User interfaces and event structures

I am just learning to use event structures to improve operation of front panels. At the moment, I am working on a tab strucrture we use to montinor/set program paramters. VArious parametera are displayed on each tab as controls (not indicators) and are constanly updated sice they can be modified elsewhere by other pieces of code. We controls are normally disabled via a property node so they behave more like indicators and the values are updated by writing to a local variable (I know this is evil but it works in this case). There is a boolean on each tab that toggles between "monitoring" and "changing" modes (or states). When toggles to "changing" (or setting) parameters the monitoring activities are suspended and the control are enabbled. After making the desired changes the controls are either read and the changes implimented or get rest based on some other boolean controls, the parameters controls are then disables and monitoring resumes. The issue is that many users fing the need to toggle between "monitorsing" and "changing" modes to be confusing. I want to replace this switch with an event. If the user clicks on any of the parameters controls in a tab, I want to go into changing mode rather than using the "monitoring"/"changing" boolean control. All other asspects are handled as before. My problem is than if I use a mousedown event to go into "changing" mode and the user clicks on several parameters to shift the keyboard focus to edit more than one parameter, these subsequent mose clciks get processed as new events when change mode is exited. This means that as soon as it returns to "monitoring" mode it jumps immeadiately back to "changing" mode. How should I structure the events to avoid this problem?
Thanks,
--Lee

Lee,
Instead of using a "mouse down" event, use a "value change" event for each
of the controls. Then your program can switch to the "changing" mode
whenever the value of the control is actually changed.
TJW
"Lee Robertson" <[email protected]> wrote in message
news:[email protected]...
>I am just learning to use event structures to improve operation of front
>panels.  At the moment, I am working on a tab strucrture we use to
>montinor/set program paramters.  VArious parametera are displayed on
>each tab as controls (not indicators) and are constanly updated sice they
>can be modified elsewhere by other pieces of code.  We controls are
>normally disabled via a property node so they behave more like indicators
>and the values are updated by writing to a local variable (I know this is
>evil but it works in this case).  There is a boolean on each tab that
>toggles between "monitoring" and "changing" modes (or states).  When
>toggles to "changing" (or setting) parameters the monitoring activities are
>suspended and the control are enabbled.  After making the desired
>changes the controls are either read and the changes implimented or get
>rest based on some other boolean controls, the parameters controls are then
>disables and monitoring resumes.  The issue is that many users fing
>the need to toggle between "monitorsing" and "changing" modes to be
>confusing.  I want to replace this switch with an event.  If the
>user clicks on any of the parameters controls in a tab, I want to go into
>changing mode rather than using the "monitoring"/"changing" boolean
>control.  All other asspects are handled as before.  My problem
>is than if I use a mousedown event to go into "changing" mode and the user
>clicks on several parameters to shift the keyboard focus to edit more than
>one parameter, these subsequent mose clciks get processed as new events
>when change mode is exited.  This means that as soon as it returns to
>"monitoring" mode it jumps immeadiately back to "changing" mode.  How
>should I structure the events to avoid this problem?
>  
> Thanks,
>  
> --Lee

Similar Messages

Vi with access via both user interface and external communication

Hi,
Maybe a little bit too general but I'll risk it:
It concerns a project to stear a tensile machine. Data acquisition is relative straightforward via a few analogue and digital channels.
The interaction of the operator with the tensile machine can be done at two levels:
1. Directly via an user interface on the computer next to the machine. Here I would typically use a state machine in combination with event structures.
2. Remotely, via TCP protocol. Here different clients (external program, written in C) should be able to connect to the LabVIEW program and send simple commands such as 'read force', 'write force', 'read status', ...
It is relatively easy to setup a small daemon that accepts TCP connections from multiple clients. Furthermore I would put the commands from the clients into a message queue (using a producer/consumer structure).
I was wondering what the best method is to allow both the user interface and the external clients to stear the machine (e.g. using an event structure to respond to interactions via the user interface, and create user events from the commands of the clients to access the same actions).
Steven
LV8.5

Hi VPS,
one way could be to build your TCP receiver in one while loop and your event structure in another while loop. if you want to send a command from the event structure you can send this over tcp to the TCP receiver.
Another way could be to use a queue or notifier to send the commands from the event structure and the TCP receiver to another while loop which work with the received commands.
Mike

INTERFACE AND EVENTS IN ABAP OO

Can some one explain me the real time need with example for use of interfaces and events in abap objects
(asking interface and events along doesnt mean that i am relating them).
<removed_by_moderator>
Pankaj Giri
Edited by: Julius Bussche on Jul 14, 2008 1:36 PM

Events :
Technically speaking :
" Events are notifications an object receives from, or transmits to, other objects or applications. Events allow objects to perform actions whenever a specific occurrence takes place. Microsoft Windows is an event-driven operating system, events can come from other objects, applications, or user input such as mouse clicks or key presses. "
Lets say you have an ALV - An editable one ...
Lats say - Once you press some button you want some kind of validation to be done.
How to do this ?
Raise an Event - Which is handled by a method and write the validation code.
Now you might argue, that I can do it in this way : Capture the function code - and call the validate method.
Yes, in this case it can be done.. But lets say .. you change a field in the ALV and you want the validation to be done as soon as he is done with typing.
Where is the function code here ? No function code... But there is an event here - The data changed event.
So you can raise a data changed event that can be handled and will do the validation.
It is not user friendly that you ask the user to press a button (to get the function code) for validation each time he enters a data.
The events can be raised by a system, or by a program also. So in this case the data changed event is raised by a system that you can handle.
Also, Lets say on a particular action you want some code to trigger. (You can take the same example of validation code). In this case the code to trigger is in a separate class. The object of which is not available here at this moment. (This case happens very frequently).
Advantage with events : Event handlers can be in a separate class also.
e.g : In the middle of some business logic .. you encounter a error. You want to send this information to the UI (to user - in form of a pop up) and then continue with some processing.
In many cases - A direct method call to trigger the pop up is not done. Because (in ideal cases) the engine must not interact with UI directly - Because the UI could be some other application - like a windows UI but the error comes from some SAP program.
So - A event is raised from the engine that is handled in the UI and a pop up is triggered.
Here -- I would have different classes (lets say for different Operating Systems). And all these classes must register to the event ERROR raised in application.
And these different classes for different Operation systems will have different code to raise a pop-up.
Now you can imagine : If you coded a pop-up for Windows (in your application logic) .. it will not work for Mac or Linux. But of you raise a event.. that is handled separately by a different UI classes for Win, Linux or Mac they will catch this event and process accordingly.
May be I complicated this explanation .... but I couldn't think of a simpler and concrete example.
Cheers.
Varun.

Run-Time Menu and Event Structure~~HELP~~

Hi, brothers,
I'm new user for LabVIEW. Use the version of LabVIEW is 8.6
One problem describe as below:
Can Run-Time Menu and Event Structure exist simultaneously?
Without Event Structure, Run-Time Menu works well, but Run-Time Menu works abnormally.after adding Event Structure,
How do I modify the problem?
Thank you for your help.
Attached is the program.
Attachments:
PC_Adjustment.zip ‏44 KB

Yes they can coexist, but not how you did it.
Every iteration of the while loop you wait for an event in the event structure, and you look if the user has selected anything from the menu.
To get this working you should add an event for <this VI>\Menu Selection (User):
Ton
PS you should add a stop button, the only way you can stop the VI is hitting the 'emergency stop' button, with hardware connected (like you have) this is not the adviced routine
Message Edited by TonP on 09-26-2008 07:34 AM
Free Code Capture Tool! Version 2.1.3 with comments, web-upload, back-save and snippets!
Nederlandse LabVIEW user groep www.lvug.nl
My LabVIEW Ideas
LabVIEW, programming like it should be!
Attachments:
Main_BD.png ‏5 KB

Nokia's user interface and software A terrible e...

(hope this is in the right section, here goes)
I got my E63 about a month ago. I thought it only fair to list my grievances with the phone so far, because I am really close to selling it and buying something else, and I think Nokia deserves to know the reasons why, and will hopefully improve in the future.
I will start off with what I love about the phone, because I do in fact, really like it. The slim size, excellent keyboard, great connectivity like 3G and Wi-fi, and the reasonable battery life, make it a great phone, but its in the user interface, and actual usage of the phone, where things really start to fall flat.
The user interface is very disorganized and difficult.
Why is everything so hard to find and use? For example, if I want to change the applications that appear on my home screen, I need to navigate the following directory tree:
Menu > Tools > Settings > General > Personalisation > Mode Settings > Home Screen Applications > Application Shortcuts > Shortcut 1
and then, when I finally get to the list, the items arent even in any order!
You must be joking, what a horrible menu! Its insane. As a user, I feel like its a terrible experience.
But thats not all. Why are there 2 notes applications on my phone? There's notes, and active notes. Why does Nokia OVI store exist twice, once in the main menu, and once under installations? Why do I have Contacts, and OVI contacts on my phone? Why can't my phone have just one app that works well, instead of 2 that differ slightly.
Why is email setup in the installations menu? Why isn't it under email, where mailboxes get set up? Its just sitting there cluttering up Installations, when its something you only use once or twice.
Not to mention the fact that the configuration settings are all over the place, i.e. there is an Email Key setting in Menu > Tools > Settings > General. Its all on its own, not near anything. Why isn't it in the same menu where the other shortcut keys get set? Or even in the email app settings?
I am not naive, I know that the E63 is on the cheaper end of the scale, so I can't expect it to be the fastest phone in the world, but scrolling through my email in HTML mode is so horribly slow, and the view tends to jump from left to right on some emails, I hardly even use the phone to check my email anymore, which is sad because that is the reason I chose the E63.( for the nice keypad and email syncing)
These are just SOME of the things that I have noticed. So please Nokia, PLEASE PLEASE PLEASE! Spend some of your R&D money on user interface and user experience!
If there is anything you take away from this letter, let it be this:
Doing anything on the phone just feels like hard work, it makes me not want to use the phone.
Your phones should be simple and easy to use, even without the mass of Help files that appear in a lot of the menus.
I know, some people may say that if I want a better experience, I should be willing to buy a more expensive phone. Well, I can say that if I do spend more money on a phone, the chance is very slim that it will be a Nokia.
Really, I am not being nasty, just trying to be honest. I know that Nokia is run by human beings like myself, so I tried to be polite, and I dont intend to hurt anyones feelings.
I am quite willing to answer any questions you may have. Its a bit of a pipe dream, but I would love to see Nokia release an entirely new version of the phone software, with everything re-organized and re-done. The sad truth is that will probably never happen, and I will end up swopping my E63 for something that is easy and pleasant to use.
So please, let me know if there is anything I can do to change the situation, or help you guys in making this better for everyone.

"If your not used to symbian you'll learn it quick enough, compared to android it's childsplay."
On the contrary, Ive been using Nokia phones for the last 8 years, and for the last 2 years Nokia exclusively, and the UI has just got worse.
The fact that "you'll learn it" is of no comfort. Why should I have to spend hours learning how to use a cellphone? Whenever I want to do something I havent done before, I have to spend time digging through obscure menu options. That is absurd.
"What phone do you have in mind to replace the E63, that would help us users alot more then telling us what is wrong with your phone. Although over all I like my e63."
I don't have any specific phone in mind, Im just going to ask and look around until I find something with an operating system that is organised in a logical manner, and that allows me to access and read my email quickly and easily.
I put this here hoping that other users would come forward with UI related things that they were unhappy about, and then Nokia would take notice. I could easily have just gone out and changed phones, but I chose to do something different.

How to implement boolean comparison and event structure?

Hello all,
I'm currently developing an undergraduate lab in which a laptop sends out a voltage via USB-6008 to a circuit board with an op-amp, the voltage is amplified, and then sent back into the laptop. The student is going to have to determine an "unknown" voltage which will run in the background (they can do this by a step test, graph V_in vs V_out and extrapolate to the x-axis).
Currently, I have two loops that are independent in my VI. The first loop is used to "Set the zero." When 0 voltage (V_in) is sent out of the laptop it returns a value around -1.40V (V_out) typically. Thus, I created the first loop to average this value. The second loop, averages the V_out values that come into the laptop as the V_in numeric control changes. Then I take the "set zero" value from the first loop and subtract it from the second loop average to get as close to 0V for V_out when V_in is 0V.
The problem I'm facing is, the event structure waits for the V_in numeric control value change, but after "SET ZERO" is pressed, if there is an unknown value, it will not be added to the average for V_out until V_in is changed by the user. So, I tried implementing a comparison algorithm in the "[0] Timeout Case." This algorithm works for step tests with positive values for V_in, but there are two problems.
1) Negative values cannot be used for V_in
2) If a user uses increasing positive values for V_in there is no trouble, but if they try to go back to 0, the value change event has been called and it is added to V_out as well as the timeout case.
Sorry for the extremely long post, but I've been banging my head over this and can't figure out how to properly implement this. I'm using LabVIEW 8.0.
Attachments:
Average Reset Test.vi ‏371 KB

OK you have bigger problems than Raven's Fan is pointing out.
When the first event loop stops ( after pressing "") (the boolean text is "Set Zero") The second loop may start- (AND PROCESSES all of the events it was registered to process before the loop started!) that would enclude the value change event from "" (The boolean text is Stop) Being pressed bebore the loop started. Of course, since the labels of "Set Zero" and Stop are identical nulls....................................................BOTH event trigger at the same time and are processed as soon as the event loop they are registerd to is available.
Get it ... The two buttons labeled "" both queue Value change events to both loops registered to act on the value change of the control labled ""!
Both loops will do what you programmed in the case of "" Value Change! This can, (as you have observered) lead to confusing code actions.
Do avoid controls with duplicate labels (There is a VI Analizer test for that!) Do avoid multiple event structures in the same hierarchy.
DO NOT bring this to your studients to help you learn LabVIEW! We get enough studii asking embarassing questions
VI Analizer will help you provide sound templates. If you need help with that hit my sig line- e-mail me and I'll always be willing to help.
Jeff

Radio Buttons and Event Structures

Guys, I am hoping for some help here on something I probably should know but cant figure out how to do. I am using a set of radio buttons and an event structure to run a UI. The radio buttons are acting as both a user input and indicator of current status selected by the user. I have the event structure set to run on value change of the radio button. All works great unless the operator decides that he wants to repeat the selection that the radio button is already set to. The event structure does not recognize this as a change of value and will not execute. Is there a way to set the event structure up to run even if the same selection of the radio button is made a second time? I attached a crude example to this post. Basically everytime I press the button, I want an entry made into the text box, regardless if the selected item was already pressed. Any thoughts.
Troy
Attachments:
Radio Button Test.vi ‏16 KB

Instead of hacking around the inherent nature of radio buttons, why not just use a Cluster? Its behavior in the UI is more customizable. Check this out:
I just replaced your radio buttons with a cluster, and added a little logic to the event structure to get it to behave the way you describe. I've attached the VI below, saved in 8.2, if you want to try it out.
Hope this helps,
-D
Message Edited by Darren on 11-07-2006 10:45 AM
Darren Nattinger, CLA
LabVIEW Artisan and Nugget Penman
Attachments:
cluster_buttons.png ‏7 KB
Cluster Test.vi ‏17 KB

Array of buttons and Event Structure

I have a the same problem... I was wondering if you were able to find a soln.
I have literally thousands of buttons and each button as a unique key code and unique name.
For Example :
Button Name    – Key code #
PUSH              - 1
POP                 - 2
PULL               - 3
The way I program this simple algorithm is deadly painful.
1)      I create 1000 buttons by going to control->buttons.
2)      Then I have to go through one by one to change the Boolean text to “PUSH,POP,PULL…”
3)      Then used the Event Structure, adding in all the controls names “PUSH, POP, PULL..” In side each event, I placed the key codes “1,2,3..”
I have search and tried for weeks to find a better way, and still nothing comes to mind.
Please help.

Hi Ben,
Thank you for your reply, I have attached my try_code and a picture of the code I currently have.  I am new in Labview, but I am confident that, there must be a easier way to make an array of buttons with different button name and an associated number to that button.
The 1st file, is to show you want I am currently doing, (painful)
The 2nd file attached is some of the things i tried, I had great hope with my try_code, array_jan25. But I couldn't get it. I think I am very close.
Thanks again,
Attachments:
ThePainfulWay.doc ‏63 KB
Array_jan25.vi ‏42 KB

Queues and event structure

Hi,
I'm having some troubles with a program that I'm doing.
I think the big problem are in the queues and the structure event. I don't know why, but the elements in the queue only increase. And I think, because of that, some things that should happen, don't happen. For example, when I press the button "Impact" an square wave should appear, and it didn't appear.
Thanks for all your help,
Sílvia
Attachments:
Sonic.zip ‏115 KB

sreis wrote:
Hi,
Of course I read your post, but like I said, I don't understand why, when I press the "Impulse" button, I have the start in the queue. I shouldn't have the previous used enum, because of the shift register?
You are right, the things are slow down because of the re-enqueues and the many Acquire state I have in the queue. But how can I fix that?
Thanks,
Sílvia
Yes the previous enum will be there because of the shift register. But look at what is happening in your Acquire loop. You are enqueueing on every loop iteration. So your action is like this:
Press Start - enqueue Initialize and Acquire
Initilaize state runs
Acquire state runs - Acquire is queued up again. Because you have no wait state, this loop runs super fast and you keep dequeueing acquire and queueing acquire again. The wait state is imperative to allow the CPU time to process some other loops. This is not done automatically, you have to allow it by adding waits to loops.
Now you happen to press impact and somehow the CPU finds time to process the event. What state gets queued? Acquire again. Now you have two acquires queued because the last Acquire state also queued an acquire. Every button you press will enque something on top of what is already there and your dequeue will never be able to keep up.
The solution is to not enqueue anything in the bottom loop. Never do this. If you want to acquire when no event is happeing, queue up the acquire in the Timeout event.
- tbob
Inventor of the WORM Global

Datasocket and Event Structure.

I have developed a labview aplication that control some pumps and send the setpoint to some PID controllers in my test laboratory. this aplication run fine with Event Structure. I have developed too, a small labview software that use Datasocket to connect to previous software, the problem is that when i press a control from client (the control change in server) the Event structure don´t detect the "value changed" event, why? How could I find a solution?
Thanks in advance

Hi!
I came up, with this simple solution and it works with my datasocket appz.
Have a nice day!!
Attachments:
prog2.JPG ‏15 KB
fronpanel.JPG ‏11 KB
prog1.JPG ‏16 KB

User Recon and Event Handler Issue in OIM11g

Hello Experts,
I am developing a Trusted Source Recon (User) for a proprietary application. I have devloped two java classes and put them into a jar file and placed the same jar in both ScheduleTask and EventHandlers directory of OIM.
1.UserRecon class -- for User Recon
2. EventHandler class -- update user's password after reconciling the user using API.
In the User Recon class , I have defined a static HashMap variable to store the user and user's password as a class variable. The flow will be like this,
1. Query trsuted source
2. Reconcile user (except password because password can not be)
3. Update the map with userid (key) and password.
4. After reconciling , In the event handler use the same map to read user's password and set the password in OIM using API. (setXelleratePassword).
5.After successful password set remove the entry from Map.
The idea is to avoid making additional call to trusted source to retrieve the user's password.
This worked in OIM9102. But when I tested the same in 11g it did not work.... Though the User Recon class updates the record into the HashMap , In the eventhandler , the HashMap is always coming as "empty".
I did the below changes for 11g,
1. Modified the EventHandler class to suite for 11g.
2. Deployed the event handler as a plugin
3. Placed the same jar file in Schedule task directory of OIM 11g
I am unable find out where it is going wrong?? and tired of troubleshooting. Can you assist me ?
Thanks and Regards
INIYA

INIYA wrote:
Hello Experts,
I am developing a Trusted Source Recon (User) for a proprietary application. I have devloped two java classes and put them into a jar file and placed the same jar in both ScheduleTask and EventHandlers directory of OIM.
1.UserRecon class -- for User Recon
2. EventHandler class -- update user's password after reconciling the user using API.
In the User Recon class , I have defined a static HashMap variable to store the user and user's password as a class variable. The flow will be like this,
1. Query trsuted source
2. Reconcile user (except password because password can not be)
3. Update the map with userid (key) and password.
4. After reconciling , In the event handler use the same map to read user's password and set the password in OIM using API. (setXelleratePassword).
5.After successful password set remove the entry from Map.
The idea is to avoid making additional call to trusted source to retrieve the user's password.
This worked in OIM9102. But when I tested the same in 11g it did not work.... Though the User Recon class updates the record into the HashMap , In the eventhandler , the HashMap is always coming as "empty".
I did the below changes for 11g,
1. Modified the EventHandler class to suite for 11g.
2. Deployed the event handler as a plugin
3. Placed the same jar file in Schedule task directory of OIM 11gHope there are two jars, if this is the case place the secondary jar in lib folder, than make it as plugin.
>
I am unable find out where it is going wrong?? and tired of troubleshooting. Can you assist me ?
put the printstacktrace, so that problem can be identify
sample code
HashMap hash= null;
try{
hash = get map from other class
//class be load at this point, if class is not loaded than you will get exception.
}catch(Throwable t){
// hope you will get some error or exception here
t.printstacktrace();
>
Thanks and Regards
INIYAPaste you sample code, if it couldn't help.

Need user exit and runtime structure to get all the unsaved entries in CAT2

Hi Friends
As per our requirement, we need to compare all the entered hours from CAT2 vs Project level assigned hours.
If user save the record in CAT2, then no problem.I can get the records from CATSDB table.If user enter the record in first week and goes to next week without saving time entries in CAT2, Runtime structure catsd_imp-sumdays in user exit CATS0009 shows the current screen hours only.But I want to get all the entries entered in CAT2(Unsaved records).Is there any other user exit or badi available to get this functionalites.I tried with CATS0006 ,CATS0003 and so many user exits.But still I didnt get any solution for this.
If anyone come accross this kind of scenario,kindly help me on this.
Thanks
Gowrishankar
Edited by: gowrishankar p on Nov 1, 2010 2:24 PM

Hi
I didn;t find any badi for this transaction , check this onc CATS0010 once.
Regards
Siva

Same time Most of the Outlook2k10 users Disconnect and Event 9646 32 MAPI Session event is logged

Exchange 2010 SP3 RU5, windows 2008 R2 SP1 and outlook 2010.
4 CAS Servers behind F5 Load balancer and 7 Mailbox servers are configured in DAG.
issue:
- For most of the users Outlook shows 'Disconnected' state and outlook will not connect even if you restart the client PC.
- All effected users are from different databases and are on different mailbox servers.
- For all the effected users, we are receiving 9646 Event for exceeding 32 MAPI Sessions. And this event is coming on all mailbox servers at same time, So I am thinking it is some common problem, but couldn't figured it out.
-- To fix the issue i have restarted the Exchange servers..
- In RPC Logs, I can see Session drop errors.
============
OUTLOOK.EXE ncacn_ip_tcp 0x6BA (rpc::Exception) 57:16.0 SessionDropped RpcEndPoint: [ServerUnavailableException] Connection must be re-established -> [SessionDeadException] Connection is severed (ConnectionFailedTransientException).
Client needs to reconnect. -> [ConnectionFailedTransientException] Cannot get synchronize buffers from Microsoft.Exchange.Data.Storage.SynchronizerProviderBase. -> [MapiExceptionNetworkError] MapiExceptionNetworkError: Synchronization failure. (hr=0x80040115,
ec=0)
OUTLOOK.EXE ncacn_http 0x6BA (rpc::Exception) 36:08.7 SessionDropped RpcEndPoint: [ServerUnavailableException] Connection must be re-established -> [SessionDeadException] Connection is severed (ConnectionFailedTransientException).
Client needs to reconnect. -> [ConnectionFailedTransientException] Cannot get recipient table. -> [MapiExceptionNetworkError] MapiExceptionNetworkError: Unable to get recipient table. (hr=0x80040115, ec=0)
=========================================
- I found KB http://support.microsoft.com/kb/2781488 for similar issue, but as per the article issue is fixed in SP3 RU1 - But we are running SP3 RU5
Any help regarding this issue will be highly appreciated.
Looking forward to the response,
Regards,
Ghouse

Hi Ghouse,
Please check whether there is a service account for a third-party application that interacts with Exchange server.
Generally, event 4969 occurs because the default maximum number(by default the value is 32) of connections that an account can open was exceeded. There should be a mapi session in your Event description. Please run the following cmdlet to grant the service
account the "View Information Store status" permission:
Add-ADPermission -Identity “Exchange Administrative Group (FYDIBOHF23SPDLT)” -User account_or_group_name
-AccessRights ExtendedRight -ExtendedRights "View information store status" -InhertanceType Descendents
For more information about it, please refer to the following KB:
http://support.microsoft.com/kb/2742012
If it doesn’t work, please check whether the problematic users can access OWA or work well in OWA. Also collect the Test E-mail AutoConfiguration results in Outlook for further anylysis.
Thanks,
Winnie Liang
TechNet Community Support

Run / quit user interface and memory clean up

Hi,
may be I simply miss a coffee, but here is my problem: I have a main function and a quit callback in a standard configuration, i.e. as follows:
int main ( int argc, char *argv [] )
    RunUserInterface ();
    DiscardMenuBar ( popup_menubar_handle );
    DiscardMenuBar ( menubar_handle );
    DiscardPanel ( panel_handle );
    return ( 0 );
int CVICALLBACK Quit ( int panel, int control, int event, void *callbackData, int eventData1, int eventData2 )
    if ( event == EVENT_COMMIT )
        FreeDynamicMemory...
        FileCloseFiles ...
        QuitUserInterface ( 0 );
    return ( 0 );
On the panel, there is a graph with plots etc. If I quit the program via the quit callback (i.e. a Quit button), everything is fine, no memory leak.
If I try a modified version as shown below, plots are not freed after the program has finished, resulting in a memory leak as shown by the resource monitor:
int main ( int argc, char *argv [] )
    RunUserInterface ();
    return ( 0 );
int CVICALLBACK Quit ( int panel, int control, int event, void *callbackData, int eventData1, int eventData2 )
    if ( event == EVENT_COMMIT )
        FreeDynamicMemory...
        FileCloseFiles ...
        DiscardMenuBar ( popup_menubar_handle );
        DiscardMenuBar ( menubar_handle );
        DiscardPanel ( panel_handle );
        QuitUserInterface ( 0 );
    return ( 0 );
I have also tried reversing the order in the quit callback, with the same result:
int CVICALLBACK Quit ( int panel, int control, int event, void *callbackData, int eventData1, int eventData2 )
    if ( event == EVENT_COMMIT )
        FreeDynamicMemory...
        FileCloseFiles ...
        QuitUserInterface ( 0 );
        DiscardMenuBar ( popup_menubar_handle );
        DiscardMenuBar ( menubar_handle );
        DiscardPanel ( panel_handle );
    return ( 0 );
So only the first variant works properly and I could be happy.
However, I do not understand why version 2/3 do not work.
Second, the first variant does not work if I also have installed a main callback, see the other thread here That's why I tried variants 2 and 3 but failed...
Thanks for the coffee

Off the top of my head, I can't think of any reason why the 2/3
alternatives wouldn't work. This might be a bug in the resource tracking
detection.
I tried to reproduce the problem with a simple program, but I wasn't
able to. I'm attaching my files. Can you try them out and see if you can
reproduce the problem with them? If not, can you try attaching some
project that reproduces it, so that I can look into it?
Thanks,
Luis
Attachments:
plotleak.zip ‏5 KB

I using an event structure for an input box and also for hardware input

Hi there!
I have many input boxes and display boxes on my user interface and i have a case for it in the event structure. If the user changes the values of the input boxes, the particular event case writes this new value to the hardware. However, in the timeout section of the event structure, we read the inputs from the hardware and we set the value for its related display object. Within the timeout section i set the value for the related user interface display component, by making use of the property node value.
Furthermore, for most of the display components, i have a event case for a value change associated with it. In that event case, i check for range errors and so forth.
My assumption was that when i read the value in from the hardware and set the property node value, the user interface component's value will change and it would then trigger the event of the value changed for that component.
My assumption worked to the point of updating the component's value, however, it did not trigger the event for the value changed for this component.
Please advise me on how to achieve what i wish to.
Thanks.
Regards,

Local variables are actually much more efficient than property nodes as far as execution speed goes. Take that same VI and delete the property node and instead write to a local variable inside the loop. You'll see the speed nearly identical to writing directly to the indicator. But note that variables also cause a copy of the data to be created. So if you're just using the "Value" property, you're much better off using a local variable. But don't use both at the same time. If you are reading/writing other properties when you update/read the value, you might as well use the value property.
Sometimes you can't get around using either variables or property nodes. Initializing controls to saved values from an ini file is one example. But in this case, they are only written to once as the application starts so the performance hit is negligible.
One thing you can do is store values in a shift register as they are read and pass this through your application so the last value read can be accessed from anywhere. You can even create a cluster that runs through the shift register to hold several values in a single register. This can be expanded a bit farther to create what's called a Functional Global. This basically a subVI with a While loop that only runs once on each call. It has a shift register to store data and a Case structure to Read or Write data into the register. There's a pretty good example that ships with LabVIEW showing how these work. Open the Example finder and search for "Functional Global Communication". This example shows how to use them to pass data between two independent loops, but they work just as well in a single loop application.
One of the big advantages of these Functional Globals over the standard LabVIEW Globals and local variables is that you can add cases to manipulate the data in addition to just storing it. Also, since it's a subVI, access to the data is protected to one caller at a time whereas variables can have multiple reads and/or writes at the same time. This can cause race conditions.
Hope that helps a bit.
Ed
Ed Dickens - Certified LabVIEW Architect - DISTek Integration, Inc. - NI Certified Alliance Partner
Using the Abort button to stop your VI is like using a tree to stop your car. It works, but there may be consequences.

User interfaces and event structures

Similar Messages

Maybe you are looking for