On idle-handler problem

i want to create an stay-open-applescript-application with "on idle"-handler.
to prevent accidental quitting of the script, i'm additional using the "on quit"-handler. but if the quit-handler will not be continued, the idle-handler doesn't continue his work. the application is still open, but nothing happens. by switching to the Finder and following switch to my stay-open-script the idle-handler of it still working again.
is this problem a bug in the current applescript-version (1.10.7)?
here is a test-script:
on idle
beep
return 2
end idle
on quit
if button returned of (display dialog "Really quit?" buttons {"Continue", "Quit"} default button {"Continue"}) is "Quit" then continue quit
end quit
Message was edited by: hohabadu
Message was edited by: hohabadu

Similar Messages

Exception Handling Problem In BPM

All
I am facing an exception handling problem I am using BPM and , I have caught exception in the transformation step but when there is any data problem in that mapping(mentioned in the transformation)
it is not throwing the exception . is there any option to collect these type of system exception in the bpm and give a alert thru mail
is there any way to collect these type of exception happened in the BPE and raise alert thru generic alert
Thanks
Jayaraman

Hi Jayaraman,
When you say there is any data problem, does that fail the message mapping that you have defined?
If the message mapping used in the tranformation fails, it should raise an exception in the BPM.
Did you test the message mapping using the payload and see if it really fails or not?
Regards,
Ravi Kanth Talagana

GUI event handling problems appear in 1.4.1 vs. 1.3.1?

Hi,
Has anyone else experienced strange event handling problems when migrating from 1.3.1 to 1.4.1? My GUI applications that make use of Swing's AbstractTableModel suddenly don't track mouse and selection events quickly anymore. Formerly zippy tables are now very unresponsive to user interactions.
I've run the code through JProbe under both 1.3 and 1.4 and see no differences in the profiles, yet the 1.4.1 version is virtually unusable. I had hoped that JProbe would show me that some low-level event-handling related or drawing method was getting wailed on in 1.4, but that was not the case.
My only guess is that the existing installation of 1.3.1 is interfering with the 1.4.1 installation is some way. Any thoughts on that before I trash the 1.3.1 installation (which I'm slightly reluctant to do)?
My platform is Windows XP Pro on a 2GHz P4 with 1GB RAM.
Here's my test case:
import javax.swing.table.AbstractTableModel;
import javax.swing.*;
import java.awt.*;
public class VerySimpleTableModel extends AbstractTableModel
private int d_rows = 0;
private int d_cols = 0;
private String[][] d_data = null;
public VerySimpleTableModel(int rows,int cols)
System.err.println("Creating table of size [" + rows + "," + cols +
d_rows = rows;
d_cols = cols;
d_data = new String[d_rows][d_cols];
int r = 0;
while (r < d_rows){
int c = 0;
while (c < d_cols){
d_data[r][c] = new String("[" + r + "," + c + "]");
c++;
r++;
System.err.println("Done.");
public int getRowCount()
return d_rows;
public int getColumnCount()
return d_cols;
public Object getValueAt(int rowIndex, int columnIndex)
return d_data[rowIndex][columnIndex];
public static void main(String[] args)
System.err.println( "1.4..." );
JFrame window = new JFrame();
window.setDefaultCloseOperation(JFrame.EXIT_ON_CLOSE);
JPanel panel = new JPanel();
Dimension size = new Dimension(500,500);
panel.setMinimumSize(size);
panel.setMaximumSize(size);
JTable table = new JTable(new VerySimpleTableModel(40,5));
panel.add(table);
window.getContentPane().add(panel);
window.setSize(new Dimension(600,800));
window.validate();
window.setVisible(true);
Thanks in advance!!
- Dean

Hi,
I've fixed the problem by upgrading to 1.4.1_02. I was on 1.4.1_01.
I did further narrow down the symptoms more. It seemed the further the distance from the previous mouse click, the longer it would take for the table row to highlight. So, clicking on row 1, then 2, was much faster than clicking on row 1, then row 40.
If no one else has seen this problem -- good! I wouldn't wish the tremendous waste of time I've had on anyone!
- Dean

Labview slider event handling problems - revisited

This topic asked a question that was very close to a problem I am having:
Labview slider event handling problems
That is, how do I, using an event structure and/or other means, only read out the value of a slider control after the value change is finalized?
The extra constraint I'd like to place on this, which I believe will invalidate the answer given in the thread above, is that my slider control also has a digital display which can also be used to change the value without ever using the mouse. I cannot look for a value change followed by a mouse-up event if the mouse was not used to change the value.
Any ideas how this might be accomplished? FWIW, I am using LabVIEW 7.0

Each and every incremental value-change event generated by the movement of the slider is still detected and queued up for use by the event structure and the event structure loop wades through them all before it's done.
I have come up the attached "fix" (LV v7.0) for this problem. While it is not as clean a solution as I had hoped it would be, it's the best I can manage given what LabVIEW offers me to work with and it does work in the situation where I need to use it.
Now, within the Finalize Slider Events subVI, I'm keeping track of the most-recent values seen by the subVI, checking to see if they have changed, and reporting out that fact. That gives me a Changed/Not-Changed bit within the event case that I can use to control what code then gets executed. If the event case is just playing catch-up to the real value of the control, I can now see that fact and ignore it.
In this version I've also dumped the variant output and limited the VI to using DBL values. I decided it added complication to something that was too complicated already and I wanted the output terminals for other purposes anyway (reporting of the correct "OldVal" of the control).Message Edited by Warren Massey on 04-28-2005 03:56 AM
Attachments:
slider_events[5].llb ‏77 KB

Delicate ProgressMonitor/ event handling problem

Hi,
I have a delicate ProgressMonitor / event handling problem.
There exists some solution on similiar problems here in the forums, but no solution to my special problem, hope anybody can help me out.
I have a rather big project, an applet, that I've written a separate JarLoader class that will load specific jar's if the ClassLoader encounter a unloaded class, all ok so far.
But, during loading, I would like to display a ProgressBar. And here is the problem. If my custom ClassLoader intercepts a findClass -request that needs to load the class from a jar from a normal thread, this is no problem, but if the ClassLoader intercepts a findClass that needs loading a jar when inside the EventQueue -thread, this is becomming tricky!
The catch is that an event posted on the EventQueue finally needs a class that needs to be loaded, but I cannot dispatch a thread to do this and end that particular event before the class itself is loaded.
So how do I hold the current EventQueue -event needing a jar -load, Pop up a ProgressBar, then process events in the EventQueue to display the ProgressBar and update repaints in it? then return from the event that now have loaded needed class-files.
I've tried a tip described earlier in the forums by trying to handle events with this code when loading the Jar:
    /** process any waiting events - use during long operations
     * to update UI */
    static public void doEvents() {
        // need to derive from EventQueue otherwise
        // don't have access to protected method dispatchEvent()
        class EvtQueue extends java.awt.EventQueue {
            public void doEvents() {
                Toolkit toolKit = Toolkit.getDefaultToolkit();
                EventQueue evtQueue = toolKit.getSystemEventQueue();
                // loop whilst there are events to process
                while (evtQueue.peekEvent() != null) {
                    try {
                        // if there are then get the event
                        AWTEvent evt = evtQueue.getNextEvent();
                        // and dispatch it
                        super.dispatchEvent(evt);
                    catch (java.lang.InterruptedException e) {
                        // if we get an exception in getNextEvent()
                        // do nothing
        // create an instance of our new class
        EvtQueue evtQueue = new EvtQueue();
        // and call the doEvents method to process the events.
        evtQueue.doEvents();
    }Then, the loader checks
java.awt.EventQueue.isDispatchThread()to see if its' inside the eventqueue, and runs doEvents after updating the ProgressMonitor with new setProgress and setNote values.
More precise;
The loader is loading the jar like this:
(this is pseudo code, not really usable, but outlines the vital parts)
public void load() {
monitor = new ProgressMonitor(null, "Loading " + jarName, ""+jarSize + " bytes", 0, jarSize);
monitor.setMillisToDecideToPopup(0);
monitor.setMillisToPopup(0);
// reading jar info code here ...
JarEntry zip = input.getNextJarEntry();
for (;zip != null;) {
     // misc file handling here... total = current bytes read
     monitor.setProgress(total);
     monitor.setNote(note);
     if (java.awt.EventQueue.isDispatchThread()) {
        doEvents();
     zip = input.getNextJarEntry();
monitor.close();
}When debugging doEvents(), there is never any events pending in peekEvents(), even if I tries to put a invokeLater() to run the setProgress on the monitor, why? If it had worked, the doEvents() code would have saved my day...
So, this is where I'm totally stuck...

Just want to describe how I did this using spin from http://spin.sourceforge.net
This example is not pretty, but it can probably help others understanding spin. Cancelling the ProgressMonitor is not implemented, but can easily be done by checking on ProgressMonitor.isCanceled() in
the implementation code.
First, I create a bean for displaying and run a ProgressMonitor:
Spin requires an Interface and an Implementation, but that's just nice programming practice :-)
import java.util.*;
public interface ProgressMonitorBean {
    public void start(); // start spinning
    public void cancel(); // cancel
    public int getProgress(); // get the current progress value
    public void setProgress(int progress); // set progress value
    public void addObserver(Observer observer); // observer on the progressValue
}Then, I created the implementation:
import java.util.*;
import javax.swing.ProgressMonitor;
import java.awt.*;
public class ProgressMonitorBeanImpl extends Observable implements ProgressMonitorBean {
    private ProgressMonitor monitor;
    private boolean cancelled;
    private int progress = 0;
    private int currentprogress = 0;
    public ProgressMonitorBeanImpl(Component parent, Object message, String note, int min, int max) {
        monitor = new ProgressMonitor(parent, message, note, min, max);
        monitor.setMillisToDecideToPopup(0);
        monitor.setMillisToPopup(0);
    public void cancel() {
        monitor.close();
    public void start() {
        cancelled = false;
        progress = 0;
        currentprogress = 0;
        while (progress < m_cMonitor.getMaximum()) {
            try {
                synchronized (this) {
                    wait(50); // Spinning with 50 ms delay
                if (progress != currentprogress) { // change only when different from previous value
                    setChanged();
                    notifyObservers(new Integer(progress));
                    monitor.setProgress(progress);
                    currentprogress = progress;
            } catch (InterruptedException ex) {
                // ignore
            if (cancelled) {
                break;
    public int getProgress() {
        return progress;
    public void setProgress(int progress) {
        this.progress = progress;
}in my class/jarloader code, something like this is done:
public void load() {
ProgressMonitorBean monitor = (ProgressMonitorBean)Spin.off(new ProgressMonitorBeanImpl(null, "Loading " + url,"", 0, jarSize));
Thread t = new Thread() {
    public void run() {
      JarEntry zip = input.getNextJarEntry();
      for (;zip != null;) {
        // misc file handling here... progress = current bytes read
     monitor.setProgress(progress);
t.start(); // fire off loadin into own thread to do time consuming work
monitor.start(); // this will spin events on monitor and block until progress = max
monitor.cancel(); // Just make sure ProgressMonitor is closed
}The beautiful thing here is that Spin is taking care of weither the load() is inside the dispatch thread or not, making the code much simpler and cleaner to understand.
The ProgressMonitorBeanImplementation could been made much nicer and more complete, but I was in a hurry to check if it worked out. And it did! This will be applied on a lot of gui -components that blocks the event-queue in our project, making it much more responsive.
Hope this will help others in similiar situations! Thanks again svenmeier for pointing me to the spin -project.

IP device tracking and idle timer problem

Hi,
We are deploying 802.1X in our network and have encountered problem with a type of payment terminal.
The problem is that the terminal do not 'speak' to the network after the first initial DHCP request, the terminal waits for incoming packets from a counter to start the payment process. After the idle-time the MAC is flushed from the switch and the port is not authorized any more.
To solve this we set 'authentication control-direction in' on the port and use 'ip device tracking' to keep the client on the network, ip device tracking sends an arp request every 30 seconds to clients.
Our ISE is sending Radius:Idle-Timeout = 300 and the timer start to count down when the client is authenticated.
In Wireshark, I can see that the ARP request is going out and the ARP reply coming back in but this does not update the inactivity timer for the client. So after 5 minutes the port is gone, and there is no way to get the port up again from the network. Traffic from the client brings up the network.
This looks like a bug to me, anyone seen this, or a similar behaviour?
Running:
ISE 1.2p6
IOS 12.2(55)SE6
From Trustsec 1.99 Wired 802.1X Deployment Guide:
Tip Enable IP Device Tracking with inactivity timers to keep quiet endpoints connected. When IP Device Tracking is enabled, the switch periodically sends ARP probes to endpoints in the IP Device Tracking table (which is initially populated by DHCP requests or ARP from the end point). As long as the endpoint is connected and responds to these probes, the inactivity timer is not triggered and the endpoint is not inadvertently removed from the network.
From CLI output
SW03#sh auth sessions int fa0/4
            Interface: FastEthernet0/4
          MAC Address: xxxx.xxxx.5289
           IP Address: 10.10.10.64
            User-Name: XX-XX-XX-XX-52-89
               Status: Authz Success
               Domain: DATA
       Oper host mode: multi-auth
     Oper control dir: both
        Authorized By: Authentication Server
           Vlan Group: N/A
      Session timeout: N/A
         Idle timeout: 300s (server), Remaining: 2s
    Common Session ID: 0A17BD07000000A925152A7B
      Acct Session ID: 0x00000458
               Handle: 0x090000A9
Runnable methods list:
       Method   State
       dot1x    Failed over
       mab      Authc Success
SW03#
SW03#
SW03#
SW03#sh auth sessions int fa0/4
            Interface: FastEthernet0/4
          MAC Address: Unknown
           IP Address: Unknown
               Status: Running
               Domain: UNKNOWN
       Oper host mode: multi-auth
     Oper control dir: both
      Session timeout: N/A
         Idle timeout: N/A
    Common Session ID: 0A17BD07000000AA251A0019
      Acct Session ID: 0x00000462
               Handle: 0x800000AA
Runnable methods list:
       Method   State
       dot1x    Running
       mab      Not run

Here is the port config.
Just to clarify, everything is working except that the terminal is losing the authentication. The terminal works again if traffic is initiated from the terminals menu, like with ping.
interface FastEthernet0/4
description Standard
switchport access vlan xxx
switchport mode access
switchport block unicast
switchport voice vlan xxx
switchport port-security maximum 2
switchport port-security
switchport port-security aging time 5
switchport port-security violation restrict
priority-queue out
authentication control-direction in
authentication event fail action next-method
authentication event server dead action reinitialize vlan xxx
authentication event server dead action authorize voice
authentication event server alive action reinitialize
authentication host-mode multi-auth
authentication order dot1x mab
authentication priority dot1x mab
authentication port-control auto
authentication periodic
authentication timer reauthenticate server
authentication timer inactivity server
authentication violation restrict
mab
no snmp trap link-status
dot1x pae authenticator
dot1x timeout tx-period 5
storm-control broadcast level pps 100
storm-control multicast level pps 100
storm-control action trap
spanning-tree portfast
service-policy input users

Request.getParameter(): special character handling problem

Hello, there:
This should be an easy question for the gurus in here.
I use a string attached to the url as the parameter string, which includes spaces. the system automatically converts them to %20, but when I receive them in my servlet using request.getParameter, a special character, Â appears; I couldn't proceed with it.
I suppose this is a common problem but seems that I didn't find topics about it on this forum.
So anyone can help? Any encoder/decoder APIs to handle it?
Thanks a lot,
Sway

hi,
escape can solve ur problem
use js function to submit form like
function subForm()
var any_spcl_value ="kdfhjdf$%@#$% 2FGSFG @%@#%@# V";
mainForm.action = "index.jsp?value="+ escape(any_spcl_value);
mainForm.submit();
and use request.getParameter("value");
u ll get kdfhjdf$%@#$% 2FGSFG @%@#%@# V as it is.
AE

BSP extension "download" - event handling problem

Folks,
First of all, I apologize if the answer is already out there, but I spent quite some time to find the solution for my problem but finally gave up:
I implemented the Download extension from Brian's and Tom's book. Since I am using an MVC based app, I put the event handling in the DO_HANDLE_EVENT of a controller instead of the OnInputProcessing event of a page. Here's how my app is designed:
I have a main controller that has an execute button to trigger a data selection based on some selection criteria. The result of this selection is displayed by means of a subcontroller with tableviews that are wrapped into my zdownload extension. Everything works fine: I make my selections, execute the app through the button of the main controller and get my result shown in the tableviews. I push the download icon, select my file format (XLS, HTML etc), get the dialog to choose whether to download or to display the file, perfect!
But now after I processed the download successfully - if I click the execute button in the main controller again, I still get the download dialog. It looks like the created event for the download does not get cleared. As I mentioned, I searched through SDN, but all I found were referrals to the call of DISPATCH_INPUT( ) in DO_REQUEST of the main controller, but unfortunately this one is already implemented.
Any ideas?
Thanks,
Guenther

There is an attribute of the download extension called display_url. You pass a value to the extension telling it what cached URL contains the binary content for your download. if this display_url attribute has a value assigned to it, then the element will render an iFrame to call the URl. However if the display_url attribute is empty, it won't render the iFrame. I suspect that you have a stateful application and after your download event you aren't clearing whatever field that you pass into the display_url attribute.

Event handler problems

Hi,
I'm having problems with a event handler that I'm trying to create.
I currently have 2 event handlers, one of which needs to only happen when Shift + a key is pressed. I currently have:
function keypressedHandler(event:KeyboardEvent):void {      if ((event.shiftKey) + (event.keyCode == 81)) {        //something would happen      } }
So my question is how can I get the shift function to work? It doesn't matter if it isn't shift but Alt or Ctrl would be the alternatives.
Any help would be great,
Thanks,

You need the ampersand (&) instead of the plus sign to use the logical AND:
stage.addEventListener( KeyboardEvent.KEY_DOWN, keypressedHandler )
function keypressedHandler(event:KeyboardEvent):void
    if ((event.shiftKey) && (event.keyCode == 65))
        //something would happen
        trace( "shift and 65" )

IDLE event problem with newer versions of the flash player

Greetings,
I have an issue that seems to be related to newer versions of the flashplayer in how they handle IDLE events. I have an IDLE event handler that works as expected with flashplayer version 10.0.45.2. With flashplayer 10.2.153.1, the IDLE event does not fire when my flex application is not in focus (browser minized, different browser tab in focus,etc...). Or when it fires, it's usually after a longer lag than normal. Has anyone experienced this issue? Thanks

Thanks for the reply. Are there any fixes/workarounds for this issue? Say I have a need to create a timer that fires every 10 minutes regardless of whether the player is in focus or not, how do I implement such a timer? Same goes for the IDLE event, I want to be notified if the application has been idle for X minutes regardless of whether it's in focus or not.Prior to 10.2 (and per your reply, 10.1), this all worked fine in flex without needing to adjust for the player being in focus.

Inheritance and mouse handler problem

I have a super class and two subclass which is extend the super class. I add a mouse handler in one of the subclass. The problem is that the other subclass also affect by the mouse handler. How can i avoid that??
here is the code
public abstract class SketchView extends JPanel implements Observer, Constants, ActionListener, Printable, Pageable
public SketchView(Sketcher theApp, SketchModel sketcherModel)
this.sketcherModel = sketcherModel;
this.theApp = theApp;
public static class DocView extends SketchView
public DocView(Sketcher theApp, SketchModel sketcherModel)
super(theApp, sketcherModel);
MouseHandler handler = new MouseHandler();
class MouseHandler extends MouseInputAdapter
public static class ReportView extends SketchView
public ReportView(Sketcher theApp, SketchModel sketcherModel)
super(theApp, sketcherModel);

to: marsian27
thank for you reply
here is the code
public abstract class SketchView extends JPanel implements Observer, Constants, ActionListener, Printable, Pageable
public SketchView(Sketcher theApp, SketchModel sketcherModel)
this.sketcherModel = sketcherModel;
this.theApp = theApp;
public static class DocView extends SketchView
public DocView(Sketcher theApp, SketchModel sketcherModel)
super(theApp, sketcherModel);
MouseHandler handler = new MouseHandler();
addMouseListener(handler; addMouseMotionListener(handler);
class MouseHandler extends MouseInputAdapter
public static class ReportView extends SketchView
public ReportView(Sketcher theApp, SketchModel sketcherModel)
super(theApp, sketcherModel);

ColdFusion.Ajax.submitForm error handler problem with AppleWebKit-based browsers

All-
When using an AppleWebKit-based browser (i.e. Safari or
Chrome), the
message argument of the Coldfusion.Ajax.submitForm error
handler always contains the string "OK", regardless of what error
was cfthrown by my CFM form handler.
Recall that the signature for the error handler is this:
MyFormErrorHandler = function( code, message )
In other browsers (i.e. IE or Firefox), the
message argument of the error handler properly contains the
message that was cfthrown by my form handler.
Can anyone offer any guidance for this problem? Thanks in
advance for your help.
-Josh

I was testing on IE8, Windows7 Starter. I don't have IE6 available to test on. Can test on IE7 a bit later on, but I have to fart-arse around with a VM to get it ready, so I'll do that after work.
I tested multiple submits both via clicking between the tabs within one main "request" as well as multiple entire requests (like via refreshing the whole browser window). I had no problems.
I think you're being quite optimistic ot expect AJAX stuff to work reliably on IE6. IE6 crashes. That's what it's most famous for doing. One needs to expect crashes if one pushes it (push it anywhere than out the window, that is). And your client company's IT dept should be lined up against a wall and shot for still running it. That's professionally negligent IMO. But... you probably don't need to be told that.
Adam

Web Service and result handler problem

Hi!
I load WSDL file and then invoke addnumbers() method without
problems. But I don't get any results. resultHandler and
faultHandler silent after addnumbers invoke. Where is the
problem???
My code:
public function useWebService():void {
WS = new WebService();
WS.wsdl = "................?wsdl"
WS.addEventListener("load", loadHandler);
WS.addEventListener("fault",faultHandler);
WS.useProxy = false;
WS.loadWSDL();
WS.addnumbers.addEventListener("result", resultHandler);
WS.addnumbers.addEventListener("fault", faultHandler);
public function loadHandler(event:LoadEvent):void {
Alert.show("WSDL is loaded");
WS.addnumbers(2,5);
public function resultHandler(event:ResultEvent):void {
Alert.show("in result handler");
myTextArea.text = event.result.toString();
public function faultHandler(event:FaultEvent):void {
Alert.show("fault: "+ event.toString());
My WSDL:
<?xml version="1.0" encoding="UTF-8"?><definitions
xmlns="
http://schemas.xmlsoap.org/wsdl/"
xmlns:tns="urn:TestWebservice/wsdl"
xmlns:ns2="urn:TestWebservice/types" xmlns:xsd="
http://www.w3.org/2001/XMLSchema"
xmlns:soap="
http://schemas.xmlsoap.org/wsdl/soap/"
name="TestWebservice" targetNamespace="urn:TestWebservice/wsdl">
<types>
<schema xmlns="
http://www.w3.org/2001/XMLSchema"
xmlns:tns="urn:TestWebservice/types" xmlns:soap11-enc="
http://schemas.xmlsoap.org/soap/encoding/"
xmlns:xsi="
http://www.w3.org/2001/XMLSchema-instance"
xmlns:wsdl="
http://schemas.xmlsoap.org/wsdl/"
targetNamespace="urn:TestWebservice/types">
<complexType name="addnumbers">
<sequence>
<element name="int_1" type="int"/>
<element name="int_2"
type="int"/></sequence></complexType>
<complexType name="addnumbersResponse">
<sequence>
<element name="result"
type="int"/></sequence></complexType>
<element name="addnumbers" type="tns:addnumbers"/>
<element name="addnumbersResponse"
type="tns:addnumbersResponse"/></schema></types>
<message name="TestWebserviceSEI_addnumbers">
<part name="parameters"
element="ns2:addnumbers"/></message>
<message name="TestWebserviceSEI_addnumbersResponse">
<part name="result"
element="ns2:addnumbersResponse"/></message>
<portType name="TestWebserviceSEI">
<operation name="addnumbers">
<input message="tns:TestWebserviceSEI_addnumbers"/>
<output
message="tns:TestWebserviceSEI_addnumbersResponse"/></operation></portType>
<binding name="TestWebserviceSEIBinding"
type="tns:TestWebserviceSEI">
<soap:binding transport="
http://schemas.xmlsoap.org/soap/http"
style="document"/>
<operation name="addnumbers">
<soap:operation soapAction=""/>
<input>
<soap:body use="literal"/></input>
<output>
<soap:body
use="literal"/></output></operation></binding>
<service name="TestWebservice">
<port name="TestWebserviceSEIPort"
binding="tns:TestWebserviceSEIBinding">
<soap:address location="
http://................../FlexWebserviceTest2/TestWebservice"
xmlns:wsdl="
http://schemas.xmlsoap.org/wsdl/"
xmlns:soap12="
http://schemas.xmlsoap.org/wsdl/soap12/"/></port></service></definitions>
--------------------

Here it is:
<mx:WebService
id="WS"
wsdl="
http://................................TestWebservice?wsdl"
useProxy="false"
fault="Alert.show(event.fault.faultString), 'Error'"
showBusyCursor="true" >
<mx:operation name="addnumbers" >
<mx:request>
<int_1>
2
</int_1>
<int_2>
5
</int_2>
</mx:request>
</mx:operation>
</mx:WebService>
<mx:Panel>
<mx:VBox >
<mx:HBox fontSize="12">
<mx:Button label="Go!" click="WS.addnumbers.send()"/>
</mx:HBox>
<mx:HDividedBox width="800">
<mx:TextArea id="myTextArea"
text="{WS.addnumbers.lastResult.toXMLString()}" width="390"
height="400" fontSize="12"/>
</mx:HDividedBox>
</mx:VBox>
</mx:Panel>
----------------

SBO2004: Event Handling problems

Hello, all
I have noticed a very peculiar thing about the way SBO handles events.
I wrote a handler for the et_VALIDATE event on an EditText item (Sales Quotation->Customer) which does nothing but waits for two minutes. Then I performed the following experiment:
Opened a sales quotation from in the Add mode, chose a Customer and added an item. Them I put the cursor into the Cusotmer field and pressed "Add". As a result, my event handler fired, so SBO was "frozen" for two minutes. But when the handler had completed nothing happened. The quotation wasn't added, although I didn't set BubbleEvent to false in the eventhandler.
However, when I reduced the wait time to 30 seconds in the event handler everything started to work as it should: after completion of the eventhandler the quotation got added.
Looks like if the handler of the preceding event (VALIDATE) executes for too long the following event (ITEM_PRESSED) may not fire at all...
I this a correct behavior?
Thanks in advance,
Anton
PS: I was testing in SBO2004

Please update your system. Version 2004 is no longer supported.
Hello, Vitor
I know this, but I can't update soon. So I hope somebody can either confirm this problem or suggest a solution/workaround. But OK, I will check this in SBO2007.
Anton

[HELP] ALE inbound process and Workflow handling problem

Hello,
first of all, i have to apologize about my english level. I will try to explain my problem (thanks for your patience ).
Well, I'm implementing an ALE inbound interface. My development at this point are:
- Customer Idoc Inbound function (with correct interface).
- Customer Basic Type
- Customer Message Type and correct assignment to the Basic Type (Tx WE81 and WE82)
- Customer Object Type (Subtype of IDOCAPPL)
- On BD51 I put "1" for my function.
- On WE57 I put this:
Processing by --> My function and type "F"
IDOC type --> My Basic type (without any extension)
Message --> My message type (without any message code or msg. function)
Object --> My object type (subtype of IDOCAPPL)
Direction --> "2" (Inbound)
- I created a process code for this interface (on WE42):
It is processed with ALE service and by a function module (my function).
On "Module(Inboud)" part I put my function and maximun number of repeats "0". On IDOC part I put my Object type with start event "INPUTERROROCURRED" and end event "INPUTFINISHED".
On application objet I put my Object type.
Rests of fields are blank.
- I defined a partner (WE20) with this new message type and process code.
- Also I created a Task (PFTC) for handle exceptions in inbound process. It has a rule for agent determinate.
I think it is all.
My problem is: when I process an IDOC (from WE19 for example) and it gets 51 status (error), it has to launch an event that trigger my task, is this way? Well I'm not getting any event.
I have all Workflow customizations OK (in this systems are running some Workflows).
In my function code, when I detect any error I put an error in the status table and put in WORKFLOW_RESULT the number '99999'.
Why I'm not getting any event? What I'm doing wrong?
Thanks in advance. Regards.

Hello again,
finally I've solved the problem. The solution for me, was a config on tx. WE42 and some missing code in the function, I only fill WORKFLOW_RESULT with '9999', but I didn't add any register to RETURN_VARIABLES table.
Regards.

On idle-handler problem

Similar Messages

Maybe you are looking for