org.systemsbiology.gaggle.core

Class GooseWorkflowManager

java.lang.Object

org.systemsbiology.gaggle.core.GooseWorkflowManager

public class GooseWorkflowManager
extends java.lang.Object

The GooseWorkflowManager class is a helper class for Gaggle Geese to carry out workflow related actions. It wraps the most common actions so that Goose does not have to implement them individually.

Here is how it works. When a goose receives a WorkflowAction from Boss, it should call addSession(org.systemsbiology.gaggle.core.datatypes.WorkflowAction), which stores the request in the processing queue. addSession generates an ID for each workflowAction received. The ID should be passed to getWorkflowAction(String) to retrieve the workflowAction stored in the processing queue.

Some goose (such as Firegoose) employes a polling mechanism and thus needs to call getCurrentRequest() to retrieve a workflowAction.

User can interact with the goose and generate data to be sent to the next component in the workflow. Once the data is generated, the goose should call addSessionTargetData(String, int, org.systemsbiology.gaggle.core.datatypes.GaggleData) to add the data for the component. Note there could be multiple components in the next step. Therefore, the goose needs to call addSessionTargetData multiple times until each component is associated with data. Finally, the goose should call CompleteWorkflowAction(Boss, String) to complete the processing of the workflowAction. CompleteWorkflowAction passes the data to to the Boss, which triggers the subsequent geese of the workflow to process the data.

Constructor Summary

Constructors

Constructor and Description
GooseWorkflowManager()

Method Summary

Methods

Modifier and Type	Method and Description
java.lang.String	addSession(WorkflowAction request) When a Goose receives a workflowAction from the Boss, it should call this function to save the workflowAction in the processing queue.
void	addSessionTargetData(java.lang.String requestID, int targetIndex, org.systemsbiology.gaggle.core.datatypes.GaggleData data) Attach data to a goose in the next step of a workflow.
boolean	CompleteWorkflowAction(org.systemsbiology.gaggle.core.Boss boss, java.lang.String requestID) All the data are ready, we submit the response to the boss.
int	dataSubmittedForSession(java.lang.String requestID) Retrieve the number of data elements that have been added for the geese in the next step of the workflow.
void	downloadFileFromUrl(java.lang.String filename, java.lang.String urlString)
boolean	finalizeSessionAction(java.lang.String requestID)
java.lang.String	getCurrentRequest() Retrieve a workflowAction that has not been processed yet.
org.systemsbiology.gaggle.core.datatypes.GaggleGooseInfo	getGooseInfo()
java.lang.String[]	getNameList(java.lang.String requestID)
WorkflowAction	getSessionResponse(java.lang.String requestID)
java.lang.String	getSize(java.lang.String requestID)
java.lang.String	getSpecies(java.lang.String requestID)
java.lang.String	getSubAction(java.lang.String requestID)
java.lang.ThreadGroup	getThreadGroup(java.lang.String name) Starting from Java 7 update 25, the Java Runtime is using a different threadgroup (i.e., javawsApplicationThreadGroup) to run web starts.
java.lang.String	getType(java.lang.String requestID)
WorkflowAction	getWorkflowAction(java.lang.String requestID) Get the workflowAction given the requestID
java.lang.Object[]	getWorkflowActionData(java.lang.String requestID)
void	invokeLater2(java.lang.ThreadGroup javawsApplicationThreadGroup, sun.awt.AppContext appContext, java.lang.Runnable rn) Starting from Java 7 update 25, the Java Runtime is using a different threadgroup (i.e., javawsApplicationThreadGroup) to run web starts.
org.systemsbiology.gaggle.core.datatypes.GaggleData	ProcessTextFile(java.lang.String fileurl) First download a text file from the server, then parse it and output a GaggleData object.
void	removeRequest(java.lang.String requestID) Remove a workflowAction from the processing queue.
void	RemoveSessionData(java.lang.String requestID) Remove the workflowAction after finishing handling it.

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

GooseWorkflowManager

public GooseWorkflowManager()

Method Detail

getWorkflowActionData

public java.lang.Object[] getWorkflowActionData(java.lang.String requestID)

getSpecies

public java.lang.String getSpecies(java.lang.String requestID)

getNameList

public java.lang.String[] getNameList(java.lang.String requestID)

getSize

public java.lang.String getSize(java.lang.String requestID)

getType

public java.lang.String getType(java.lang.String requestID)

getSubAction

public java.lang.String getSubAction(java.lang.String requestID)

getGooseInfo

public org.systemsbiology.gaggle.core.datatypes.GaggleGooseInfo getGooseInfo()

addSession

public java.lang.String addSession(WorkflowAction request)

When a Goose receives a workflowAction from the Boss, it should call this function to save the workflowAction in the processing queue.

Parameters:

request - workflowAction received from the Boss

Returns:

the requestID generated by the function

getWorkflowAction

public WorkflowAction getWorkflowAction(java.lang.String requestID)

Get the workflowAction given the requestID

Parameters:

requestID - the ID genmerated by addSession.

Returns:

the workflowAction corresponding to the requestID

getCurrentRequest

public java.lang.String getCurrentRequest()

Retrieve a workflowAction that has not been processed yet.

Returns:

the requestID of the workflowAction

removeRequest

public void removeRequest(java.lang.String requestID)

Remove a workflowAction from the processing queue.

Parameters:

requestID - the ID generated by addSession

addSessionTargetData

public void addSessionTargetData(java.lang.String requestID,
                        int targetIndex,
                        org.systemsbiology.gaggle.core.datatypes.GaggleData data)

Attach data to a goose in the next step of a workflow.

Parameters:

requestID - requestID generated by addSession

targetIndex - the index of the target goose in the workflowAction

data - data to be associated with the target Goose

finalizeSessionAction

public boolean finalizeSessionAction(java.lang.String requestID)

dataSubmittedForSession

public int dataSubmittedForSession(java.lang.String requestID)

Retrieve the number of data elements that have been added for the geese in the next step of the workflow.

Parameters:

requestID - the ID generated by addSession

Returns:

the number of data elements that have been added for the geese in the next step of the workflow

getSessionResponse

public WorkflowAction getSessionResponse(java.lang.String requestID)

RemoveSessionData

public void RemoveSessionData(java.lang.String requestID)

Remove the workflowAction after finishing handling it.

Parameters:

requestID - the ID generated by addSession

CompleteWorkflowAction

public boolean CompleteWorkflowAction(org.systemsbiology.gaggle.core.Boss boss,
                             java.lang.String requestID)

All the data are ready, we submit the response to the boss.

Parameters:

boss - Boss that supports the Boss3 interface

requestID - requestID generated by addSession

Returns:

true if data is submitted successfully. false if error is encountered

downloadFileFromUrl

public void downloadFileFromUrl(java.lang.String filename,
                       java.lang.String urlString)
                         throws java.lang.Exception

Throws:

java.lang.Exception

ProcessTextFile

public org.systemsbiology.gaggle.core.datatypes.GaggleData ProcessTextFile(java.lang.String fileurl)

First download a text file from the server, then parse it and output a GaggleData object. For now we assume the first column of the text file contains the gene names.

Parameters:

fileurl - the url of the file to be parsed.

Returns:

getThreadGroup

public java.lang.ThreadGroup getThreadGroup(java.lang.String name)

Starting from Java 7 update 25, the Java Runtime is using a different threadgroup (i.e., javawsApplicationThreadGroup) to run web starts. This causes some applications such as Cytoscape and MeV to freeze on Mac OS. It turns out that we have to invoke the function calls on the javawsApplicationThreadGroup to avoid potential deadlocks on different thread groups. This API is provided to get the ThreadGroup given the name. User should pass "javawsApplicationThreadGroup" as the input parameter.

Parameters:

name -

Returns:

invokeLater2

public void invokeLater2(java.lang.ThreadGroup javawsApplicationThreadGroup,
                sun.awt.AppContext appContext,
                java.lang.Runnable rn)

Starting from Java 7 update 25, the Java Runtime is using a different threadgroup (i.e., javawsApplicationThreadGroup) to run web starts. This causes some applications such as Cytoscape and MeV to freeze on Mac OS. It turns out that we have to invoke the function calls on the javawsApplicationThreadGroup to avoid potential deadlocks on different thread groups. This API is a wrapper to start a function call on the javawsApplicationThreadGroup. User applications should call the getThreadGroup API first to get the javawsApplicationThreadGroup. Then create a Runnnable task to wrap the target function call, use the AppContext.getAppContext to get the current application context, and call this API.

Parameters:

javawsApplicationThreadGroup -

appContext -

rn -