass instance!
I did some debugging and discovered that:
- in org.jbpm.context.exe.VariableContainer.getVariable(String), when org.jbpm.context.exe.VariableContainer.hasVariableLocally(String) is called, Hibernate lazily retrieves a org.jbpm.context.exe.variableinstance.ByteArrayInstance instance from the database and creates it using its empty constructor
- unfortunately, when a org.jbpm.context.exe.VariableInstance (extended by ByteArrayInstance) is constructed using its constructor, its converter instance variable remains null
- the VariableInstance converter instance variable is set to something only when a VariableInstance is created using org.jbpm.context.exe.JbpmType.newVariableInstance(), but this is not the case when a VariableInstance is lazily loaded by Hibernate; in fact, in my case, org.jbpm.context.exe.JbpmType.newVariableInstance() is never called and the converter field of my ByteArrayInstance instance is always null!
- this causes org.jbpm.context.exe.VariableInstance.getValue() to return the value given by org.jbpm.context.exe.variableinstance.ByteArrayInstance.getObject() without applying the necessary conversion (deserialization)
Please note that my variable values are stored in jBPM database in JBPM_VARIABLEINSTANCE table with the CONVERTER column correctly populated with "R".
I think this is a critical problem. By now, the only workaround I have found is to always apply the following utility method when retrieving variable values, before returning them to the client code:
public static Object extractVariableValue(final Object rawValue)
{
Object result = CardinisJbpmUtilities.getCGLIBProxiedObject(rawValue);
if(result instanceof ByteArray)
try
{
result =
new ObjectInputStream(new ByteArrayInputStream(((ByteArray) result)
.getBytes())).readObject();
}
catch(final Exception e)
{
throw new RuntimeException(e);
}
return result;
}
--
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators: https://jira.jboss.org/jira/secure/Administrators.jspa
-
For more information on JIRA, see: http://www.atlassian.com/software/jira
15 years