Author: tom.baeyens(a)jboss.com Date: 2008-09-18 09:55:29 -0400 (Thu, 18 Sep 2008) New Revision: 2279 Added: jbpm4/pvm/trunk/modules/core/src/test/java/org/jbpm/pvm/example04/ jbpm4/pvm/trunk/modules/core/src/test/java/org/jbpm/pvm/example04/Display.java jbpm4/pvm/trunk/modules/core/src/test/java/org/jbpm/pvm/example04/DisplayActivityTest.java jbpm4/pvm/trunk/modules/core/src/test/java/org/jbpm/pvm/example04/TestConsole.java jbpm4/pvm/trunk/modules/core/src/test/java/org/jbpm/pvm/example05/ jbpm4/pvm/trunk/modules/core/src/test/java/org/jbpm/pvm/example05/ExternalActivityExampleTest.java jbpm4/pvm/trunk/modules/core/src/test/java/org/jbpm/pvm/example05/WaitState.java jbpm4/pvm/trunk/modules/core/src/test/java/org/jbpm/pvm/example06/ jbpm4/pvm/trunk/modules/core/src/test/java/org/jbpm/pvm/example06/BasicProcessExecutionTest.java jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/images/execution.loan.archive.png jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/images/execution.loan.end.png jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/images/execution.loan.evaluate.png jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/images/process.ab.png jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/images/process.loan.png jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/modules/ch04-ImplementingBasicActivities.xml jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/modules/ch05-ProcessAnatomy.xml jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/modules/ch06-ImplementingAdvancedActivities.xml jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/modules/ch07-Variables.xml jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/modules/ch08-Timers.xml jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/modules/ch09-AsynchronousContinuations.xml jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/modules/ch10-SoftwareLogging.xml Removed: jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/images/execution.state.accept.png jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/images/execution.state.end.png jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/images/execution.state.evaluation.png jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/images/loan.process.png jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/images/process.png jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/modules/ch04-BasicProcessExecutionConcepts.xml jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/modules/chxx-AdvancedGraphExecution.xml jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/modules/chxx-AsynchronousContinuations.xml jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/modules/chxx-SoftwareLogging.xml jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/modules/chxx-Timers.xml jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/modules/chxx-Variables.xml Modified: jbpm4/pvm/trunk/modules/manual/src/main/diagrams/diagrams.mdzip jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/master.xml jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/modules/ch01-Introduction.xml jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/modules/ch02-ExecutionModes.xml Log: docs Added: jbpm4/pvm/trunk/modules/core/src/test/java/org/jbpm/pvm/example04/Display.java =================================================================== --- jbpm4/pvm/trunk/modules/core/src/test/java/org/jbpm/pvm/example04/Display.java (rev 0) +++ jbpm4/pvm/trunk/modules/core/src/test/java/org/jbpm/pvm/example04/Display.java 2008-09-18 13:55:29 UTC (rev 2279) @@ -0,0 +1,40 @@ +/* + * JBoss, Home of Professional Open Source + * Copyright 2005, JBoss Inc., and individual contributors as indicated + * by the @authors tag. See the copyright.txt in the distribution for a + * full listing of individual contributors. + * + * This is free software; you can redistribute it and/or modify it + * under the terms of the GNU Lesser General Public License as + * published by the Free Software Foundation; either version 2.1 of + * the License, or (at your option) any later version. + * + * This software is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * Lesser General Public License for more details. + * + * You should have received a copy of the GNU Lesser General Public + * License along with this software; if not, write to the Free + * Software Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA + * 02110-1301 USA, or see the FSF site: http://www.fsf.org. + */ +package org.jbpm.pvm.example04; + +import org.jbpm.pvm.activity.Activity; +import org.jbpm.pvm.activity.ActivityExecution; + +public class Display implements Activity { + + private static final long serialVersionUID = 1L; + + String message; + + public Display(String message) { + this.message = message; + } + + public void execute(ActivityExecution execution) { + System.out.println(message); + } +} \ No newline at end of file Added: jbpm4/pvm/trunk/modules/core/src/test/java/org/jbpm/pvm/example04/DisplayActivityTest.java =================================================================== --- jbpm4/pvm/trunk/modules/core/src/test/java/org/jbpm/pvm/example04/DisplayActivityTest.java (rev 0) +++ jbpm4/pvm/trunk/modules/core/src/test/java/org/jbpm/pvm/example04/DisplayActivityTest.java 2008-09-18 13:55:29 UTC (rev 2279) @@ -0,0 +1,34 @@ +package org.jbpm.pvm.example04; + +import org.jbpm.pvm.Execution; +import org.jbpm.pvm.client.ClientProcessDefinition; +import org.jbpm.pvm.model.ProcessFactory; +import org.jbpm.pvm.test.base.JbpmTestCase; + + +public class DisplayActivityTest extends JbpmTestCase { + + TestConsole testConsole; + + public void setUp() { + testConsole = TestConsole.install(); + } + + public void tearDown() { + TestConsole.uninstall(); + testConsole = null; + } + + public void testHelloWorld() { + ClientProcessDefinition processDefinition = ProcessFactory.build() + .node("a").initial().behaviour(new Display("hello")) + .transition().to("b") + .node("b").behaviour(new Display("world")) + .done(); + + Execution execution = processDefinition.startProcessInstance(); + + assertEquals("hello", testConsole.getLine(0)); + assertEquals("world", testConsole.getLine(1)); + } +} Added: jbpm4/pvm/trunk/modules/core/src/test/java/org/jbpm/pvm/example04/TestConsole.java =================================================================== --- jbpm4/pvm/trunk/modules/core/src/test/java/org/jbpm/pvm/example04/TestConsole.java (rev 0) +++ jbpm4/pvm/trunk/modules/core/src/test/java/org/jbpm/pvm/example04/TestConsole.java 2008-09-18 13:55:29 UTC (rev 2279) @@ -0,0 +1,38 @@ +package org.jbpm.pvm.example04; + +import java.io.PrintStream; +import java.util.ArrayList; +import java.util.List; + +/** wraps the System.out and buffers the println(String) invocations + * so that those can be asserted */ +public class TestConsole extends PrintStream { + + List<String> lines = new ArrayList<String>(); + + public TestConsole() { + super(System.out); + } + + public static TestConsole install() { + TestConsole testConsole = new TestConsole(); + System.setOut(testConsole); + return testConsole; + } + + public static void uninstall() { + if (System.out instanceof TestConsole) { + TestConsole testConsole = (TestConsole) System.out; + System.setOut((PrintStream) testConsole.out); + } + } + + public void println(String x) { + lines.add(x); + super.println(x); + } + + public String getLine(int i) { + return lines.get(i); + } +} Added: jbpm4/pvm/trunk/modules/core/src/test/java/org/jbpm/pvm/example05/ExternalActivityExampleTest.java =================================================================== --- jbpm4/pvm/trunk/modules/core/src/test/java/org/jbpm/pvm/example05/ExternalActivityExampleTest.java (rev 0) +++ jbpm4/pvm/trunk/modules/core/src/test/java/org/jbpm/pvm/example05/ExternalActivityExampleTest.java 2008-09-18 13:55:29 UTC (rev 2279) @@ -0,0 +1,27 @@ +package org.jbpm.pvm.example05; + +import org.jbpm.pvm.client.ClientExecution; +import org.jbpm.pvm.client.ClientProcessDefinition; +import org.jbpm.pvm.model.ProcessFactory; +import org.jbpm.pvm.test.base.JbpmTestCase; + + +public class ExternalActivityExampleTest extends JbpmTestCase { + + public void testExternalActivityExample() { + ClientProcessDefinition processDefinition = ProcessFactory.build() + .node("a").initial().behaviour(new WaitState()) + .transition().to("b") + .node("b").behaviour(new WaitState()) + .done(); + + ClientExecution execution = processDefinition.startProcessInstance(); + + assertEquals("a", execution.getNodeName()); + + execution.signal(); + + assertEquals("b", execution.getNodeName()); + } + +} Added: jbpm4/pvm/trunk/modules/core/src/test/java/org/jbpm/pvm/example05/WaitState.java =================================================================== --- jbpm4/pvm/trunk/modules/core/src/test/java/org/jbpm/pvm/example05/WaitState.java (rev 0) +++ jbpm4/pvm/trunk/modules/core/src/test/java/org/jbpm/pvm/example05/WaitState.java 2008-09-18 13:55:29 UTC (rev 2279) @@ -0,0 +1,21 @@ +package org.jbpm.pvm.example05; + +import java.util.Map; + +import org.jbpm.pvm.activity.ActivityExecution; +import org.jbpm.pvm.activity.ExternalActivity; + +public class WaitState implements ExternalActivity { + + private static final long serialVersionUID = 1L; + + public void execute(ActivityExecution execution) { + execution.waitForSignal(); + } + + public void signal(ActivityExecution execution, + String signalName, + Map<String, Object> parameters) { + execution.take(signalName); + } +} Added: jbpm4/pvm/trunk/modules/core/src/test/java/org/jbpm/pvm/example06/BasicProcessExecutionTest.java =================================================================== --- jbpm4/pvm/trunk/modules/core/src/test/java/org/jbpm/pvm/example06/BasicProcessExecutionTest.java (rev 0) +++ jbpm4/pvm/trunk/modules/core/src/test/java/org/jbpm/pvm/example06/BasicProcessExecutionTest.java 2008-09-18 13:55:29 UTC (rev 2279) @@ -0,0 +1,32 @@ +package org.jbpm.pvm.example06; + +import org.jbpm.pvm.client.ClientExecution; +import org.jbpm.pvm.client.ClientProcessDefinition; +import org.jbpm.pvm.example04.Display; +import org.jbpm.pvm.example05.WaitState; +import org.jbpm.pvm.model.ProcessFactory; +import org.jbpm.pvm.test.base.JbpmTestCase; + + +public class BasicProcessExecutionTest extends JbpmTestCase { + + public void testBasicProcessExecution() { + ClientProcessDefinition processDefinition = ProcessFactory.build("loan") + .node("submit loan request").initial().behaviour(new Display("loan request submitted")) + .transition().to("evaluate") + .node("evaluate").behaviour(new WaitState()) + .transition("approve").to("wire money") + .transition("reject").to("end") + .node("wire money").behaviour(new Display("wire the money")) + .transition().to("archive") + .node("archive").behaviour(new WaitState()) + .transition().to("end") + .node("end").behaviour(new WaitState()) + .done(); + + ClientExecution execution = processDefinition.startProcessInstance(); + + + + } +} Modified: jbpm4/pvm/trunk/modules/manual/src/main/diagrams/diagrams.mdzip =================================================================== (Binary files differ) Added: jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/images/execution.loan.archive.png =================================================================== (Binary files differ) Property changes on: jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/images/execution.loan.archive.png ___________________________________________________________________ Name: svn:mime-type + application/octet-stream Added: jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/images/execution.loan.end.png =================================================================== (Binary files differ) Property changes on: jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/images/execution.loan.end.png ___________________________________________________________________ Name: svn:mime-type + application/octet-stream Added: jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/images/execution.loan.evaluate.png =================================================================== (Binary files differ) Property changes on: jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/images/execution.loan.evaluate.png ___________________________________________________________________ Name: svn:mime-type + application/octet-stream Deleted: jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/images/execution.state.accept.png =================================================================== (Binary files differ) Deleted: jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/images/execution.state.end.png =================================================================== (Binary files differ) Deleted: jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/images/execution.state.evaluation.png =================================================================== (Binary files differ) Deleted: jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/images/loan.process.png =================================================================== (Binary files differ) Copied: jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/images/process.ab.png (from rev 2276, jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/images/process.png) =================================================================== (Binary files differ) Property changes on: jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/images/process.ab.png ___________________________________________________________________ Name: svn:mime-type + application/octet-stream Name: svn:mergeinfo + Copied: jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/images/process.loan.png (from rev 2276, jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/images/loan.process.png) =================================================================== (Binary files differ) Property changes on: jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/images/process.loan.png ___________________________________________________________________ Name: svn:mime-type + application/octet-stream Name: svn:mergeinfo + Deleted: jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/images/process.png =================================================================== (Binary files differ) Modified: jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/master.xml =================================================================== --- jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/master.xml 2008-09-18 07:25:16 UTC (rev 2278) +++ jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/master.xml 2008-09-18 13:55:29 UTC (rev 2279) @@ -4,7 +4,13 @@ <!ENTITY ch01-Introduction SYSTEM "modules/ch01-Introduction.xml"> <!ENTITY ch02-ExecutionModes SYSTEM "modules/ch02-ExecutionModes.xml"> <!ENTITY ch03-Architecture SYSTEM "modules/ch03-Architecture.xml"> - <!ENTITY ch04-BasicProcessExecutionConcepts SYSTEM "modules/ch04-BasicProcessExecutionConcepts.xml"> + <!ENTITY ch04-ImplementingBasicActivities SYSTEM "modules/ch04-ImplementingBasicActivities.xml"> + <!ENTITY ch05-ProcessAnatomy SYSTEM "modules/ch05-ProcessAnatomy.xml"> + <!ENTITY ch06-ImplementingAdvancedActivities SYSTEM "modules/ch06-ImplementingAdvancedActivities.xml"> + <!ENTITY ch07-Variables SYSTEM "modules/ch07-Variables.xml"> + <!ENTITY ch08-Timers SYSTEM "modules/ch08-Timers.xml"> + <!ENTITY ch09-AsynchronousContinuations SYSTEM "modules/ch09-AsynchronousContinuations.xml"> + <!ENTITY ch10-SoftwareLogging SYSTEM "modules/ch10-SoftwareLogging.xml"> ]> <book lang="en"> @@ -21,6 +27,12 @@ &ch01-Introduction; &ch02-ExecutionModes; &ch03-Architecture; - &ch04-BasicProcessExecutionConcepts; + &ch04-ImplementingBasicActivities; + &ch05-ProcessAnatomy; + &ch06-ImplementingAdvancedActivities; + &ch07-Variables; + &ch08-Timers; + &ch09-AsynchronousContinuations; + &ch10-SoftwareLogging; </book> \ No newline at end of file Modified: jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/modules/ch01-Introduction.xml =================================================================== --- jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/modules/ch01-Introduction.xml 2008-09-18 07:25:16 UTC (rev 2278) +++ jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/modules/ch01-Introduction.xml 2008-09-18 13:55:29 UTC (rev 2279) @@ -24,7 +24,7 @@ </para> <figure id="example.process.graph"> <title>Example process definition</title> - <mediaobject><imageobject><imagedata align="center" fileref="images/loan.process.png"/></imageobject></mediaobject> + <mediaobject><imageobject><imagedata align="center" fileref="images/process.loan.png"/></imageobject></mediaobject> </figure> <para>Typically, process definitions are static. A process definition is composed of nodes and transitions. The runtime behaviour of a node is encapsulated in a so called Modified: jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/modules/ch02-ExecutionModes.xml =================================================================== --- jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/modules/ch02-ExecutionModes.xml 2008-09-18 07:25:16 UTC (rev 2278) +++ jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/modules/ch02-ExecutionModes.xml 2008-09-18 13:55:29 UTC (rev 2279) @@ -25,7 +25,7 @@ </para> <figure id="loan.process"> <title>The loan process</title> - <mediaobject><imageobject><imagedata align="center" fileref="images/loan.process.png"/></imageobject></mediaobject> + <mediaobject><imageobject><imagedata align="center" fileref="images/process.loan.png"/></imageobject></mediaobject> </figure> <programlisting>ClientProcessDefinition processDefinition = ProcessFactory.build("loan") .node("submit loan request").initial().behaviour(AutomaticActivity.class) Deleted: jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/modules/ch04-BasicProcessExecutionConcepts.xml =================================================================== --- jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/modules/ch04-BasicProcessExecutionConcepts.xml 2008-09-18 07:25:16 UTC (rev 2278) +++ jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/modules/ch04-BasicProcessExecutionConcepts.xml 2008-09-18 13:55:29 UTC (rev 2279) @@ -1,544 +0,0 @@ -<chapter id="basicprocessexecutionconcepts"> - <title>Basic process execution concepts</title> - - <para>This chapter explains the basics of process definitions, the features offered by - the Process Virtual Machine and how activity implementations can be build. At the same - time the client API is shown to execute processes with those activity implementations. - </para> - - <!-- ### Activity ####################################################### --> - <section id="activity"> - <title>Activity</title> - <para>The PVM library doesn't have a fixed set of process constructs. - Instead, runtime behaviour of a node is delegated to an <literal>Activity</literal>. - In other words, <literal>Activity</literal> is an interface to implement the runtime - behaviour of process constructs in plain Java. - </para> - <programlisting>public <emphasis role="bold">interface Activity</emphasis> extends Serializable { - - void <emphasis role="bold">execute</emphasis>(ActivityExecution execution) throws Exception; - -}</programlisting> - <para>When an activity is used as the node behaviour, it is in full control of the further - propagation of the execution. In other words, a node behaviour can decide what the execution - should do next. For example, it can take a transition with - <literal>execution.take(Transition)</literal> or go into a wait state with - <literal>execution.waitForSignal()</literal>. In case the node behaviour does not invoke - any of the above execution propagation methods, the execution will - <link linkend="defaultproceedbehaviour">proceed in a default way</link>. - </para> - </section> - - <!-- ### Activity example ############################################### --> - <section id="activityexample"> - <title>Activity example</title> - <para>We'll start with a very original hello world example. A Display - activity will print a message to the console: - </para> - <programlisting>public <emphasis role="bold">class Display</emphasis> implements <emphasis role="bold">Activity</emphasis> { - - String message; - - public Display(String message) { - this.message = message; - } - - public void execute(ActivityExecution execution) { - <emphasis role="bold">System.out.println(message);</emphasis> - } -}</programlisting> - <para>Let' build our first process definition with this activity:</para> - <figure id="activity.example"> - <title>Activity example process</title> - <mediaobject><imageobject><imagedata align="center" fileref="images/process.png"/></imageobject></mediaobject> - </figure> - <programlisting>ProcessDefinition processDefinition = ProcessFactory.build() - .<emphasis role="bold">node("a").initial()</emphasis>.behaviour(<emphasis role="bold">new Display("hello")</emphasis>) - .transition().to("b") - .<emphasis role="bold">node("b")</emphasis>.behaviour(<emphasis role="bold">new Display("world")</emphasis>) -.done();</programlisting> - <para>Now we can execute this process as follows:</para> - <programlisting>Execution execution = processDefinition.startExecution();</programlisting> - <para>The invocation of <literal>startExecution</literal> will print hello world to the console:</para> - <programlisting>hello -world</programlisting> - <para>One thing already worth noticing is that activities can be configured - with properties. In the Display example, you can see that the message property - is configured differently in the two usages. With configuration properties - it becomes possible to write reusable activities. They can then be configured - differently each time they are used in a process. That is an essential part of - how process languages can be build on top of the Process Virtual Machine. - </para> - </section> - - <!-- ### ExternalActivity ############################################### --> - <section id="externalactivity"> - <title>ExternalActivity</title> - <para>External activities are activities for which the responsibility for proceeding - the execution is transferred externally, meaning outside the process system. This - means that for the system that is executing the process, it's a wait state. The - execution will wait until an external trigger is given. - </para> - <para>For dealing with external triggers, <literal>ExternalActivity</literal> - adds two methods to the <literal>Activity</literal>:</para> - <programlisting>public <emphasis role="bold">interface ExternalActivity</emphasis> extends <emphasis role="bold">Activity</emphasis> { - - void <emphasis role="bold">signal</emphasis>(Execution execution, - String signal, - Map&lt;String, Object&gt; parameters) throws Exception; - -}</programlisting> - <para>Just like with plain activities, when an execution arrives in a node, the - <literal>execute</literal>-method of the node behaviour is invoked. - In external activities, the execute method typically does something to - transfer the responsibility to another system and then enters a wait - state by invoking <literal>execution.waitForSignal()</literal>. For - example in the execute method, responsibility could be transferred to a - person by creating a task entry in a task management system and then - wait until the person completes the task. - </para> - <para>In case a node behaves as a wait state, then the execution will - wait in that node until the execution's <literal>signal</literal> method - is invoked. The execution will delegate that signal to the behaviour Activity - of the current node. - </para> - <para>So the Activity's <literal>signal</literal>-method is invoked - when the execution receives an external trigger during the wait state. With the - signal method, responsibility is transferred back to the process execution. For - example, when a person completes a task, the task management system calls the - signal method on the execution. - </para> - <para>A signal can optionally have a signal name and a map of parameters. Most - common way on how node behaviours interprete the signal and parameters is that - the signal relates to the outgoing transition that needs to be taken and that the - parameters are set as variables on the execution. But those are just examples, it - is up to the activity to use the signal and the parameters as it pleases. - </para> - </section> - - <!-- ### ExternalActivity example ####################################### --> - <section id="externalactivityexample"> - <title>ExternalActivity example</title> - <para>Here's a first example of a simple wait state implementation: - </para> - <programlisting>public <emphasis role="bold">class WaitState</emphasis> implements ExternalActivity { - - public void <emphasis role="bold">execute</emphasis>(Execution execution) { - execution.waitForSignal(); - } - - public void <emphasis role="bold">signal</emphasis>(Execution execution, - String signal, - Map&lt;String, Object&gt; parameters) { - execution.take(signal); - } -}</programlisting> - <para>The <literal>execute</literal>-method calls - <literal>execution.waitForSignal()</literal>. This call is - necessary to prevent automatic propagation of the execution. By calling - <literal>execution.waitForSignal()</literal>, the node will - behave as a wait state. - </para> - <para><literal>signal</literal>-method takes the transition with - the signal parameter as the transition name. So when an execution receives an - external trigger, the signal name is interpreted as the name of an outgoing - transition and the execution will be propagated over that transition. - </para> - <para>Here's the same simple process that has a transition from a to b. This - time, the behaviour of the two nodes will be WaitState's. - </para> - <figure id="process.diagram"> - <title>Process diagram</title> - <mediaobject><imageobject><imagedata align="center" fileref="images/process.png"/></imageobject></mediaobject> - </figure> - <programlisting>ProcessDefinition processDefinition = ProcessFactory.build() - .<emphasis role="bold">node("a").initial()</emphasis>.behaviour(<emphasis role="bold">new WaitState()</emphasis>) - .transition().to("b") - .<emphasis role="bold">node("b")</emphasis>.behaviour(<emphasis role="bold">new WaitState()</emphasis>) -.done();</programlisting> - -<programlisting>Execution execution = processDefinition.startExecution();</programlisting> -<programlisting>execution.signal();</programlisting> - </section> - - <!-- ### BASIC PROCESS EXECUTION ######################################## --> - <section id="basicprocessexecution"> - <title>Basic process execution</title> - <para>In this next example, we'll combine automatic activities and wait states. - This example is a simplified version of a loan approval process. Graphically, - it looks like this: - </para> - <figure id="the.first.graph.process"> - <title>The first graph process</title> - <mediaobject><imageobject><imagedata align="center" fileref="images/first.graph.process.png"/></imageobject></mediaobject> - </figure> - <para>Building process graphs in Java code can be tedious because you have to keep track of all the - references in local variables. To resolve that, the Process Virtual Machine comes with a - ProcessFactory. The ProcessFactory is a kind of domain specific language (DSL) that is embedded - in Java and eases the construction of process graphs. This pattern is also known as - a <ulink url="http://martinfowler.com/bliki/FluentInterface.html">fluent - interface</ulink>. - </para> - <programlisting>ProcessDefinition processDefinition = ProcessFactory.build() - .<emphasis role="bold">node("accept loan request").initial()</emphasis>.behaviour(new WaitState()) - .transition().to("loan evaluation") - .<emphasis role="bold">node("loan evaluation")</emphasis>.behaviour(new WaitState()) - .transition("approve").to("wire the money") - .transition("reject").to("end") - .<emphasis role="bold">node("wire the money")</emphasis>.behaviour(new Display("automatic payment")) - .transition().to("end") - .<emphasis role="bold">node("end")</emphasis>.behaviour(new WaitState()) -.done();</programlisting> - <para>For more details about the ProcessFactory, see the javadocs. An alternative for - the ProcessFactory would be to create an XML language and an XML parser for expressing - processes. The XML parser can then instantiate the classes of package - <literal>org.jbpm.pvm.internal.model</literal> directly. That approach is typically taken by - process languages. - </para> - <para>The node <literal>wire the money</literal> is an automatic node. The - <literal>Display</literal> implementation uses the Java API's to just print a - message to the console. But the witty reader can imagine an alternative - <literal>Activity</literal> implementation that uses the Java API of a payment - processing library to make a real automatic payment. All the other nodes are - wait states. - </para> - <para>A new execution for the process above can be started like this - </para> - <programlisting>Execution execution = processDefinition.startExecution();</programlisting> - <para>Starting a new execution implies that the initial node is executed. Since in - this case it's a wait state, the new execution will be positioned in the - node 'accept loan request' when the <literal>startExecution</literal>-method returns. - </para> - <figure id="first.execution.state.accept"> - <title>Execution positioned in 'accept loan request'</title> - <mediaobject><imageobject><imagedata align="center" fileref="images/execution.state.accept.png"/></imageobject></mediaobject> - </figure> - <para>Now we can give this execution an external trigger with the <literal>signal</literal>- - method on the execution. Invoking the signal method will take the execution to the next - wait state. - </para> - <programlisting>execution.signal();</programlisting> - <figure id="execution.state.evaluation"> - <title>Execution positioned in 'loan evaluation'</title> - <mediaobject><imageobject><imagedata align="center" fileref="images/execution.state.evaluation.png"/></imageobject></mediaobject> - </figure> - <para>Now, the execution is at an interesting point. There are two transitions out of - the state 'loan evaluation'. One transition is called 'approve' and one transition - is called 'reject'. As we explained above in the WaitState implementation, the transition - taken corresponds to the signal that is given. Let's feed in the 'approve' signal like this: - </para> - <programlisting>execution.signal("approve");</programlisting> - <para>The 'approve' signal will cause the execution to take the 'approve' transition and - it will arrive in the node 'wire the money'. - </para> - <para>In <literal>wire the money</literal>, the message will be printed to the console. - Since, the <literal>Display</literal> activity didn't invoke the - <literal>execution.waitForSignal()</literal>, nor any of the other execution propagation - methods, the default behaviour will be to just proceed. - </para> - <para><link linkend="defaultproceedbehaviour">Proceeding</link> in this case means that - the default outgoing transition is taken and the execution will arrive in the <literal>end</literal> - node, which is a wait state.</para> - <para>So only when the <literal>end</literal> wait state is reached, the <literal>signal("approve")</literal> - returns. That is because all of the things that needed to be done between the original - state and this new state could be executed by the process system. Executing till the - next wait state is the default behaviour and that behaviour can be changed with - </para> - <para><emphasis role="bold">TODO: add link to async continuations</emphasis></para> - <para>asynchronous continuations in case - transactions should not include all calculations till the next wait state. For - more about this, see <xref linkend="executionandthreads" />. - </para> - <para>Another signal invocation will bring it eventually in the end state.</para> - <figure id="execution.state.end"> - <title>Execution positioned in 'end'</title> - <mediaobject><imageobject><imagedata align="center" fileref="images/execution.state.end.png"/></imageobject></mediaobject> - </figure> - </section> - - <!-- ### MOTIVATION ##################################################### --> - <section id="motivation"> - <title>Motivation</title> - <para>There are basically two forms of process languages: graph based and composite - process languages. First of all, this design supports both. Even graph based execution - and node composition can be used in combination to implement something like UML super states. - </para> - <para>In this design, control flow activity implementations will have to be - aware of whether they are dependent on transitions (graph based) or whether they are - using the composite node structure. The goal of this design is that all non-control - flow activities can be implemented in the same way so that you can use them in graph - based process languages as well as in composite process languages. - </para> - </section> - - <!-- ### EVENTS ######################################################### --> - <section id="events"> - <title>Events</title> - <para>Events are points in the process definition to which a list of - activities can be subscribed as listeners. The motivation for events is to allow for - developers to add programming logic to a process without changing the process diagram. - This is a very valuable instrument in - facilitating the collaboration between business analysts and developers. Business analysts - are responsible for expressing the requirements. When they use a process graph to document - those requirements, developers can take this diagram and make it executable. Events can - be a very handy to insert technical details into a process (like e.g. some database insert) - in which the business analyst is not interested. - </para> - <para>Most common events are fired by the execution automatically: - </para> - <itemizedlist> - <listitem><literal>Transition.EVENT_TRANSITION_TAKE = "transition-take"</literal> : - fired on transitions when transitions are taken. - </listitem> - <listitem><literal>Node.EVENT_NODE_ENTER = "node-enter"</literal> : fired on the node - when execution enters that node. This happens when execution takes a transition - to that node, when a child node is being executed with <literal>execution.execute(Node)</literal> - or when a transition is taken from a node outside that node to a contained node. - The latter refers to super states in state machines. - </listitem> - <listitem><literal>Node.EVENT_NODE_LEAVE = "node-leave"</literal> : fired on the node - when a transition is taken out of that node or when a child node execution is - finished and the execution is propagated to the parent node. - </listitem> - <listitem><literal>ProcessDefinition.EVENT_PROCESS_START = "process-start"</literal> : fired on a process - when a new process is started. - </listitem> - <listitem><literal>ProcessDefinition.EVENT_PROCESS_END = "process-end"</literal> : fired on a process - when a new process is ended. This might include a executions that are ended - with a cancelled or error state. - </listitem> - </itemizedlist> - <para>Events are identified by the combination of a process element - and an event name. Users and process languages can also fire events - programmatically with the fire method on the Execution: - </para> - <programlisting>public interface Execution extends Serializable { - ... - void fire(String eventName, ProcessElement eventSource); - ... -}</programlisting> - - <para>A list of <link linkend="activity"><literal>Activity</literal>s</link> can be associated to an - event. But activities on events can not influence the control flow of the execution since - they are merely listeners to an execution wich is already in progress. This is different from - activities that serve as the behaviour for nodes. Node behaviour activities are responsible - for propagating the execution. So if an activity in an event invokes any of the - following methods, then it will result in an exception. - </para> - - <itemizedlist> - <listitem><literal>waitForSignal()</literal></listitem> - <listitem><literal>take(Transition)</literal></listitem> - <listitem><literal>end(*)</literal></listitem> - <listitem><literal>execute(Node)</literal></listitem> - </itemizedlist> - - <para>We'll reuse the <literal><link linkend="activity">Display</link></literal> - activity from above in a simple process: two nodes connected by a - transition. The Display listener will be subscribed to the transition event.</para> - <figure id="action.process"> - <title>The process to which a listener activity will be associated</title> - <mediaobject><imageobject><imagedata align="center" fileref="images/process.png"/></imageobject></mediaobject> - </figure> - <programlisting>ProcessDefinition processDefinition = ProcessFactory.build() - .node("a").initial().behaviour(new WaitState()) - <emphasis role="bold">.event("node-leave") - .listener(new Display("leaving a")) - .listener(new Display("second message while leaving a"))</emphasis> - .transition().to("b") - <emphasis role="bold">.listener(new Display("taking transition"))</emphasis> - .node("b").behaviour(new WaitState()) - <emphasis role="bold">.event("node-enter") - .listener(new Display("entering b"))</emphasis> -.done();</programlisting> - <para>The first event shows how to register multiple listeners to the same - event. They will be notified in the order as they are specified. - </para> - <para>Then, on the transition, there is only one type of event. So in that case, - the event type must not be specified and the listeners can be added directly on - the transition. - </para> - <para>A listeners will be called each time an execution fires the event to - which the listener is subscribed. The execution will be provided in the activity - interface as a parameter and can be used by listeners except for the methods that - control the propagation of execution. - </para> - </section> - - <!-- ### EVENT PROPAGATION ############################################## --> - <section> - <title>Event propagation</title> - <para>Events are by default propagated to enclosing process elements. The motivation - is to allow for listeners on process definitions or composite nodes that get executed - for all events that occur within that process element. For example this feature - allows to register a listener on a process definition or a composite node on - <literal>node-leave</literal> events. Such action will be executed if that node is - left. And if that listener is registered on a composite node, it will also be executed - for all nodes that are left within that composite node. - </para> - <para>To show this clearly, we'll create a <literal>DisplaySource</literal> activity - that will print the message <literal>leaving</literal> and the source of the event - to the console. - </para> - <programlisting>public class <emphasis role="bold">DisplaySource</emphasis> implements Activity { - - public void execute(Execution execution) { - <emphasis role="bold">System.out.println("leaving "+execution.getEventSource());</emphasis> - } -}</programlisting> - <para>Note that the purpose of event listeners is not to be visible, that's why the activity - itself should not be displayed in the diagram. A <literal>DisplaySource</literal> activity - will be added as a listener to the event <literal>node-leave</literal> on the composite node. - </para> - <para>The next process shows how the <literal>DisplaySource</literal> activity is registered - as a listener to to the 'node-leave' event on the <literal>composite</literal> node:</para> - <figure id="propagation.process"> - <title>A process with an invisible activity on a node-leave event on a composite node.</title> - <mediaobject><imageobject><imagedata align="center" fileref="images/propagation.process.png"/></imageobject></mediaobject> - </figure> - <programlisting>ProcessDefinition processDefinition = ProcessFactory.build("propagate") - .compositeNode("composite") - <emphasis role="bold">.event(Node.EVENT_NODE_LEAVE) - .listener(new DisplaySource())</emphasis> - .node("a").initial().behaviour(new WaitState()) - .transition().to("b") - .node("b").behaviour(new WaitState()) - .transition().to("c") - .compositeEnd() - .node("c").behaviour(new WaitState()) -.done();</programlisting> - <para>Next we'll start an execution.</para> - <programlisting>Execution execution = processDefinition.startExecution();</programlisting> - <para>After starting a new execution, the execution will be in node <literal>a</literal> as - that is the initial node. No nodes have been left so no message is logged. Next a signal - will be given to the execution, causing it to take the transition from <literal>a</literal> - to <literal>b</literal>. - </para> - <programlisting>execution.signal();</programlisting> - <para>When the signal method returns, the execution will have taken the transition and - the node-leave event will be fired on node a. That event will be propagated to the - composite node and to the process definition. Since our propagation logger is placed - on node composite it will receive the event and print the following message: - </para> - <programlisting>leaving node(a)</programlisting> - <para>Another</para> - <programlisting>execution.signal();</programlisting> - <para>will take the transition from b to c. That will fire two node-leave events. One on - node b and one on node composite. So the following lines will be appended to the console - output:</para> - <programlisting>leaving node(b) -leaving node(composite)</programlisting> - <para>Event propagation is build on the hierarchical composition structure of the process - definition. The top level element is always the process definition. The process - definition contains a list of nodes. Each node can be a leaf node or it can be a - composite node, which means that it contains a list of nested nodes. Nested nodes - can be used for e.g. super states or composite activities in nested process languages like BPEL. - </para> - <para>So the even model also works similarly for composite nodes as it did for the process - definition above. Suppose that 'Phase one' models - a super state as in state machines. Then event propagation allows to subscribe to all events - within that super state. The idea is that the hierarchical composition corresponds to - diagram representation. If an element 'e' is drawn inside another element 'p', then p - is the parent of e. A process definition has a set of top level nodes. Every node can have - a set of nested nodes. The parent of a transition is considered as the first common - parent for it's source and destination. - </para> - <para>If an event listener is not interested in propagated events, propagation can be disabled - with <literal>propagationDisabled()</literal>. The next process is the same process - as above except that propagated events will be disabled on the event listener. The graph diagram - remains the same. - </para> - <figure id="propagation.disabled.process"> - <title>A process with a listener to 'node-leave' events with propagation disabled.</title> - <mediaobject><imageobject><imagedata align="center" fileref="images/propagation.process.png"/></imageobject></mediaobject> - </figure> - <para>Building the process with the process factory: - </para> - <programlisting>ProcessDefinition processDefinition = ProcessFactory.build("propagate") - .compositeNode("composite") - <emphasis role="bold">.event(Node.EVENT_NODE_LEAVE) - .listener(new DisplaySource()) - .propagationDisabled()</emphasis> - .node("a").initial().behaviour(new WaitState()) - .transition().to("b") - .node("b").behaviour(new WaitState()) - .transition().to("c") - .nodesEnd() - .node("c").behaviour(new WaitState()) -.done();</programlisting> - <para>So when the first signal is given for this process, again the node-leave event will be - fired on node a, but now the listener on the composite node will not be executed cause - propagated events have been disabled. Disabling propagation is a property on the listener - and doesn't influence the other listeners. The event will always be fired and propagated - over the whole parent hierarchy. - </para> - <programlisting>Execution execution = processDefinition.startExecution(); -execution.signal(); -</programlisting> - <para>Next, the second signal will take the transition from b to c. - </para> - <programlisting>execution.signal()</programlisting> - <para>Again two node-leave - events are fired just like above on nodes b and composite respectively. The first event - is the node-leave event on node b. That will be propagated to the composite node. So the - listener will not be executed for this event cause it has propagation disabled. But the - listener will be executed for the node-leave event on the composite node. That is not - propagated, but fired directly on the composite node. So the listener will now be executed - only once for the composite node as shown in the following console output: - </para> - <programlisting>leaving node(composite)</programlisting> - </section> - - <!-- ### PROCESS STRUCTURE ############################################## --> - <section> - <title>Process structure</title> - <para>Above we already touched briefly on the two main process constructs: - Nodes, transitions and node composition. This section will elaborate - on all the basic combination possibilities. - </para> - <figure id="process.structure"> - <title>UML class diagram of the basic process structure</title> - <mediaobject><imageobject><imagedata align="center" fileref="images/process.structure.classes.png"/></imageobject></mediaobject> - </figure> - <para>Next is a series of example diagram structures that can be formed - with the PVM process model. - </para> - <figure id="transition"> - <title>Any two nodes can be connected with a transition.</title> - <mediaobject><imageobject><imagedata align="center" fileref="images/transition.png"/></imageobject></mediaobject> - </figure> - <figure id="self.transition"> - <title>A self transition.</title> - <mediaobject><imageobject><imagedata align="center" fileref="images/self.transition.png"/></imageobject></mediaobject> - </figure> - <figure id="composite.node"> - <title>Composite node is a list of nested nodes.</title> - <mediaobject><imageobject><imagedata align="center" fileref="images/composite.node.png"/></imageobject></mediaobject> - </figure> - <figure id="transition.into.composite"> - <title>Transition to a node inside a composite.</title> - <mediaobject><imageobject><imagedata align="center" fileref="images/transition.into.composite.png"/></imageobject></mediaobject> - </figure> - <figure id="transition.out.of.composite"> - <title>Transition from a node inside a composite to a node outside the composite.</title> - <mediaobject><imageobject><imagedata align="center" fileref="images/transition.out.of.composite.png"/></imageobject></mediaobject> - </figure> - <figure id="transition.inheritence"> - <title>Transition of composite nodes are inherited. The node inside can take the transition of the composite node.</title> - <mediaobject><imageobject><imagedata align="center" fileref="images/transition.inheritence.png"/></imageobject></mediaobject> - </figure> - <figure id="transition.to.outer"> - <title>Transition from a node to an outer composite.</title> - <mediaobject><imageobject><imagedata align="center" fileref="images/transition.to.outer.png"/></imageobject></mediaobject> - </figure> - <figure id="transition.to.inner"> - <title>Transition from a composite node to an inner composed node.</title> - <mediaobject><imageobject><imagedata align="center" fileref="images/transition.to.inner.png"/></imageobject></mediaobject> - </figure> - <figure id="initial.in.composite"> - <title>An initial node inside a composite node.</title> - <mediaobject><imageobject><imagedata align="center" fileref="images/initial.in.composite.png"/></imageobject></mediaobject> - </figure> - </section> - -</chapter> Copied: jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/modules/ch04-ImplementingBasicActivities.xml (from rev 2276, jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/modules/ch04-BasicProcessExecutionConcepts.xml) =================================================================== --- jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/modules/ch04-ImplementingBasicActivities.xml (rev 0) +++ jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/modules/ch04-ImplementingBasicActivities.xml 2008-09-18 13:55:29 UTC (rev 2279) @@ -0,0 +1,576 @@ +<chapter id="implementingbasicactivities"> + <title>Implementing basic activities</title> + + <para>This chapter explains the basics of process definitions, the features offered by + the Process Virtual Machine and how activity implementations can be build. At the same + time the client API is shown to execute processes with those activity implementations. + </para> + + <!-- ### Activity ####################################################### --> + <section id="activity"> + <title>Activity</title> + <para>The PVM library doesn't have a fixed set of process constructs. + Instead, runtime behaviour of a node is delegated to an <literal>Activity</literal>. + In other words, <literal>Activity</literal> is an interface to implement the runtime + behaviour of process constructs in plain Java. + </para> + <programlisting>public <emphasis role="bold">interface Activity</emphasis> extends Serializable { + + void <emphasis role="bold">execute</emphasis>(ActivityExecution execution) throws Exception; + +}</programlisting> + <para>When an activity is used as the node behaviour, it is in full control of the further + propagation of the execution. In other words, a node behaviour can decide what the execution + should do next. For example, it can take a transition with + <literal>execution.take(Transition)</literal> or go into a wait state with + <literal>execution.waitForSignal()</literal>. In case the node behaviour does not invoke + any of the above execution propagation methods, the execution will + <link linkend="implicitproceedbehaviour">proceed in a default way</link>. + </para> + </section> + + <!-- ### Activity example ############################################### --> + <section id="activityexample"> + <title>Activity example</title> + <para>We'll start with a very original hello world example. A Display + activity will print a message to the console: + </para> + <programlisting>public <emphasis role="bold">class Display</emphasis> implements <emphasis role="bold">Activity</emphasis> { + + String message; + + public Display(String message) { + this.message = message; + } + + public void execute(ActivityExecution execution) { + <emphasis role="bold">System.out.println(message);</emphasis> + } +}</programlisting> + <para>Let' build our first process definition with this activity:</para> + <figure id="activity.example"> + <title>Display example process</title> + <mediaobject><imageobject><imagedata align="center" fileref="images/process.ab.png"/></imageobject></mediaobject> + </figure> + <programlisting>ClientProcessDefinition processDefinition = ProcessFactory.build() + .<emphasis role="bold">node("a").initial()</emphasis>.behaviour(<emphasis role="bold">new Display("hello")</emphasis>) + .transition().to("b") + .<emphasis role="bold">node("b")</emphasis>.behaviour(<emphasis role="bold">new Display("world")</emphasis>) +.done();</programlisting> + <para>Now we can execute this process as follows:</para> + <programlisting>Execution execution = processDefinition.startExecution();</programlisting> + <para>The invocation of <literal>startExecution</literal> will print hello world to the console:</para> + <programlisting>hello +world</programlisting> + <para>One thing already worth noticing is that activities can be configured + with properties. In the Display example, you can see that the message property + is configured differently in the two usages. With configuration properties + it becomes possible to write reusable activities. They can then be configured + differently each time they are used in a process. That is an essential part of + how process languages can be build on top of the Process Virtual Machine. + </para> + <para>The other part that needs explanation is that this activity + implementation does not contain any instructions for the propagation of the + execution. When a new process instance is started, the execution is positioned + in the initial node and that activity is executed. The method + <literal>Display.execute</literal> makes use of what is called implicit propagation + of execution. Concretely this means that the activity itself does not + invoke any of the methods on the execution to propagate it. In that case + implicit propagation kicks in. Implicit propagation will take the first + transition if there is one. If not, it will end the execution. This explains + why both nodes <literal>a</literal> and <literal>b</literal> are executed and that + the execution stops after node <literal>b</literal> is executed. + </para> + <para>More details about the implicit proceed behaviour can be found + in <xref linkend="implicitproceedbehaviour" /></para> + </section> + + <!-- ### ExternalActivity ############################################### --> + <section id="externalactivity"> + <title>ExternalActivity</title> + <para>External activities are activities for which the responsibility for proceeding + the execution is transferred externally, meaning outside the process system. This + means that for the system that is executing the process, it's a wait state. The + execution will wait until an external trigger is given. + </para> + <para>For dealing with external triggers, <literal>ExternalActivity</literal> + adds two methods to the <literal>Activity</literal>:</para> + <programlisting>public <emphasis role="bold">interface ExternalActivity</emphasis> extends <emphasis role="bold">Activity</emphasis> { + + void <emphasis role="bold">signal</emphasis>(Execution execution, + String signal, + Map&lt;String, Object&gt; parameters) throws Exception; + +}</programlisting> + <para>Just like with plain activities, when an execution arrives in a node, the + <literal>execute</literal>-method of the node behaviour is invoked. + In external activities, the execute method typically does something to + transfer the responsibility to another system and then enters a wait + state by invoking <literal>execution.waitForSignal()</literal>. For + example in the execute method, responsibility could be transferred to a + person by creating a task entry in a task management system and then + wait until the person completes the task. + </para> + <para>In case a node behaves as a wait state, then the execution will + wait in that node until the execution's <literal>signal</literal> method + is invoked. The execution will delegate that signal to the behaviour Activity + of the current node. + </para> + <para>So the Activity's <literal>signal</literal>-method is invoked + when the execution receives an external trigger during the wait state. With the + signal method, responsibility is transferred back to the process execution. For + example, when a person completes a task, the task management system calls the + signal method on the execution. + </para> + <para>A signal can optionally have a signal name and a map of parameters. Most + common way on how node behaviours interprete the signal and parameters is that + the signal relates to the outgoing transition that needs to be taken and that the + parameters are set as variables on the execution. But those are just examples, it + is up to the activity to use the signal and the parameters as it pleases. + </para> + </section> + + <!-- ### ExternalActivity example ####################################### --> + <section id="externalactivityexample"> + <title>ExternalActivity example</title> + <para>Here's a first example of a simple wait state implementation: + </para> + <programlisting>public <emphasis role="bold">class WaitState</emphasis> implements <emphasis role="bold">ExternalActivity</emphasis> { + + public void execute(ActivityExecution execution) { + <emphasis role="bold">execution.waitForSignal();</emphasis> + } + + public void signal(ActivityExecution execution, + String signalName, + Map&lt;String, Object&gt; parameters) { + <emphasis role="bold">execution.take(signalName);</emphasis> + } +}</programlisting> + <para>The <literal>execute</literal>-method calls + <literal>execution.waitForSignal()</literal>. The invocation of + <literal>execution.waitForSignal()</literal> will bring the process execution + into a wait state until an external trigger is given. + </para> + <para><literal>signal</literal>-method takes the transition with + the signal parameter as the transition name. So when an execution receives an + external trigger, the signal name is interpreted as the name of an outgoing + transition and the execution will be propagated over that transition. + </para> + <para>Here's the same simple process that has a transition from a to b. This + time, the behaviour of the two nodes will be WaitState's. + </para> + <figure id="process.diagram"> + <title>The external activity example process</title> + <mediaobject><imageobject><imagedata align="center" fileref="images/process.ab.png"/></imageobject></mediaobject> + </figure> + <programlisting>ClientProcessDefinition processDefinition = ProcessFactory.build() + <emphasis role="bold">.node("a").initial().behaviour(new WaitState())</emphasis> + .transition().to("b") + <emphasis role="bold">.node("b").behaviour(new WaitState())</emphasis> +.done();</programlisting> + <para>Let's start a new process instance for this process definition:</para> + <programlisting>ClientExecution execution = processDefinition.startProcessInstance();</programlisting> + <para>Starting this process will execute the <literal>WaitState</literal> activity + in node <literal>a</literal>. <literal>WaitState.execute</literal> will invoke + <literal>ActivityExecution.waitForSignal</literal>. So when the + <literal>processDefinition.startProcessInstance()</literal> returns, the execution + will still be positioned in node a. + </para> + <programlisting>assertEquals("a", execution.getNodeName());</programlisting> + <para>Then we provide the external trigger by calling the <literal>signal</literal> + method. + </para> + <programlisting>execution.signal();</programlisting> + <para>The <literal>execution.signal()</literal> will delegate to the activity + of the current node. So in this case that is the <literal>WaitState</literal> + activity in node <literal>a</literal>. The <literal>WaitState.signal</literal> + will invoke the <literal>ActivityExecution.take(String transitionName)</literal>. + Since we didn't supply a signalName, the first transition with name <literal>null</literal> + will be taken. The only transition we specified out of node <literal>a</literal> + didn't get a name so that one will be taken. And that transition points to node + <literal>b</literal>. When the execution arrives in node <literal>b</literal>, + the <literal>WaitState</literal> in node <literal>b</literal> is executed. + Similar as we saw above, the execution will wait in node <literal>b</literal> + and this time the <literal>signal</literal> method will return, leaving the + execution positioned in node <literal>b</literal>. + </para> + <programlisting>assertEquals("b", execution.getNodeName());</programlisting> + </section> + + <!-- ### BASIC PROCESS EXECUTION ######################################## --> + <section id="basicprocessexecution"> + <title>Basic process execution</title> + <para>In this next example, we'll combine automatic activities and wait states. + This example builds upon the loan approval process with the <literal>WaitState</literal> + and <literal>Display</literal> activities that we've just created. Graphically, + the loan process looks like this: + </para> + <figure id="basicprocessexecution.loan.process"> + <title>The loan process</title> + <mediaobject><imageobject><imagedata align="center" fileref="images/process.loan.png"/></imageobject></mediaobject> + </figure> + <para>Building process graphs in Java code can be tedious because you have to keep track of all the + references in local variables. To resolve that, the Process Virtual Machine comes with a + ProcessFactory. The ProcessFactory is a kind of domain specific language (DSL) that is embedded + in Java and eases the construction of process graphs. This pattern is also known as + a <ulink url="http://martinfowler.com/bliki/FluentInterface.html">fluent + interface</ulink>. + </para> + <programlisting>ClientProcessDefinition processDefinition = ProcessFactory.build("loan") + .node(<emphasis role="bold">"submit loan request"</emphasis>).initial().behaviour(new Display("loan request submitted")) + .transition().to("evaluate") + <emphasis role="bold">.node("evaluate").behaviour(new WaitState())</emphasis> + .transition("approve").to("wire money") + .transition("reject").to("end") + <emphasis role="bold">.node("wire money").behaviour(new Display("wire the money"))</emphasis> + .transition().to("archive") + <emphasis role="bold">.node("archive").behaviour(new WaitState())</emphasis> + .transition().to("end") + <emphasis role="bold">.node("end").behaviour(new WaitState())</emphasis> +.done();</programlisting> + <para>For more details about the ProcessFactory, see <ulink url="../../api/org/jbpm/pvm/package-summary.html">the + api docs</ulink>. An alternative for + the ProcessFactory would be to create an XML language and an XML parser for expressing + processes. The XML parser can then instantiate the classes of package + <literal>org.jbpm.pvm.internal.model</literal> directly. That approach is typically taken by + process languages. + </para> + <para>The initial node <literal>submit loan request</literal> and the node + <literal>wire the money</literal> are automatic nodes. In this example, + the <literal>Display</literal> implementation of node + <literal>wire the money</literal> uses the Java API's to just print a + message to the console. But the witty reader can imagine an alternative + <literal>Activity</literal> implementation that uses the Java API of a payment + processing library to make a real automatic payment. + </para> + <para>A new execution for the process above can be started like this + </para> + <programlisting>ClientExecution execution = processDefinition.startProcessInstance();</programlisting> + <para>When the <literal>startExecution</literal>-method returns, the node + <literal>submit loan request</literal> will be executed and the execution will be + positioned in the node <literal>evaluate</literal>. + </para> + <figure id="execution.loan.evaluate"> + <title>Execution positioned in the 'evaluate' node</title> + <mediaobject><imageobject><imagedata align="center" fileref="images/execution.loan.evaluate.png"/></imageobject></mediaobject> + </figure> + <para>Now, the execution is at an interesting point. There are two transitions out of + the state <literal>evaluate</literal>. One transition is called <literal>approve</literal> + and one transition is called <literal>reject</literal>. As we explained above, the WaitState + implementation will take the transition that corresponds to the signal that is given. + Let's feed in the 'approve' signal like this: + </para> + <programlisting>execution.signal("approve");</programlisting> + <para>The <literal>approve</literal> signal will cause the execution to take the <literal>approve</literal> + transition and it will arrive in the node <literal>wire money</literal>. + </para> + <para>In node <literal>wire money</literal>, the message will be printed to the console. + Since, the <literal>Display</literal> activity didn't invoke the + <literal>execution.waitForSignal()</literal>, nor any of the other execution propagation + methods, the implicit proceed behaviour will just make the execution continue + over the outgoing transition to node <literal>archive</literal>, which is again + a <literal>WaitState</literal>. + </para> + <figure id="execution.loan.archive"> + <title>Execution positioned in 'archive' node</title> + <mediaobject><imageobject><imagedata align="center" fileref="images/execution.loan.archive.png"/></imageobject></mediaobject> + </figure> + <para>So only when the <literal>archive</literal> wait state is reached, + the <literal>signal("approve")</literal> returns. + </para> + <para>Another signal like this:</para> + <programlisting>execution.signal("approve");</programlisting> + <para>will bring the execution eventually in the end state.</para> + <figure id="execution.loan.end"> + <title>Execution positioned in the 'end' node</title> + <mediaobject><imageobject><imagedata align="center" fileref="images/execution.loan.end.png"/></imageobject></mediaobject> + </figure> + </section> + + <!-- ### MOTIVATION ##################################################### --> + <section id="motivation"> + <title>Motivation</title> + <para>There are basically two forms of process languages: graph based and composite + process languages. First of all, this design supports both. Even graph based execution + and node composition can be used in combination to implement something like UML super states. + </para> + <para>In this design, control flow activity implementations will have to be + aware of whether they are dependent on transitions (graph based) or whether they are + using the composite node structure. The goal of this design is that all non-control + flow activities can be implemented in the same way so that you can use them in graph + based process languages as well as in composite process languages. + </para> + </section> + + <!-- ### EVENTS ######################################################### --> + <section id="events"> + <title>Events</title> + <para>Events are points in the process definition to which a list of + activities can be subscribed as listeners. The motivation for events is to allow for + developers to add programming logic to a process without changing the process diagram. + This is a very valuable instrument in + facilitating the collaboration between business analysts and developers. Business analysts + are responsible for expressing the requirements. When they use a process graph to document + those requirements, developers can take this diagram and make it executable. Events can + be a very handy to insert technical details into a process (like e.g. some database insert) + in which the business analyst is not interested. + </para> + <para>Most common events are fired by the execution automatically: + </para> + <itemizedlist> + <listitem><literal>Transition.EVENT_TRANSITION_TAKE = "transition-take"</literal> : + fired on transitions when transitions are taken. + </listitem> + <listitem><literal>Node.EVENT_NODE_ENTER = "node-enter"</literal> : fired on the node + when execution enters that node. This happens when execution takes a transition + to that node, when a child node is being executed with <literal>execution.execute(Node)</literal> + or when a transition is taken from a node outside that node to a contained node. + The latter refers to super states in state machines. + </listitem> + <listitem><literal>Node.EVENT_NODE_LEAVE = "node-leave"</literal> : fired on the node + when a transition is taken out of that node or when a child node execution is + finished and the execution is propagated to the parent node. + </listitem> + <listitem><literal>ProcessDefinition.EVENT_PROCESS_START = "process-start"</literal> : fired on a process + when a new process is started. + </listitem> + <listitem><literal>ProcessDefinition.EVENT_PROCESS_END = "process-end"</literal> : fired on a process + when a new process is ended. This might include a executions that are ended + with a cancelled or error state. + </listitem> + </itemizedlist> + <para>Events are identified by the combination of a process element + and an event name. Users and process languages can also fire events + programmatically with the fire method on the Execution: + </para> + <programlisting>public interface Execution extends Serializable { + ... + void fire(String eventName, ProcessElement eventSource); + ... +}</programlisting> + + <para>A list of <link linkend="activity"><literal>Activity</literal>s</link> can be associated to an + event. But activities on events can not influence the control flow of the execution since + they are merely listeners to an execution wich is already in progress. This is different from + activities that serve as the behaviour for nodes. Node behaviour activities are responsible + for propagating the execution. So if an activity in an event invokes any of the + following methods, then it will result in an exception. + </para> + + <itemizedlist> + <listitem><literal>waitForSignal()</literal></listitem> + <listitem><literal>take(Transition)</literal></listitem> + <listitem><literal>end(*)</literal></listitem> + <listitem><literal>execute(Node)</literal></listitem> + </itemizedlist> + + <para>We'll reuse the <literal><link linkend="activity">Display</link></literal> + activity from above in a simple process: two nodes connected by a + transition. The Display listener will be subscribed to the transition event.</para> + <figure id="action.process"> + <title>The process to which a listener activity will be associated</title> + <mediaobject><imageobject><imagedata align="center" fileref="images/process.png"/></imageobject></mediaobject> + </figure> + <programlisting>ProcessDefinition processDefinition = ProcessFactory.build() + .node("a").initial().behaviour(new WaitState()) + <emphasis role="bold">.event("node-leave") + .listener(new Display("leaving a")) + .listener(new Display("second message while leaving a"))</emphasis> + .transition().to("b") + <emphasis role="bold">.listener(new Display("taking transition"))</emphasis> + .node("b").behaviour(new WaitState()) + <emphasis role="bold">.event("node-enter") + .listener(new Display("entering b"))</emphasis> +.done();</programlisting> + <para>The first event shows how to register multiple listeners to the same + event. They will be notified in the order as they are specified. + </para> + <para>Then, on the transition, there is only one type of event. So in that case, + the event type must not be specified and the listeners can be added directly on + the transition. + </para> + <para>A listeners will be called each time an execution fires the event to + which the listener is subscribed. The execution will be provided in the activity + interface as a parameter and can be used by listeners except for the methods that + control the propagation of execution. + </para> + </section> + + <!-- ### EVENT PROPAGATION ############################################## --> + <section> + <title>Event propagation</title> + <para>Events are by default propagated to enclosing process elements. The motivation + is to allow for listeners on process definitions or composite nodes that get executed + for all events that occur within that process element. For example this feature + allows to register a listener on a process definition or a composite node on + <literal>node-leave</literal> events. Such action will be executed if that node is + left. And if that listener is registered on a composite node, it will also be executed + for all nodes that are left within that composite node. + </para> + <para>To show this clearly, we'll create a <literal>DisplaySource</literal> activity + that will print the message <literal>leaving</literal> and the source of the event + to the console. + </para> + <programlisting>public class <emphasis role="bold">DisplaySource</emphasis> implements Activity { + + public void execute(Execution execution) { + <emphasis role="bold">System.out.println("leaving "+execution.getEventSource());</emphasis> + } +}</programlisting> + <para>Note that the purpose of event listeners is not to be visible, that's why the activity + itself should not be displayed in the diagram. A <literal>DisplaySource</literal> activity + will be added as a listener to the event <literal>node-leave</literal> on the composite node. + </para> + <para>The next process shows how the <literal>DisplaySource</literal> activity is registered + as a listener to to the 'node-leave' event on the <literal>composite</literal> node:</para> + <figure id="propagation.process"> + <title>A process with an invisible activity on a node-leave event on a composite node.</title> + <mediaobject><imageobject><imagedata align="center" fileref="images/propagation.process.png"/></imageobject></mediaobject> + </figure> + <programlisting>ProcessDefinition processDefinition = ProcessFactory.build("propagate") + .compositeNode("composite") + <emphasis role="bold">.event(Node.EVENT_NODE_LEAVE) + .listener(new DisplaySource())</emphasis> + .node("a").initial().behaviour(new WaitState()) + .transition().to("b") + .node("b").behaviour(new WaitState()) + .transition().to("c") + .compositeEnd() + .node("c").behaviour(new WaitState()) +.done();</programlisting> + <para>Next we'll start an execution.</para> + <programlisting>Execution execution = processDefinition.startExecution();</programlisting> + <para>After starting a new execution, the execution will be in node <literal>a</literal> as + that is the initial node. No nodes have been left so no message is logged. Next a signal + will be given to the execution, causing it to take the transition from <literal>a</literal> + to <literal>b</literal>. + </para> + <programlisting>execution.signal();</programlisting> + <para>When the signal method returns, the execution will have taken the transition and + the node-leave event will be fired on node a. That event will be propagated to the + composite node and to the process definition. Since our propagation logger is placed + on node composite it will receive the event and print the following message: + </para> + <programlisting>leaving node(a)</programlisting> + <para>Another</para> + <programlisting>execution.signal();</programlisting> + <para>will take the transition from b to c. That will fire two node-leave events. One on + node b and one on node composite. So the following lines will be appended to the console + output:</para> + <programlisting>leaving node(b) +leaving node(composite)</programlisting> + <para>Event propagation is build on the hierarchical composition structure of the process + definition. The top level element is always the process definition. The process + definition contains a list of nodes. Each node can be a leaf node or it can be a + composite node, which means that it contains a list of nested nodes. Nested nodes + can be used for e.g. super states or composite activities in nested process languages like BPEL. + </para> + <para>So the even model also works similarly for composite nodes as it did for the process + definition above. Suppose that 'Phase one' models + a super state as in state machines. Then event propagation allows to subscribe to all events + within that super state. The idea is that the hierarchical composition corresponds to + diagram representation. If an element 'e' is drawn inside another element 'p', then p + is the parent of e. A process definition has a set of top level nodes. Every node can have + a set of nested nodes. The parent of a transition is considered as the first common + parent for it's source and destination. + </para> + <para>If an event listener is not interested in propagated events, propagation can be disabled + with <literal>propagationDisabled()</literal>. The next process is the same process + as above except that propagated events will be disabled on the event listener. The graph diagram + remains the same. + </para> + <figure id="propagation.disabled.process"> + <title>A process with a listener to 'node-leave' events with propagation disabled.</title> + <mediaobject><imageobject><imagedata align="center" fileref="images/propagation.process.png"/></imageobject></mediaobject> + </figure> + <para>Building the process with the process factory: + </para> + <programlisting>ProcessDefinition processDefinition = ProcessFactory.build("propagate") + .compositeNode("composite") + <emphasis role="bold">.event(Node.EVENT_NODE_LEAVE) + .listener(new DisplaySource()) + .propagationDisabled()</emphasis> + .node("a").initial().behaviour(new WaitState()) + .transition().to("b") + .node("b").behaviour(new WaitState()) + .transition().to("c") + .nodesEnd() + .node("c").behaviour(new WaitState()) +.done();</programlisting> + <para>So when the first signal is given for this process, again the node-leave event will be + fired on node a, but now the listener on the composite node will not be executed cause + propagated events have been disabled. Disabling propagation is a property on the listener + and doesn't influence the other listeners. The event will always be fired and propagated + over the whole parent hierarchy. + </para> + <programlisting>Execution execution = processDefinition.startExecution(); +execution.signal(); +</programlisting> + <para>Next, the second signal will take the transition from b to c. + </para> + <programlisting>execution.signal()</programlisting> + <para>Again two node-leave + events are fired just like above on nodes b and composite respectively. The first event + is the node-leave event on node b. That will be propagated to the composite node. So the + listener will not be executed for this event cause it has propagation disabled. But the + listener will be executed for the node-leave event on the composite node. That is not + propagated, but fired directly on the composite node. So the listener will now be executed + only once for the composite node as shown in the following console output: + </para> + <programlisting>leaving node(composite)</programlisting> + </section> + + <!-- ### PROCESS STRUCTURE ############################################## --> + <section> + <title>Process structure</title> + <para>Above we already touched briefly on the two main process constructs: + Nodes, transitions and node composition. This section will elaborate + on all the basic combination possibilities. + </para> + <figure id="process.structure"> + <title>UML class diagram of the basic process structure</title> + <mediaobject><imageobject><imagedata align="center" fileref="images/process.structure.classes.png"/></imageobject></mediaobject> + </figure> + <para>Next is a series of example diagram structures that can be formed + with the PVM process model. + </para> + <figure id="transition"> + <title>Any two nodes can be connected with a transition.</title> + <mediaobject><imageobject><imagedata align="center" fileref="images/transition.png"/></imageobject></mediaobject> + </figure> + <figure id="self.transition"> + <title>A self transition.</title> + <mediaobject><imageobject><imagedata align="center" fileref="images/self.transition.png"/></imageobject></mediaobject> + </figure> + <figure id="composite.node"> + <title>Composite node is a list of nested nodes.</title> + <mediaobject><imageobject><imagedata align="center" fileref="images/composite.node.png"/></imageobject></mediaobject> + </figure> + <figure id="transition.into.composite"> + <title>Transition to a node inside a composite.</title> + <mediaobject><imageobject><imagedata align="center" fileref="images/transition.into.composite.png"/></imageobject></mediaobject> + </figure> + <figure id="transition.out.of.composite"> + <title>Transition from a node inside a composite to a node outside the composite.</title> + <mediaobject><imageobject><imagedata align="center" fileref="images/transition.out.of.composite.png"/></imageobject></mediaobject> + </figure> + <figure id="transition.inheritence"> + <title>Transition of composite nodes are inherited. The node inside can take the transition of the composite node.</title> + <mediaobject><imageobject><imagedata align="center" fileref="images/transition.inheritence.png"/></imageobject></mediaobject> + </figure> + <figure id="transition.to.outer"> + <title>Transition from a node to an outer composite.</title> + <mediaobject><imageobject><imagedata align="center" fileref="images/transition.to.outer.png"/></imageobject></mediaobject> + </figure> + <figure id="transition.to.inner"> + <title>Transition from a composite node to an inner composed node.</title> + <mediaobject><imageobject><imagedata align="center" fileref="images/transition.to.inner.png"/></imageobject></mediaobject> + </figure> + <figure id="initial.in.composite"> + <title>An initial node inside a composite node.</title> + <mediaobject><imageobject><imagedata align="center" fileref="images/initial.in.composite.png"/></imageobject></mediaobject> + </figure> + </section> + +</chapter> Added: jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/modules/ch05-ProcessAnatomy.xml =================================================================== --- jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/modules/ch05-ProcessAnatomy.xml (rev 0) +++ jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/modules/ch05-ProcessAnatomy.xml 2008-09-18 13:55:29 UTC (rev 2279) @@ -0,0 +1,4 @@ +<chapter id="timers"> + <title>Timers</title> + <para>TODO</para> +</chapter> \ No newline at end of file Copied: jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/modules/ch06-ImplementingAdvancedActivities.xml (from rev 2276, jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/modules/chxx-AdvancedGraphExecution.xml) =================================================================== --- jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/modules/ch06-ImplementingAdvancedActivities.xml (rev 0) +++ jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/modules/ch06-ImplementingAdvancedActivities.xml 2008-09-18 13:55:29 UTC (rev 2279) @@ -0,0 +1,323 @@ +<chapter id="advancedgraphexecution"> + <title>Advanced graph execution</title> + + <!-- ### LOOPS ########################################################## --> + <section> + <title>Loops</title> + <para>Loops can be based on transitions or on node composition. Loops can contain wait + states. + </para> + <para>To support high numbers of automatic loop executions, the Process Virtual Machine + tranformed the propagation of execution from tail recursion to a while loop. This means + that all the methods in the <literal>Execution</literal> class that propagate the + execution like <literal>take</literal> or <literal>execute</literal> will not be + executed when you call them. Instead, the method invocations will be appended to a list. + The first invocation of such a method will start a loop that will execute all invocations + till that list is empty. These invocations are called atomic operations. + </para> + </section> + + <!-- ### SUB PROCESSES ################################################## --> + <section> + <title>Sub processes</title> + <para>TODO: sub processes</para> + </section> + + <!-- ### DEFAULT PROCEED BEHAVIOUR ###################################### --> + <section id="implicitproceedbehaviour"> + <title>Implicit proceed behaviour</title> + <para>When an <literal>Activity</literal> is used as node behaviour, it can + explicitely propagate the execution with following methods: + </para> + <itemizedlist> + <listitem><literal>waitForSignal()</literal></listitem> + <listitem><literal>take(Transition)</literal></listitem> + <listitem><literal>end(*)</literal></listitem> + <listitem><literal>execute(Node)</literal></listitem> + <listitem><literal>createExecution(*)</literal></listitem> + </itemizedlist> + <para>When <literal>Activity</literal> implementations used for node behviour + don't call any of the following execution propagation methods, then, after + the activity is executed, the execution will apply the implicit proceed behaviour. + </para> + <para>The implicit proceed behaviour is defined as follows:</para> + <itemizedlist> + <listitem>If the current node has a default outgoing transition, take it.</listitem> + <listitem>If the current node has a parent node, move back to the parent node.</listitem> + <listitem>Otherwise, end this execution.</listitem> + </itemizedlist> + <para>Process languages can overwrite the implicit proceed behaviour + by overriding the <literal>proceed</literal> method in + <literal>ExecutionImpl</literal>. + </para> + </section> + + <!-- ### FUNCTIONAL ACTIVITIES ################################ --> + <section id="functionalactivities"> + <title>Functional activities</title> + <para>Activities that also can be used as event listeners are called functional + activities. Examples of automatic activities are sending an email, doing a database + update, generating a pdf, calculating an average, etc. All of these are automatic + activities that do not change the execution flow. Here's how such activities can + be implemented: + </para> + <programlisting>public class FunctionalActivity implements Activity, EventListener { + public void execute(ActivityExecution execution) { + perform(execution); + } + public void notify(EventListenerExecution execution) { + perform(execution); + } + void perform(OpenExecution execution) { + ...do functional work... + } + }</programlisting> + <para>None of these methods actually invoke execution propagation methods. + So after the perform method is completed, the execution will + <link linkend="implicitproceedbehaviour">proceed in the default way</link>. + </para> + </section> + + + <!-- ### EXECUTION AND THREADS ########################################## --> + <section id="executionandthreads"> + <title>Execution and threads</title> + <para>This section explains how the Process Virtual Machine boroughs the thread + from the client to bring an execution from one wait state to another. + </para> + <para>When a client invokes a method (like e.g. the signal method) on an execution, + by default, the Process Virtual Machine will use that thread to progress the execution + until it reached a wait state. Once the next wait state has been reached, the + method returns and the client gets the thread back. This is the default way + for the Process Virtual Machine to operate. Two more levels of asynchonous + execution complement this default behaviour: + <link linkend="asynchronouscontinuations">Asynchronous continuations</link> + and the <link linkend="architecture">asynchronous command service</link>. + </para> + <para>The next process will show the basics concretely. It has three wait states + and four automatic nodes. + </para> + <figure id="automatic"> + <title>Process with many sequential automatic activities.</title> + <mediaobject><imageobject><imagedata align="center" fileref="images/ch04.automatic.png"/></imageobject></mediaobject> + </figure> + <para>Here's how to build the process:</para> + <programlisting>ProcessDefinition processDefinition = ProcessFactory.build("automatic") + .<emphasis role="bold">node("wait 1").initial()</emphasis>.behaviour(new <emphasis role="bold">WaitState</emphasis>()) + .transition().to("automatic 1") + .<emphasis role="bold">node("automatic 1")</emphasis>.behaviour(new <emphasis role="bold">Display("one")</emphasis>) + .transition().to("wait 2") + .<emphasis role="bold">node("wait 2")</emphasis>.behaviour(new <emphasis role="bold">WaitState</emphasis>()) + .transition().to("automatic 2") + .<emphasis role="bold">node("automatic 2")</emphasis>.behaviour(new <emphasis role="bold">Display("two")</emphasis>) + .transition().to("automatic 3") + .<emphasis role="bold">node("automatic 3")</emphasis>.behaviour(new <emphasis role="bold">Display("three")</emphasis>) + .transition().to("automatic 4") + .<emphasis role="bold">node("automatic 4")</emphasis>.behaviour(new <emphasis role="bold">Display("four")</emphasis>) + .transition().to("wait 3") + .<emphasis role="bold">node("wait 3")</emphasis>.behaviour(new <emphasis role="bold">WaitState</emphasis>()) +.done();</programlisting> + <para>Let's walk you through one execution of this process. + </para> + <programlisting>Execution execution = processDefinition.startExecution();</programlisting> + <para>Starting a new execution means that the initial node is executed. So if an automatic + activity would be configured as the behaviour in the initial node, the process will start executing + immediatly in the startExecution. In this case however, the initial node is a wait state. So + the startExecution method returns immediately and the execution will be positioned in the initial + node 'wait 1'. + </para> + <figure id="automatic.wait1"> + <title>A new execution will be positioned in 'wait 1'.</title> + <mediaobject><imageobject><imagedata align="center" fileref="images/ch04.automatic.wait1.png"/></imageobject></mediaobject> + </figure> + <para>Then an external trigger is given with the signal method.</para> + <programlisting>execution.signal();</programlisting> + <para>As explained above when introducing the WaitState, that signal will cause the default transition to be taken. The + transition will move the execution to node <literal>automatic 1</literal> and execute it. The execute method + of the <literal>Display</literal> activity in <literal>automatic 1</literal> print a line to the console and it + will <emphasis role="bold">not</emphasis> call <literal>execution.waitForSignal()</literal>. Therefore, the execution + will proceed by taking the default transition out of <literal>automatic 1</literal>. The signal method is still blocking + cause this action and the transitions are taken + by that same thread. Then the execution arrives in <literal>wait 2</literal> and executes the <literal>WaitState</literal> + activity. That method will invoke the <literal>execution.waitForSignal()</literal>, which will cause the signal method + to return. That is when the thread is given back to the client that invoked the signal method. + </para> + <para>So when the signal method returns, the execution is positioned in <literal>wait 2</literal>.</para> + <figure id="automatic.wait2"> + <title>One signal brought the execution from 'initial' to 'wait 2'.</title> + <mediaobject><imageobject><imagedata align="center" fileref="images/ch04.automatic.wait2.png"/></imageobject></mediaobject> + </figure> + <para>Then the execution is now waiting for an external trigger just as an object (more precisely an object graph) in + memory until the next external trigger is given with the signal method. + </para> + <programlisting>execution.signal();</programlisting> + <para>This second invocation of signal will take the execution similarly all the way to <literal>wait 3</literal> before + it returns. + </para> + <figure id="automatic.wait3"> + <title>The second signal brought the execution all the way to 'wait 3'.</title> + <mediaobject><imageobject><imagedata align="center" fileref="images/ch04.automatic.wait3.png"/></imageobject></mediaobject> + </figure> + <para>To make executable processes, developers need to know exactly what the automatic activities, + what the wait states are and which threads will be allocated to the process execution. For business + analysts that draw the analysis process, things are a bit simpler. For the + activities they draw, they usually know whether it's a human or a system that is responsible. + But they typically don't not how this translates to threads and transactions. + </para> + <para>So for the developer, the first job is to analyse what needs to be executed + within the thread of control of the process and what is outside. Looking for the external + triggers can be a good start to find the wait states in a process, just like verbs and nouns + can be the rule of thumb in building UML class diagrams. + </para> + </section> + + <!-- ### PROCESS CONCURRENCY ############################################ --> + <section> + <title>Process concurrency</title> + <para>To model process concurrency, there is a parent-child tree structure on the + execution. The idea is that the main path of execution is the root of that tree. + This implies that on the level of the Process Virtual Machine, there is no differentiation + between complete process instances and paths of execution within a process instance. + One of the main motivations for this design is that the API actually is not made more + complex then necessary for simple processes with only one single path of execution. + </para> + <figure id="execution.structure"> + <title>UML class diagram of the basic execution structure</title> + <mediaobject><imageobject><imagedata align="center" fileref="images/ch04.execution.structure.classes.png"/></imageobject></mediaobject> + </figure> + <para>To extablish multiple concurrent paths of execution, child executions can be + created. Only leaf executions can be active. Non-leave executions should be + inactive. This tree structure of executions doesn't enforce a particular type of + concurrency or join behaviour. It's up to the forks or and-splits and to the joins + or and-merges to use the execution tree structure in any way they want to define + the wanted concurrency behaviour. Here you see an example + of concurrent executions. + </para> + <figure id="concurrency"> + <title>Concurrent paths of execution</title> + <mediaobject><imageobject><imagedata align="center" fileref="images/ch04.concurrency.png"/></imageobject></mediaobject> + </figure> + <para>There is a billing and a shipping path of execution. In this case, the + flat bar nodes represent nodes that fork and join. The execution shows a three + executions. The main path of execution is inactive (represented as gray) and the + billing and shipping paths of execution are active and point to the node + <literal>bill</literal> and <literal>ship</literal> respectively. + </para> + <para>It's up to the node behaviour implementations how they want to use this + execution structure. Suppose that multiple tasks have to be completed before the + execution is to proceed. The node behaviour can spawn a series of child executions + for this. Or alternatively, the task component could support task groups that + are associated to one single execution. In that case, the task component becomes + responsible for synchronizing the tasks, thereby moving this responsibility + outside the scope of the execution tree structure. + </para> + </section> + + <!-- ### EXCEPTION HANDLERS ############################################# --> + <section> + <title>Exception handlers</title> + <para>In all the <link linkend="delegationclasses">code that is associated to a process</link> + like Activity's, Actions and Conditions, it's possible to include try-catch blocks in + the method implementations to handle exceptions. But in order to build more reusable building + blocks for both the delegation classes and the exception handling logic, exception handlers are + added to the core process model. + </para> + <para>An exception handler can be associated to any process element. When an exception + occurs in a delegation class, a matching exception handler will be searched for. If + such an exception handler is found, it will get a chance to handle the exception. + </para> + <para>If an exception handler completes without problems, then the exception is considered + handled and the execution resumes right after the delegation code that was called. For example, + a transition has three actions and the second action throws an exception that is handled + by an exception handler, then + </para> + <para>Writing automatic activities that are exception handler aware is easy. The + default is to proceed anyway. No method needs to be called on the execution. So + if an automatic activity throws an exception that is handled by an exception handler, + the execution will just proceed after that activity. It becomes a big more difficult + for control flow activities. They might have to include try-finally blocks to + invoke the proper methods on the execution before an exception handler gets a + chance to handle the exception. For example, if an activity is a wait state and + an exception occurs, then there is a risk that the thread jumps over the + invocation of <literal>execution.waitForSignal()</literal>, causing the execution + to proceed after the activity. + </para> + <para>TODO: exceptionhandler.isRethrowMasked</para> + <para>TODO: transactional exception handlers</para> + <para>TODO: we never catch errors</para> + </section> + + <!-- ### PROCESS MODIFICATIONS ########################################## --> + <section> + <title>Process modifications</title> + <para>TODO: process modifications</para> + </section> + + <!-- ### LOCKING AND EXECUTION STATE #################################### --> + <section> + <title>Locking and execution state</title> + <para>The state of an execution is either active or locked. An active + execution is either executing or waiting for an external trigger. If an + execution is not in STATE_ACTIVE, then it is locked. A locked execution + is read only. + </para> + <para>When a new execution is created, it is in STATE_ACTIVE. To change + the state to a locked state, use lock(String). Some STATE_* constants + are provided that represent the most commonly used locked states. But + the state '...' in the picture indicates that any string can be provided + as the state in the lock method. + </para> + <figure id="execution.states"> + <title>States of an execution</title> + <mediaobject><imageobject><imagedata align="center" fileref="images/ch04.execution.states.png"/></imageobject></mediaobject> + </figure> + <para>If an execution is locked, methods that change the execution will + throw a PvmException and the message will reference the actual locking state. + Firing events, updating variables, updating priority and adding comments + are not considered to change an execution. Also creation and removal of child + executions are unchecked, which means that those methods can be invoked by + external API clients and node behaviour methods, even while the execution + is in a locked state. + </para> + <para>Make sure that comparisons between getState() and the STATE_* constants + are done with .equals and not with '==' because if executions are loaded + from persistent storage, a new string is created instead of the constants. + </para> + <para>An execution implementation will be locked: + </para> + <itemizedlist> + <listitem>When it is ended</listitem> + <listitem>When it is suspended</listitem> + <listitem>During asynchronous continuations</listitem> + </itemizedlist> + <para>Furthermore, locking can be used by Activity implementations to make + executions read only during wait states hen responsibility for the execution is + transferred to an external entity such as: + </para> + <itemizedlist> + <listitem>A human task</listitem> + <listitem>A service invocation</listitem> + <listitem>A wait state that ends when a scanner detects that a file appears</listitem> + </itemizedlist> + <para>In these situations the strategy is that the external entity should get + full control over the execution because it wants to control what is allowed + and what not. To get that control, they lock the execution so that all interactions + have to go through the external entity. + </para> + <para>One of the main reasons to create external entities is that they can live + on after the execution has already proceeded. For example, in case + of a service invocation, a timer could cause the execution to take the timeout transition. + When the response arrives after the timeout, the service invocation entity should + make sure it doesn't signal the execution. So the service invocation can be + seen as a node instance (aka activity instance) and is unique for every execution + of the node. + </para> + <para>External entities themselves are responsible for managing the execution + lock. If the timers and client applications are consequent in addressing the + external entities instead of the execution directly, then locking is in theory + unnecessary. It's up to the node behaviour implementations whether they want + to take the overhead of locking and unlocking. + </para> + </section> + +</chapter> Copied: jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/modules/ch07-Variables.xml (from rev 2276, jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/modules/chxx-Variables.xml) =================================================================== --- jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/modules/ch07-Variables.xml (rev 0) +++ jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/modules/ch07-Variables.xml 2008-09-18 13:55:29 UTC (rev 2279) @@ -0,0 +1,4 @@ +<chapter id="variables"> + <title>Variables</title> + <para>TODO</para> +</chapter> \ No newline at end of file Copied: jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/modules/ch08-Timers.xml (from rev 2276, jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/modules/chxx-Timers.xml) =================================================================== --- jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/modules/ch08-Timers.xml (rev 0) +++ jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/modules/ch08-Timers.xml 2008-09-18 13:55:29 UTC (rev 2279) @@ -0,0 +1,4 @@ +<chapter id="timers"> + <title>Timers</title> + <para>TODO</para> +</chapter> \ No newline at end of file Copied: jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/modules/ch09-AsynchronousContinuations.xml (from rev 2276, jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/modules/chxx-AsynchronousContinuations.xml) =================================================================== --- jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/modules/ch09-AsynchronousContinuations.xml (rev 0) +++ jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/modules/ch09-AsynchronousContinuations.xml 2008-09-18 13:55:29 UTC (rev 2279) @@ -0,0 +1,4 @@ +<chapter id="asynchronouscontinuations"> + <title>Asynchronous continuations</title> + <para>TODO</para> +</chapter> Copied: jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/modules/ch10-SoftwareLogging.xml (from rev 2276, jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/modules/chxx-SoftwareLogging.xml) =================================================================== --- jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/modules/ch10-SoftwareLogging.xml (rev 0) +++ jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/modules/ch10-SoftwareLogging.xml 2008-09-18 13:55:29 UTC (rev 2279) @@ -0,0 +1,99 @@ +<chapter id="softwarelogging"> + <title>Software logging</title> + + <section> + <title>Configuration</title> + <para>PVM can use JDK logging (java.util.logging) or log4j. When the first message is + logged, PVM logging will make the selection with following procedure: + <orderedlist> + <listitem>If a <literal>logging.properties</literal> resource is found + on the classpath (using the context classloader), then JDK logging will + be used and that file will be used to initialize the JDK logging. + </listitem> + <listitem>If log4j is found on the classpath, then log4j will be used. + The check for log4j will be done by checking availability of class + <literal>org.apache.log4j.LogManager</literal> with the context classloader. + </listitem> + <listitem>If none of the above, JDK logging will be used.</listitem> + </orderedlist> + </para> + </section> + + <section> + <title>Categories</title> + <para>The PVM classes use their class name as the category for the logger. + </para> + <para>To have a basic understanding of what the PVM classes are doing, + turning on the <literal>debug</literal> level is great. Level + <literal>trace</literal> might be spitting out too much for that + purpose. + </para> + </section> + + <section> + <title>JDK logging</title> + <para>In JDK logging, <literal>debug</literal>maps to <literal>fine</literal> + and <literal>trace</literal> maps to <literal>finest</literal>. + Level <literal>finer</literal> is not used. + </para> + <para><literal>org.jbpm.pvm.internal.log.LogFormatter</literal> is part of + the pvm library and it can create a nice one-line output for log messages. + It also has a neat feature that creates a unique indentation per thread. + To configure it, this is a typical <literal>logging.properties</literal> + </para> + <programlisting>handlers = java.util.logging.ConsoleHandler +java.util.logging.ConsoleHandler.level = FINEST +java.util.logging.ConsoleHandler.formatter = org.jbpm.pvm.internal.log.LogFormatter + +# For example, set the com.xyz.foo logger to only log SEVERE messages: +# com.xyz.foo.level = SEVERE + +.level = SEVERE +org.jbpm.level=FINE +org.jbpm.tx.level=FINE +org.jbpm.pvm.internal.wire.level=FINE</programlisting> + +<!-- + <para>For production usage, jBPM also includes an error triggered log handler. This is + a log handler that will only keep the most recent log messages in + memory and these will only be flushed to a file in case an error occurs. + </para> + <para>to configure it, add <literal>org.jbpm.util.ErrorTriggeredFileHandler</literal> + to the handlers in the logging properties like this: + </para> + <programlisting>handlers = java.util.logging.ConsoleHandler org.jbpm.util.ErrorTriggeredFileHandler</programlisting> + <para>Next snippet shows how in the same logging.properties, the error + triggered file handler can be configured. The given values are the default + values. + </para> + <programlisting>org.jbpm.util.ErrorTriggeredFileHandler.size = 500 +org.jbpm.util.ErrorTriggeredFileHandler.push = SEVERE +org.jbpm.util.ErrorTriggeredFileHandler.pattern = %h/jbpm%u.log</programlisting> + <para>Alternatively to using the org.jbpm.util.ErrorTriggeredFileHandler, the + JDK handlers FileHandler and MemoryHandler can used in combination to get + similar results with a bit more configuration. + </para> + +--> + </section> + + <section> + <title>Debugging persistence</title> + <para>When testing the persistence, following logging configurations can be + valuable. Category <literal>org.hibernate.SQL</literal> shows the SQL statement that is executed + and category <literal>org.hibernate.type</literal> shows the values of the parameters that are + set in the queries. + </para> + <programlisting>org.hibernate.SQL.level=FINEST +org.hibernate.type.level=FINEST</programlisting> + <para>And in case you get a failed batch as a cause in a hibernate exception, + you might want to set the batch size to 0 like this in the hibernate properties: + </para> + <programlisting>hibernate.jdbc.batch_size = 0</programlisting> + <para>Also in the hibernate properties, the following properties allow for + detailed logs of the SQL that hibernate spits out:</para> + <programlisting>hibernate.show_sql = true +hibernate.format_sql = true +hibernate.use_sql_comments = true</programlisting> + </section> +</chapter> \ No newline at end of file Deleted: jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/modules/chxx-AdvancedGraphExecution.xml =================================================================== --- jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/modules/chxx-AdvancedGraphExecution.xml 2008-09-18 07:25:16 UTC (rev 2278) +++ jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/modules/chxx-AdvancedGraphExecution.xml 2008-09-18 13:55:29 UTC (rev 2279) @@ -1,324 +0,0 @@ -<chapter id="advancedgraphexecution"> - <title>Advanced graph execution</title> - - <!-- ### LOOPS ########################################################## --> - <section> - <title>Loops</title> - <para>Loops can be based on transitions or on node composition. Loops can contain wait - states. - </para> - <para>To support high numbers of automatic loop executions, the Process Virtual Machine - tranformed the propagation of execution from tail recursion to a while loop. This means - that all the methods in the <literal>Execution</literal> class that propagate the - execution like <literal>take</literal> or <literal>execute</literal> will not be - executed when you call them. Instead, the method invocations will be appended to a list. - The first invocation of such a method will start a loop that will execute all invocations - till that list is empty. These invocations are called atomic operations. - </para> - </section> - - <!-- ### SUB PROCESSES ################################################## --> - <section> - <title>Sub processes</title> - <para>TODO: sub processes</para> - </section> - - <!-- ### DEFAULT PROCEED BEHAVIOUR ###################################### --> - <section id="defaultproceedbehaviour"> - <title>Default proceed behaviour</title> - <para>When an <literal>Activity</literal> is used as node behaviour, it can - explicitely propagate the execution with following methods: - </para> - <itemizedlist> - <listitem><literal>waitForSignal()</literal></listitem> - <listitem><literal>take(Transition)</literal></listitem> - <listitem><literal>end(*)</literal></listitem> - <listitem><literal>execute(Node)</literal></listitem> - <listitem><literal>createExecution(*)</literal></listitem> - </itemizedlist> - <para>When <literal>Activity</literal> implementations used for node behviour - don't call any of the following execution propagation methods, then, after - the activity is executed, the execution will just proceed. - </para> - <para>By default proceeding will perform the first action that applies - in the following list:</para> - <itemizedlist> - <listitem>If the current node has a default outgoing transition, take it.</listitem> - <listitem>If the current node has a parent node, move back to the parent node.</listitem> - <listitem>Otherwise, end this execution.</listitem> - </itemizedlist> - <para>Process languages can overwrite the default proceed behaviour - by overriding the <literal>proceed</literal> method in - <literal>ExecutionImpl</literal>. - </para> - </section> - - <!-- ### FUNCTIONAL ACTIVITIES ################################ --> - <section id="functionalactivities"> - <title>Functional activities</title> - <para>Activities that also can be used as event listeners are called functional - activities. Examples of automatic activities are sending an email, doing a database - update, generating a pdf, calculating an average, etc. All of these are automatic - activities that do not change the execution flow. Here's how such activities can - be implemented: - </para> - <programlisting>public class FunctionalActivity implements Activity, EventListener { - public void execute(ActivityExecution execution) { - perform(execution); - } - public void notify(EventListenerExecution execution) { - perform(execution); - } - void perform(OpenExecution execution) { - ...do functional work... - } - }</programlisting> - <para>None of these methods actually invoke execution propagation methods. - So after the perform method is completed, the execution will - <link linkend="defaultproceedbehaviour">proceed in the default way</link>. - </para> - </section> - - - <!-- ### EXECUTION AND THREADS ########################################## --> - <section id="executionandthreads"> - <title>Execution and threads</title> - <para>This section explains how the Process Virtual Machine boroughs the thread - from the client to bring an execution from one wait state to another. - </para> - <para>When a client invokes a method (like e.g. the signal method) on an execution, - by default, the Process Virtual Machine will use that thread to progress the execution - until it reached a wait state. Once the next wait state has been reached, the - method returns and the client gets the thread back. This is the default way - for the Process Virtual Machine to operate. Two more levels of asynchonous - execution complement this default behaviour: - <link linkend="asynchronouscontinuations">Asynchronous continuations</link> - and the <link linkend="architecture">asynchronous command service</link>. - </para> - <para>The next process will show the basics concretely. It has three wait states - and four automatic nodes. - </para> - <figure id="automatic"> - <title>Process with many sequential automatic activities.</title> - <mediaobject><imageobject><imagedata align="center" fileref="images/ch04.automatic.png"/></imageobject></mediaobject> - </figure> - <para>Here's how to build the process:</para> - <programlisting>ProcessDefinition processDefinition = ProcessFactory.build("automatic") - .<emphasis role="bold">node("wait 1").initial()</emphasis>.behaviour(new <emphasis role="bold">WaitState</emphasis>()) - .transition().to("automatic 1") - .<emphasis role="bold">node("automatic 1")</emphasis>.behaviour(new <emphasis role="bold">Display("one")</emphasis>) - .transition().to("wait 2") - .<emphasis role="bold">node("wait 2")</emphasis>.behaviour(new <emphasis role="bold">WaitState</emphasis>()) - .transition().to("automatic 2") - .<emphasis role="bold">node("automatic 2")</emphasis>.behaviour(new <emphasis role="bold">Display("two")</emphasis>) - .transition().to("automatic 3") - .<emphasis role="bold">node("automatic 3")</emphasis>.behaviour(new <emphasis role="bold">Display("three")</emphasis>) - .transition().to("automatic 4") - .<emphasis role="bold">node("automatic 4")</emphasis>.behaviour(new <emphasis role="bold">Display("four")</emphasis>) - .transition().to("wait 3") - .<emphasis role="bold">node("wait 3")</emphasis>.behaviour(new <emphasis role="bold">WaitState</emphasis>()) -.done();</programlisting> - <para>Let's walk you through one execution of this process. - </para> - <programlisting>Execution execution = processDefinition.startExecution();</programlisting> - <para>Starting a new execution means that the initial node is executed. So if an automatic - activity would be configured as the behaviour in the initial node, the process will start executing - immediatly in the startExecution. In this case however, the initial node is a wait state. So - the startExecution method returns immediately and the execution will be positioned in the initial - node 'wait 1'. - </para> - <figure id="automatic.wait1"> - <title>A new execution will be positioned in 'wait 1'.</title> - <mediaobject><imageobject><imagedata align="center" fileref="images/ch04.automatic.wait1.png"/></imageobject></mediaobject> - </figure> - <para>Then an external trigger is given with the signal method.</para> - <programlisting>execution.signal();</programlisting> - <para>As explained above when introducing the WaitState, that signal will cause the default transition to be taken. The - transition will move the execution to node <literal>automatic 1</literal> and execute it. The execute method - of the <literal>Display</literal> activity in <literal>automatic 1</literal> print a line to the console and it - will <emphasis role="bold">not</emphasis> call <literal>execution.waitForSignal()</literal>. Therefore, the execution - will proceed by taking the default transition out of <literal>automatic 1</literal>. The signal method is still blocking - cause this action and the transitions are taken - by that same thread. Then the execution arrives in <literal>wait 2</literal> and executes the <literal>WaitState</literal> - activity. That method will invoke the <literal>execution.waitForSignal()</literal>, which will cause the signal method - to return. That is when the thread is given back to the client that invoked the signal method. - </para> - <para>So when the signal method returns, the execution is positioned in <literal>wait 2</literal>.</para> - <figure id="automatic.wait2"> - <title>One signal brought the execution from 'initial' to 'wait 2'.</title> - <mediaobject><imageobject><imagedata align="center" fileref="images/ch04.automatic.wait2.png"/></imageobject></mediaobject> - </figure> - <para>Then the execution is now waiting for an external trigger just as an object (more precisely an object graph) in - memory until the next external trigger is given with the signal method. - </para> - <programlisting>execution.signal();</programlisting> - <para>This second invocation of signal will take the execution similarly all the way to <literal>wait 3</literal> before - it returns. - </para> - <figure id="automatic.wait3"> - <title>The second signal brought the execution all the way to 'wait 3'.</title> - <mediaobject><imageobject><imagedata align="center" fileref="images/ch04.automatic.wait3.png"/></imageobject></mediaobject> - </figure> - <para>To make executable processes, developers need to know exactly what the automatic activities, - what the wait states are and which threads will be allocated to the process execution. For business - analysts that draw the analysis process, things are a bit simpler. For the - activities they draw, they usually know whether it's a human or a system that is responsible. - But they typically don't not how this translates to threads and transactions. - </para> - <para>So for the developer, the first job is to analyse what needs to be executed - within the thread of control of the process and what is outside. Looking for the external - triggers can be a good start to find the wait states in a process, just like verbs and nouns - can be the rule of thumb in building UML class diagrams. - </para> - </section> - - <!-- ### PROCESS CONCURRENCY ############################################ --> - <section> - <title>Process concurrency</title> - <para>To model process concurrency, there is a parent-child tree structure on the - execution. The idea is that the main path of execution is the root of that tree. - This implies that on the level of the Process Virtual Machine, there is no differentiation - between complete process instances and paths of execution within a process instance. - One of the main motivations for this design is that the API actually is not made more - complex then necessary for simple processes with only one single path of execution. - </para> - <figure id="execution.structure"> - <title>UML class diagram of the basic execution structure</title> - <mediaobject><imageobject><imagedata align="center" fileref="images/ch04.execution.structure.classes.png"/></imageobject></mediaobject> - </figure> - <para>To extablish multiple concurrent paths of execution, child executions can be - created. Only leaf executions can be active. Non-leave executions should be - inactive. This tree structure of executions doesn't enforce a particular type of - concurrency or join behaviour. It's up to the forks or and-splits and to the joins - or and-merges to use the execution tree structure in any way they want to define - the wanted concurrency behaviour. Here you see an example - of concurrent executions. - </para> - <figure id="concurrency"> - <title>Concurrent paths of execution</title> - <mediaobject><imageobject><imagedata align="center" fileref="images/ch04.concurrency.png"/></imageobject></mediaobject> - </figure> - <para>There is a billing and a shipping path of execution. In this case, the - flat bar nodes represent nodes that fork and join. The execution shows a three - executions. The main path of execution is inactive (represented as gray) and the - billing and shipping paths of execution are active and point to the node - <literal>bill</literal> and <literal>ship</literal> respectively. - </para> - <para>It's up to the node behaviour implementations how they want to use this - execution structure. Suppose that multiple tasks have to be completed before the - execution is to proceed. The node behaviour can spawn a series of child executions - for this. Or alternatively, the task component could support task groups that - are associated to one single execution. In that case, the task component becomes - responsible for synchronizing the tasks, thereby moving this responsibility - outside the scope of the execution tree structure. - </para> - </section> - - <!-- ### EXCEPTION HANDLERS ############################################# --> - <section> - <title>Exception handlers</title> - <para>In all the <link linkend="delegationclasses">code that is associated to a process</link> - like Activity's, Actions and Conditions, it's possible to include try-catch blocks in - the method implementations to handle exceptions. But in order to build more reusable building - blocks for both the delegation classes and the exception handling logic, exception handlers are - added to the core process model. - </para> - <para>An exception handler can be associated to any process element. When an exception - occurs in a delegation class, a matching exception handler will be searched for. If - such an exception handler is found, it will get a chance to handle the exception. - </para> - <para>If an exception handler completes without problems, then the exception is considered - handled and the execution resumes right after the delegation code that was called. For example, - a transition has three actions and the second action throws an exception that is handled - by an exception handler, then - </para> - <para>Writing automatic activities that are exception handler aware is easy. The - default is to proceed anyway. No method needs to be called on the execution. So - if an automatic activity throws an exception that is handled by an exception handler, - the execution will just proceed after that activity. It becomes a big more difficult - for control flow activities. They might have to include try-finally blocks to - invoke the proper methods on the execution before an exception handler gets a - chance to handle the exception. For example, if an activity is a wait state and - an exception occurs, then there is a risk that the thread jumps over the - invocation of <literal>execution.waitForSignal()</literal>, causing the execution - to proceed after the activity. - </para> - <para>TODO: exceptionhandler.isRethrowMasked</para> - <para>TODO: transactional exception handlers</para> - <para>TODO: we never catch errors</para> - </section> - - <!-- ### PROCESS MODIFICATIONS ########################################## --> - <section> - <title>Process modifications</title> - <para>TODO: process modifications</para> - </section> - - <!-- ### LOCKING AND EXECUTION STATE #################################### --> - <section> - <title>Locking and execution state</title> - <para>The state of an execution is either active or locked. An active - execution is either executing or waiting for an external trigger. If an - execution is not in STATE_ACTIVE, then it is locked. A locked execution - is read only. - </para> - <para>When a new execution is created, it is in STATE_ACTIVE. To change - the state to a locked state, use lock(String). Some STATE_* constants - are provided that represent the most commonly used locked states. But - the state '...' in the picture indicates that any string can be provided - as the state in the lock method. - </para> - <figure id="execution.states"> - <title>States of an execution</title> - <mediaobject><imageobject><imagedata align="center" fileref="images/ch04.execution.states.png"/></imageobject></mediaobject> - </figure> - <para>If an execution is locked, methods that change the execution will - throw a PvmException and the message will reference the actual locking state. - Firing events, updating variables, updating priority and adding comments - are not considered to change an execution. Also creation and removal of child - executions are unchecked, which means that those methods can be invoked by - external API clients and node behaviour methods, even while the execution - is in a locked state. - </para> - <para>Make sure that comparisons between getState() and the STATE_* constants - are done with .equals and not with '==' because if executions are loaded - from persistent storage, a new string is created instead of the constants. - </para> - <para>An execution implementation will be locked: - </para> - <itemizedlist> - <listitem>When it is ended</listitem> - <listitem>When it is suspended</listitem> - <listitem>During asynchronous continuations</listitem> - </itemizedlist> - <para>Furthermore, locking can be used by Activity implementations to make - executions read only during wait states hen responsibility for the execution is - transferred to an external entity such as: - </para> - <itemizedlist> - <listitem>A human task</listitem> - <listitem>A service invocation</listitem> - <listitem>A wait state that ends when a scanner detects that a file appears</listitem> - </itemizedlist> - <para>In these situations the strategy is that the external entity should get - full control over the execution because it wants to control what is allowed - and what not. To get that control, they lock the execution so that all interactions - have to go through the external entity. - </para> - <para>One of the main reasons to create external entities is that they can live - on after the execution has already proceeded. For example, in case - of a service invocation, a timer could cause the execution to take the timeout transition. - When the response arrives after the timeout, the service invocation entity should - make sure it doesn't signal the execution. So the service invocation can be - seen as a node instance (aka activity instance) and is unique for every execution - of the node. - </para> - <para>External entities themselves are responsible for managing the execution - lock. If the timers and client applications are consequent in addressing the - external entities instead of the execution directly, then locking is in theory - unnecessary. It's up to the node behaviour implementations whether they want - to take the overhead of locking and unlocking. - </para> - </section> - -</chapter> Deleted: jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/modules/chxx-AsynchronousContinuations.xml =================================================================== --- jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/modules/chxx-AsynchronousContinuations.xml 2008-09-18 07:25:16 UTC (rev 2278) +++ jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/modules/chxx-AsynchronousContinuations.xml 2008-09-18 13:55:29 UTC (rev 2279) @@ -1,3 +0,0 @@ -<chapter id="asynchronouscontinuations"> - <title>Asynchronous continuations</title> -</chapter> Deleted: jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/modules/chxx-SoftwareLogging.xml =================================================================== --- jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/modules/chxx-SoftwareLogging.xml 2008-09-18 07:25:16 UTC (rev 2278) +++ jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/modules/chxx-SoftwareLogging.xml 2008-09-18 13:55:29 UTC (rev 2279) @@ -1,99 +0,0 @@ -<chapter id="softwarelogging"> - <title>Software logging</title> - - <section> - <title>Configuration</title> - <para>PVM can use JDK logging (java.util.logging) or log4j. When the first message is - logged, PVM logging will make the selection with following procedure: - <orderedlist> - <listitem>If a <literal>logging.properties</literal> resource is found - on the classpath (using the context classloader), then JDK logging will - be used and that file will be used to initialize the JDK logging. - </listitem> - <listitem>If log4j is found on the classpath, then log4j will be used. - The check for log4j will be done by checking availability of class - <literal>org.apache.log4j.LogManager</literal> with the context classloader. - </listitem> - <listitem>If none of the above, JDK logging will be used.</listitem> - </orderedlist> - </para> - </section> - - <section> - <title>Categories</title> - <para>The PVM classes use their class name as the category for the logger. - </para> - <para>To have a basic understanding of what the PVM classes are doing, - turning on the <literal>debug</literal> level is great. Level - <literal>trace</literal> might be spitting out too much for that - purpose. - </para> - </section> - - <section> - <title>JDK logging</title> - <para>In JDK logging, <literal>debug</literal>maps to <literal>fine</literal> - and <literal>trace</literal> maps to <literal>finest</literal>. - Level <literal>finer</literal> is not used. - </para> - <para><literal>org.jbpm.pvm.internal.log.LogFormatter</literal> is part of - the pvm library and it can create a nice one-line output for log messages. - It also has a neat feature that creates a unique indentation per thread. - To configure it, this is a typical <literal>logging.properties</literal> - </para> - <programlisting>handlers = java.util.logging.ConsoleHandler -java.util.logging.ConsoleHandler.level = FINEST -java.util.logging.ConsoleHandler.formatter = org.jbpm.pvm.internal.log.LogFormatter - -# For example, set the com.xyz.foo logger to only log SEVERE messages: -# com.xyz.foo.level = SEVERE - -.level = SEVERE -org.jbpm.level=FINE -org.jbpm.tx.level=FINE -org.jbpm.pvm.internal.wire.level=FINE</programlisting> - -<!-- - <para>For production usage, jBPM also includes an error triggered log handler. This is - a log handler that will only keep the most recent log messages in - memory and these will only be flushed to a file in case an error occurs. - </para> - <para>to configure it, add <literal>org.jbpm.util.ErrorTriggeredFileHandler</literal> - to the handlers in the logging properties like this: - </para> - <programlisting>handlers = java.util.logging.ConsoleHandler org.jbpm.util.ErrorTriggeredFileHandler</programlisting> - <para>Next snippet shows how in the same logging.properties, the error - triggered file handler can be configured. The given values are the default - values. - </para> - <programlisting>org.jbpm.util.ErrorTriggeredFileHandler.size = 500 -org.jbpm.util.ErrorTriggeredFileHandler.push = SEVERE -org.jbpm.util.ErrorTriggeredFileHandler.pattern = %h/jbpm%u.log</programlisting> - <para>Alternatively to using the org.jbpm.util.ErrorTriggeredFileHandler, the - JDK handlers FileHandler and MemoryHandler can used in combination to get - similar results with a bit more configuration. - </para> - ---> - </section> - - <section> - <title>Debugging persistence</title> - <para>When testing the persistence, following logging configurations can be - valuable. Category <literal>org.hibernate.SQL</literal> shows the SQL statement that is executed - and category <literal>org.hibernate.type</literal> shows the values of the parameters that are - set in the queries. - </para> - <programlisting>org.hibernate.SQL.level=FINEST -org.hibernate.type.level=FINEST</programlisting> - <para>And in case you get a failed batch as a cause in a hibernate exception, - you might want to set the batch size to 0 like this in the hibernate properties: - </para> - <programlisting>hibernate.jdbc.batch_size = 0</programlisting> - <para>Also in the hibernate properties, the following properties allow for - detailed logs of the SQL that hibernate spits out:</para> - <programlisting>hibernate.show_sql = true -hibernate.format_sql = true -hibernate.use_sql_comments = true</programlisting> - </section> -</chapter> \ No newline at end of file Deleted: jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/modules/chxx-Timers.xml =================================================================== --- jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/modules/chxx-Timers.xml 2008-09-18 07:25:16 UTC (rev 2278) +++ jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/modules/chxx-Timers.xml 2008-09-18 13:55:29 UTC (rev 2279) @@ -1,3 +0,0 @@ -<chapter id="timers"> - <title>Timers</title> -</chapter> \ No newline at end of file Deleted: jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/modules/chxx-Variables.xml =================================================================== --- jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/modules/chxx-Variables.xml 2008-09-18 07:25:16 UTC (rev 2278) +++ jbpm4/pvm/trunk/modules/manual/src/main/docbook/en/modules/chxx-Variables.xml 2008-09-18 13:55:29 UTC (rev 2279) @@ -1,3 +0,0 @@ -<chapter id="variables"> - <title>Variables</title> -</chapter> \ No newline at end of file