James Wagner

Hello, basically I am setting up a jsp page that will take in the user's input files and the email address, then upload the files to another server (high performance cluster) for an analysis that can take on the order of days to run. When done several output files are created and I want these emailed to the user.

The best idea that I have come up with is to set up an hourly cron script on the intermediate server that will then take care of this (actually in turn calling a Java class that checks the status on the cluster and then obtains the results and emails them via the Java Mail API if the job is finished).
The mechanics of everything is working ok, when I just call the script directly it does properly call the Java class which does properly email the results if the job is finished on the cluster. Before I actually ask for permission from the administrator to set up this script on cron of the intermediate server I was just looking for any opinions as to whether this is an "ugly" solution to the problem and if there is any other cleaner way of accomplishing this, I don't really know where to even start looking for other approaches that will check the status on another server for a job, and then email the results if finished. Heck I can't even think of a good search term for this problem to check on Google for potential solutions.

Hi thanks for the link, though tweaking the manifest really did not seem to help with the problem at all. Ended up just figuring out the one jar file signed by another authority was non-essential and removed it from the jnlp file, not too sure what I could do if it were essential.

Hi, been trying to deploy an application on web start. It contains multiple third party jar files, some of which have been signed with a certificate that has since expired, once this problem was discovered I followed advice from Java forums and deleted DSA and RS files from the META-INF folder. To avoid conflicting problems with multiple jars, and since just in the testing stage I decided to just unpack all the jars and repack into a blob.jar, signing and verifying it with jarsigner, and listing this then as the only resource in the jnlp file. Presently just testing on localhost, with the jnlp and the blob.jar being deployed via Tomcat.


Any ideas what reason there is for this exception still being thrown? Thanks.


Here is the exception being thrown:

com.sun.deploy.net.JARSigningException: Could not verify signing in resource: http://localhost:8080/tmev/lib/blob.jar
at com.sun.javaws.security.SigningInfo.verifyAllEntriesSigned(Unknown Source)
at com.sun.javaws.security.SigningInfo.checkSigning(Unknown Source)
at com.sun.javaws.LaunchDownload.checkSignedResourcesHelper(Unknown Source)
at com.sun.javaws.LaunchDownload.checkSignedResources(Unknown Source)
at com.sun.javaws.Launcher.prepareLaunchFile(Unknown Source)
at com.sun.javaws.Launcher.prepareToLaunch(Unknown Source)
at com.sun.javaws.Launcher.launch(Unknown Source)
at com.sun.javaws.Main.launchApp(Unknown Source)
at com.sun.javaws.Main.continueInSecureThread(Unknown Source)
at com.sun.javaws.Main$1.run(Unknown Source)
at java.lang.Thread.run(Unknown Source)




Here is the jnlp file :

<?xml version="1.0" encoding="utf-8"?> <jnlp codebase="http://localhost:8080/tmev" href="tmev.jnlp"> <information> <title>MeV</title> <vendor>http://www.tm4.org/</vendor> <homepage href="http://www.tm4.org/"/> <offline-allowed/> </information> <security> <all-permissions/> </security> <resources> <j2se version="1.5+" max-heap-size="1024M"/> <jar href ="lib/blob.jar"/> </resources> <application-desc main-class="org.tigr.microarray.mev.TMEV" /> </jnlp>

Hi, after further playing around with it just a quick follow-up to maybe provide another clue as to problems with behaviour, if I keep the subset tree that i am adding to collapsed, I can add nodes to it to my heart's content, and then when I expand it I can see everything that was added. It is only after I expand it that nodes added subsequently are not showing up in the display on the subset tree.

Still not sure what I could do to ensure nodes are added and displayed properly on the tree when it has already been expanded.

Hi, basically what I am working on is there is one JTree that always contains the full tree (depth 2) for a particular dataset. There is next to this a JTree that is initially empty (except for a root) to which the user can add or remove nodes from the master tree arbitrarily.

If the user clicks on a node in the master tree, the "Select" button is enabled, any selections in the subset tree are cleared, and the user can add to the subset tree, if the user selects a node in the subset tree, the "Unselect" button is enabled, any seletions in the master tree are cleared, and the user can remove from the subset tree, but none of this makes any change to the master tree.

Will want to be able to work at any depth, such that clicking the root node in the master tree and then select will transfer the entire tree to the subset tree, selecting a mid-level node will transfer that node and all of its children to the subset, and selecting a leaf node will transfer only that leaf node (and its parent if it's not already there) to the subset tree. Right now I am focusing on the second of the three problems.

I am finding that doing it once through things work ok, you can add a node and then remove it from the subset tree. Trying to add a second node, or re-adding the original node after it has been added and removed once, seems to change the model yet I cannot seem to get it to display, can't really figure out what would be different in the logic flow the second time around.

Here is the method that is called once the "Select" button has been clicked, and it is determined which node in the master tree was selected (which is passed as the argument nodeToAdd. Thanks if you can shed any light on this.
public void addTranscriptExpressionStudy(DefaultMutableTreeNode nodeToAdd) { //We have a root node, the children nodes of the root will be TranscriptExpressionStudy //The children of TranscriptExpressionStudy nodes are leaf nodes, BiologicalSamples //selNode is the node corresponding to a TranscriptExpressionStudy that was selected //from the master list TranscriptExpressionStudy studyToAdd = (TranscriptExpressionStudy)nodeToAdd.getUserObject(); DefaultTreeModel subsetModel = (DefaultTreeModel)subsetTree.getModel(); DefaultMutableTreeNode rootNode = (DefaultMutableTreeNode)subsetModel.getRoot(); Enumeration<DefaultMutableTreeNode> enumeration = rootNode.children(); boolean found = false; DefaultMutableTreeNode newNode = null; //Check to see if the node corresponding to the sample has already been added while(enumeration.hasMoreElements()) { DefaultMutableTreeNode nodeAlreadyAdded = enumeration.nextElement(); TranscriptExpressionStudy studyAlreadyAdded = (TranscriptExpressionStudy)nodeAlreadyAdded.getUserObject(); //The study has already been added to the subset tree if(studyAlreadyAdded.getName().equals(studyToAdd.getName())) { newNode = nodeAlreadyAdded; found = true; break; } } //If not found then we create a new node that we will add to the subset tree if(!found) { newNode = new DefaultMutableTreeNode(studyToAdd); } //Now we will add the child nodes of nodeToAdd. Some of these may have been //added previously so before adding one we need to check that it hasn't already been added //I know this procedure for searching for nodes already added is not efficient //and involves a linear scan each time, but for simplicity for now doing it this way Enumeration<DefaultMutableTreeNode> nodeToAddChildren = nodeToAdd.children(); while(nodeToAddChildren.hasMoreElements()) { Enumeration<DefaultMutableTreeNode> alreadyAddedChildren = newNode.children(); DefaultMutableTreeNode sampleNodeToAdd = nodeToAddChildren.nextElement(); BiologicalSample sampleToAdd = (BiologicalSample)sampleNodeToAdd.getUserObject(); boolean sampleFound = false; while(alreadyAddedChildren.hasMoreElements()) { DefaultMutableTreeNode nodeAlreadyAdded = alreadyAddedChildren.nextElement(); BiologicalSample sampleAlreadyAdded = (BiologicalSample)nodeAlreadyAdded.getUserObject(); if(sampleAlreadyAdded.getName().equals(sampleToAdd.getName())) { sampleFound = true; break; } } if(!sampleFound) { DefaultMutableTreeNode newSampleNode = new DefaultMutableTreeNode(sampleToAdd); newNode.add(newSampleNode); } } rootNode.add(newNode); revalidate(); }

Thanks Mike:

It seems that including the option "ReferencedColumnName" in both the OneToMany and ManyToOne definitions is what finally did the trick.

Hi, I was just trying to work on an O/R mapping in EJB3. Unfortunately altering the database schema is not an option (or at the very least is a last resort). Basically there are a whole bunch of tables in which one of the columns is called "Attributes_ID". Though not explicitly defined in the schema, this attributes_id maps to the column "ID" in the table "Attributetext". Because multiple tables have this relationship, no foreign key can be defined within the Attributetext table.

I was wondering whether it is still possible to have a OneToMany mapping done correctly with this. In a java file defining a pojo corresponding to a table (let's call it "Feature") I include the following:

@OneToMany(fetch = FetchType.EAGER)
@JoinColumn(name="Attributes_ID")
public Collection<Attributetext> getAttributeText() {
return attributeText;
}

public void setAttributeText(Collection<Attributetext> attributeText) {
this.attributeText =attributeText;
}


The primary key of the "Feature" table is a column called "ID" and what happens with this is for each feature the attributeText row(s) with attributes_id equal to the feature's ID value are retrieved. What I want is for the attributeText row(s) with attributes_id equal to the features attributes_id value to be retrieved and put into the Collection. It seems that any example of a OneToMany mapping I can dig up, the join is being done with the primary key column in your "One" table (in this example Feature), to that column specified by "@JoinColumn" in your "Many" table (in this example AttributeText).

Is there any option that can be set to instead do the join on a non-key column in the "One" table?

Thanks sincerely