Weird DocumentBuilderFactory.newInstance() behaviour

problem summary
==============
jvm breaks when I call DocumentBuilderFactory.newInstance(). I suspect its something to do with security but no exception is caught in the try...catch block where DocumentBuilderFactory.newInstance() is called.
Other info about the problem
======================
I'm working on an application that use 'presence service' and message queue service. When I use Sun's message queue service, jvm doesn't break. But when I use a SKype client (jskype) jvm always fail even though jskype doesn't reference DocumentBuilderFactory.newInstance() at all. Anyone?

That's great thinking Fred.
I've working on a much cut down and simplified version with just program changes to find out the cause of the behaviour.
I tried your note off suggestion without success, but your comment on the original multi being "blocked" led me to route the final events to a second copy of the multi instrument rather than the original, and it works - the custom bank changes are correctly output!
Maybe there's some kind of internal message validation or something, I'm not quite sure. But using a second copy of the multi instrument - though not ideal - is a perfectly acceptible workaround.
So thanks for the pointer - you really helped me out..!

Similar Messages

DocumentBuilderFactory.newInstance() only works after garbage collection

Hi all,
I am stucked with a strange behaviour of "DocumentBuilderFactory.newInstance()".
I use the DocumentBuilderFactory in an applet
to parse an xml-file.
My applet starts ok until the line 5 (newInstance) gets called.
At this time it seems that nothing happens anymore and that
the applet fails to continue.
But when I open up the java console and hit
g (garbage collection) the xml parsing continues immediatly
and my applet completes loading and runs ok.
1 private void genarteDomNodes() {
2 try {
3 // ---- Parse XML file ----
4 DocumentBuilderFactory factory =
5 DocumentBuilderFactory.newInstance();
6 DocumentBuilder builder =
7 factory.newDocumentBuilder();
8 Document document = builder.parse(
9 new URL(this.myModules_xml).openStream() );
10 // ---- Get list of nodes to given element tag name ----
11 NodeList ndList =
12 document.getElementsByTagName("meter-group");
13 printNodesFromList( ndList );
14 } catch( SAXParseException spe ) {
15 ...
16 } catch( SAXException sxe ) {
17 ...
18 } catch( ParserConfigurationException pce ) {
19 ...
20 } catch( IOException ioe ) {
21 ...
22 }
23 }

I am still stucked with this problem. But I found out how to enable JAXP debug output for applets (setting system properties isn't allowed for applets).
This puts somemore output to the java console. It might help someone to understand my problem. I also printed some debug messages to track down my problem.
Following is the console output of my applet:
URL of XML to parse = "http://10.110.132.195/c8000-modules.xml"
entering "genarteDomNodes"
just before calling "DocumentBuilderFactory.newInstance()"
JAXP: find factoryId =javax.xml.parsers.DocumentBuilderFactory
!!! at this time the applet "hangs" and nothing happens;
until I hit the "g" button. Then applet continues immediatly and prints:
JAXP: loaded from fallback value: com.sun.org.apache.xerces.internal.jaxp.DocumentBuilderFactoryImpl
JAXP: created new instance of class com.sun.org.apache.xerces.internal.jaxp.DocumentBuilderFactoryImpl using ClassLoader: sun.plugin.security.PluginClassLoader@32060c
After invoking the garbage collector the applet continue to parse the xml file and runs as expected.
Can someone help me please.

DocumentBuilderFactory.newInstance(); OR TransformerFactory.newInstance();

Usually when we create a reference to an class, we'll do:
Test tst = new Test();
But when we create a DocumentBuilderFactory or a TransformerFactory, we use:
DocumentBuilderFactory.newInstance();
or
TransformerFactory.newInstance();
Why use the newInstance() instead?
Thanks
Scott

Those are "factory" classes (as you might guess from their names). The DocumentBuilderFactory's newInstance() method returns an object that is an instance of DocumentBuilderFactory. If you read the documentation for the method, you will see the logic it goes through to decide which class it will produce an instance of.
This allows the user to configure the choice of DocumentBuilderFactory by changing the environment, rather than by having to modify source code and recompile.

DocumentBuilderFactory.newInstance()

Hi Guys,
I am writing a utility that does access a database table and an xml file, like the following.
public void reads(){
tableMap = readTable(tableName); // here I am making a DB connection.
xmlMap = readXML(tableName); //
public Map readXML(){
docFact=DocumentBuilderFactory.newInstance();
//��creates an xml document
Here the problem is while I am using that readTable function the application exits by the line �docFact=DocumentBuilderFactory.newInstance()�. and if the readTable function is not called it works fine!.
Biju.

Did you run if from GUI of from console? One of the possibilities is that your JRE is crashed (in that case it usually create a log file).

DocumentBuilderFactory.newInstance().newDocumentBuilder() error

import javax.xml.parsers.DocumentBuilder;
import javax.xml.parsers.DocumentBuilderFactory;
import org.w3c.dom.*;
public class FrameRecorder extends Thread {
     public Document dmt;
     public FrameRecorder() {
          DocumentBuilderFactory fctry = DocumentBuilderFactory.newInstance();
          fctry.setValidating(true);
          fctry.setNamespaceAware(true);
          fctry.setIgnoringElementContentWhitespace(true);
//the error is on the next line.
          DocumentBuilder bldr = fctry.newDocumentBuilder();
When I run it, I get an error saying:
Exception in thread "AWT-EventQueue-0" java.lang.Error: Unresolved compilation problem:
     Unhandled exception type ParserConfigurationException
     at FrameRecorder.<init>(FrameRecorder.java:11)
     at BaseFrame.<init>(BaseFrame.java:17)
     at Base.createGUI(Base.java:10)
     at Base$1.run(Base.java:7)
     at java.awt.event.InvocationEvent.dispatch(InvocationEvent.java:226)
     at java.awt.EventQueue.dispatchEvent(EventQueue.java:602)
     at java.awt.EventDispatchThread.pumpOneEventForFilters(EventDispatchThread.java:275)
     at java.awt.EventDispatchThread.pumpEventsForFilter(EventDispatchThread.java:200)
     at java.awt.EventDispatchThread.pumpEventsForHierarchy(EventDispatchThread.java:190)
     at java.awt.EventDispatchThread.pumpEvents(EventDispatchThread.java:185)
     at java.awt.EventDispatchThread.pumpEvents(EventDispatchThread.java:177)
     at java.awt.EventDispatchThread.run(EventDispatchThread.java:138)
I don't know what to do to fix this.
I'm running 64-bit Ubuntu 10.4 so could it be a compatibility problem? The only headway I could make was that .newDocumentBuilder() throws a ParserConfigurationException (that's the error I'm getting), and the docs tell me that .newDocumentBuilder() throws this when it "DocumentBuilder cannot be created which satisfies the configuration requested".
Any help?

Yes, somewhere in your code you have a compile error. I'm guessing you're using Eclipse and running your code from within it (or at least, running the Eclipse compiled classes from the command line).
You need to go and fix the compiler error. You already have all the information you need. You know where it is, you know what it is, and the documentation confirms it.

Address Book - Weird text field behaviour

In Address Book when I try to edit any text field weird things happen.
When I start typing all the previous characters are repeated with each keystroke - for example typing "Thomas" would become TThThoThomThomaThomas". Pressing the backspace, delete or the ESC key repeats only the keystrokes made, i.e. "Thomas" would be added to the end of "TThThoThomThomaThomas".
If I click into another field and start typing it still remembers all the previous keystrokes and puts all of them in front of the key I just pressed.
Additionally, nothing can be typed in the notes field.
I've tried deleting:
~/Library/Preferences/com.apple.AddressBook.abd.plist
~/Library/Preferences/com.apple.AddressBook.plist
but that doesn't fix it.
Anything in ~/Library/Application Support/Address Book be the problem?
When I log in a a guest user this behaviour is not present.

It's good that you tried this on a different account. That suggests that the problem exists somewhere in your user account directory. I don't think anything in ~/Library/Application Support/AddressBook/ would cause this. Do you have any input managers installed? These might be in ~/Library/Input Managers/ or something like that. I've never installed one myself, so I am not certain where they are. I think if you find one, you can just remove it from that folder by dragging it out, and then restarting Address Book might solve the issue.

X131e - weird browser tabs behaviour

Hello all. Took delivery of an X131e a few days ago. OVer the past few days I have noticed an intermittent issue with the behaviour of tabs in both Explorer and the latest version of Firefox I installed. Essentially, what happens is that every time I click on a link it opens in a new browser window. This didn't happen to start with and it goes back to normal after I restart Windows. This 'fault' seems to start up randomly. It happens in both Explorer and Firefox (so does not appear to be a browser-specific setting).
I am aware that some links are specifically set to open in a new tab/browser window by the page's designer, but this is every single link, even just Google search results. Select a link with left button above track pad and link opens in a new tab. Extremely frustrating. Restart computer, run exactly the same search and clicking the link works as it should. a day or two later and the tabbing thing has started again.
Machine is running Windows 7 Home Premium. I don't shut it down when I have finished with it; I just close the lid.
Can anybody suggest what might be causing it and how I might prevent it?
Solved!
Go to Solution.

Hi there,
From what I'm gathering, it's almost like you're clicking the link and even though you're not pressing Ctrl, a new tab opens like it would if you were holding that key, is that a correct assessment of the situation?
If so, have you checked Sticky Keys in Control Panel?
Ed
Was this or another post on the forum helpful? Click the star on the left side of the screen to give kudos! Did someone solve a problem you encountered? Mark it as "solution provided" to help others with the same problem.

Weird OT font behaviour in InDesign and Illustrator

Hi font experts.
I've been directed here by someone at the InDesign forum, who thought you might be able to help.
Can anyone explain what's happening here?
I've loaded an OT version of Zapf Dingbats into Suitcase Fusion on my Mac Pro and it seems just fine. Font Doctor says its OK. Suitcase Fusion says its OK. The font characters display correctly in Fusion.
I can use both the TT and OT versions of the font in Mail and Word. Both versions print OK, too.
But when I apply the OT font in Adobe InDesign or Illustrator, all I get to see is boxes with a cross through them and that "missing font" pinkish hue.
No other OT fonts pose this problem. And deleting the TrueType font of Zapf Dingbats makes no difference so it's not a duplicates problem.
As a result of the above I've temporarily ditched Fusion today to try out FontExplorer (which I like). Comparing the two fonts in there, I can see that there are lots of "missing" characters in the OT version if I select "Show missing characters", but none in the TT version. Not sure if that's relevant.
Also, each character in the OT version has a lot more information, including Character Code, Unicode, Glyph ID, HTML and Keystroke, whereas the TT version simply has the Glyph ID. Don't know if that's relevant either. Incidentally, the keystroke doesn't work in InDesign or Illustrator.
Thanks for any suggestions or solutions to this annoying issue. It seems very weird but maybe there's a simple explanation. Cheers.

Dang! It was so obvious. The OT version has the symbols in the correct place in the glyph palette. The letters and numbers that I used to type to get the bullets and scissors are empty characters in OT. Thanks to Carl over at Extensis for enlightening me. Boy was I being dim.

Weird Image Processor Behaviour

The image processor has started acting really weird a couple of hours ago. I have been trying to figure the problem out since then, but to no avail. It all started when I tried to run batch processor from the Bridge with a really simple action that is just meant to save the images as a TIFF. This has worked before and I did not change the setup in any way. But this time many of the images just stayed open in Photoshop. They were neither saved nor closed - others were, though. I played around with the settings, repaired permissions, deleted cache files, etc. Then at one point, when I ran the image processor again, it suddenly started to open random raw files! These files are on the same hard drive as the .psd files I was trying to convert, but they are not even nearly in the same directory. In fact, these happened to be really old pictures that I hadn't seen in a while. That scared me. Also I am at the end of my wits. The only thing left to do would be to uninstall and then reinstall Photoshop, I guess. But that would mean, that I would have to go through all the settings again, import my actions, keyboard shortcuts, etc. I'd like to avoid that if possible.
I am using Photoshop CS 5.0.1 on an iMac running MacOS 10.5.8.

You mention Image Processor and Batch, which are two different things – which did you employ?
Could you post a screenshot of the Actions Panel with that specific Action clicked open just to make sure which steps are included?
And have you tried trashing prefs (after making sure your presets are saved)?

Very weird af:forEach behaviour

I've got a page with a table in it and some filter code so the user can specify a string and only rows that have that string in any of several fields will be displayed. To refine the search, it can be repeated with another word which gets added to any that are already in there. Behind it is an ArrayList in a backing bean containing the individual search terms. The ArrayList gets passed to a method in an application module which builds a predicate out of them all and calls .setWhereClause on the relevant VO. This all works OK.
Then to indicate that the displayed list is filtered, I want an af:showDetail at the top which summarises the filter that's been applied and lets the user expand it to see the list of search terms. Something like this;
<af:panelBox rendered="#{backing_UserManagement_UserManagement.resultsFiltered}"
width="100%">
<af:panelTip>
<af:showDetail disclosed="false">
<af:panelList binding="#{backing_UserManagement_UserManagement.filterList}">
<af:forEach items="#{backing_UserManagement_UserManagement.criteria}"
var="thisSearchCriterium">
<af:outputText value="Username or real names contain '#{thisSearchCriterium}'"/>
</af:forEach>
</af:panelList>
<f:facet name="prompt">
<af:outputText value="#{backing_UserManagement_UserManagement.filterSummary}"/>
</f:facet>
</af:showDetail>
</af:panelTip>
</af:panelBox>
This in itself works OK, however elsewhere on the page very odd things happen whenever there is anything non-trivial inside the af:forEach. Columns become duplicated inside the table, and panelBox elements elsewhere on the page- and their contents- also become duplicated. If I reduce this right down to the bare minimum, the odd behaviour occurs with
<af:forEach items="#{backing_UserManagement_UserManagement.criteria}"
var="thisSearchCriterium">
<af:commandButton text="commandButton 1"/>
</af:forEach>
but not with
<af:forEach items="#{backing_UserManagement_UserManagement.criteria}"
var="thisSearchCriterium">
<pre> Hello World </pre><hr>
</af:forEach>
Though the "Hello World" lines appear at the top of the page, outside any ADF elements (even when defined inside an af:panelBox or similar) so this doesn't give me a way of getting to the result I want.
Has anyone seen this kind of thing before and solved it?

Hi,
put a print statement into your managed bean and see how often it gets executed. Usually this is 2 times the case. You can set a flag after the first run , or just set refresh as needed in the iterator binding with a refresh condition of prepareRender or prepareModel
Frank

Weird flash plugin behaviour

Hello guys and girls.
Recently I switched back from catalyst to the opensource driver. Everything worked fine after that.
Now I updated my system a week ago and for some reason my flash player behaves weirdly since then.
When I got 2 or more tabs open and all of them having flash active, flash stops displaying correctly when I close one of them.
Is this a known problem? I couldn't find any posts mentioning this, so I thought I should post a topic here.
Thanks in advance,
szt

I've been having flash issues lately too, although different ones from yours. Flash videos on lots of sites refuse to go fullscreen. sometimes they will go fullscreen once, but if I unfullscreen and try again it will be broken until I restart firefox. youtube is the only site it works reliably on. I resolved my issues by using chromium and chromium-pepper-flash, which is far more up to date and less buggy than flash 11.2

Weird Inspector hotkey behaviour

The built in Firefox web developer tool inspector can be called with CTRL+SHIFT+I. which will show the web developer toolbar and you can also select a desired element on the page. Now, when I want to select another element and press CTRL+SHIFT+I again, it closes the toolbar. WHY ON EARTH? Is there any way to eliminate this overly annoying toggle behaviour?
I don't want to click every time on an icon, to inspect an element. I want a shortcut which always brings up the toolbar and lets me inspect any element on the page. Like firebug's CTRL+SHIFT+C does it.
Ideas?

Ctrl+Shift+i works like Ctrl+Shift+k and other toggles that open/shut the relevant pane. I don't think it was ever intended to reactivate inspect mode. Someone needs to define a new keyboard shortcut for that.
I looked into using [http://forums.mozillazine.org/viewtopic.php?f=48&t=72994 keyconfig 20110522] for that, but I don't know how to invoke the inspector.highlighter.unlockAndFocus() method from an object available to the extension. I'm sure someone does.

Weird window decoration behaviour

Has anyone else experienced windows, particularly in Finder, though sometimes in Safari, that appear to be only "semi" active as in the screenshot linked below?
http://dl.dropbox.com/u/2459380/weird.png

Try selecting one of the Devices or Places in the Sidebar on the left.
If that doesn't help, go to ~/Library/Preferences and move the com.apple.finder.plist file to the Trash.
~ (Tilde) represents your Home Folder.
Restart your Mac.

Weird Quick Time Behaviour

Hi,
I have to translate a fla into DVD. The picture become soft
and blurry when exported in AVI, but the export in .mov produce a
very clean and crisp image. Re-exporting from QT Pro for import in
a video editor program works perfectly well. The problem comes from
randomly "sliding" frames. I'm not talking about flickering... It
looks as the picture has been spilt in the middle and the 2 parts
are sliding on top of each other.
The Fla is a 15 minutes corporate film designed for
projection with no action script involved. On these 15 minutes this
weird effect appear 4 or 5 times, lasting approx. 1/3 of a second.
Did anybody came accross this before? Is there an answer?
Thanks in advance,
Dominique

Hi,
I have to translate a fla into DVD. The picture become soft
and blurry when exported in AVI, but the export in .mov produce a
very clean and crisp image. Re-exporting from QT Pro for import in
a video editor program works perfectly well. The problem comes from
randomly "sliding" frames. I'm not talking about flickering... It
looks as the picture has been spilt in the middle and the 2 parts
are sliding on top of each other.
The Fla is a 15 minutes corporate film designed for
projection with no action script involved. On these 15 minutes this
weird effect appear 4 or 5 times, lasting approx. 1/3 of a second.
Did anybody came accross this before? Is there an answer?
Thanks in advance,
Dominique

Weird csv import behaviour...

Hi guys!
I was wondering if any of you encountered a werid csv import behaviour:
If the csv file I'd like to import has columns shortened...I mean sometimes you have to click twice on a column to see its full content..and I'm talking about the situation where you don't see it in a full spectrum. Then if im trying to import it I receive an error...if immediately after that I click twice on every column of that csv file...the import works fine...
Any ideas?
With regards,
PsmakR

Hi guys!
I was wondering if any of you encountered a werid csv import behaviour:
If the csv file I'd like to import has columns shortened...I mean sometimes you have to click twice on a column to see its full content..and I'm talking about the situation where you don't see it in a full spectrum. Then if im trying to import it I receive an error...if immediately after that I click twice on every column of that csv file...the import works fine...
Any ideas?
With regards,
PsmakR

Weird DocumentBuilderFactory.newInstance() behaviour

Similar Messages

Maybe you are looking for