CompositeElement - a bug or a feature?

Hello there,
I've noticed something strange with CompositeElement - unlike proxy element, which sets container to proxied element, this one does not send any container-related notifications to children. This breaks long element trees, like it breaks MASTNew plugin (I'm debugging it right now, trying to make click throughs work - they don't, at least they don't with OpenX VAST). My idea is that CompositeElement should probably send ContainerChangeEvent.CONTAINER_CHANGE to all children when they added, or when element's container changes. This kinda makes sense, but I don't see this even in the recent commits. What do you, guys, think?

Hello Suzi,
provided I understand your question correctly: One way to interest Adobe's development team is to use the "wishform" and maybe your "requesting fixes or changes in future updates of the software"  can serve as a suggestion. Here at https://www.adobe.com/cfusion/mmform/index.cfm?name=wishform, where you will find this: "Use this form to request new features or suggest modifications to existing features."
Hans-Günter

Similar Messages

  • FAQ: How do I submit a bug report or feature request or otherwise give feedback about Premiere Pro?

    The best way to submit a bug report or feature request is to use the bug-report/feature-request form.
    The Premiere Pro team doesn’t  necessarily see and record every post on every forum and social network, but we do see, record, and track every entry  through the the official feature-request/bug-report form.
    It also helps a lot if you opt into the Product Improvement Program.
    Also, be sure to use the crash reporter.
    We really do read all of the bug reports and feature  requests, and the software really does get a lot of benefit from  detailed crash reports.
    Don’t forget about contacting Adobe Technical Support or Customer  Service, too. For information on how to contact Adobe Technical Support,  see this page. (Note that you must register your product before you can open a technical support case for it.)
    If you have tried to get help from our support staff, but the service  was inadequate, I can help you to escalate your issue if you send me  your case number at kopriva at adobe dot com. You must provide me  a case number. I am not offering to solve your problem myself; rather, I  am willing to forward your information to someone if you have already  hit a dead end with our Technical Support or Customer Service.
    If you have feedback about the content of the Premiere Pro Help document,  then please add a comment at the bottom of the relevant page. You can  add comments to add information, to add links, to make corrections, or  to ask for clarification.
    If you want to keep up with the Premiere Pro team, then you can follow our blogs, Facebook page, and Twitter feed and so on.
    Oh, and I can’t resist this opportunity to remind you to update to  the most recent version of the application. Don’t be surprised when the  first thing that you hear back from us is that you need to install the  latest updates. We fix a lot of things in updates, and we don’t want to go chasing bugs that we’ve already fixed. We’ll keep you posted on our blogs, Facebook page, and Twitter feed about updates as they become available.

    Hi qwerty,
    Dun tink that creative customer support sucks cos I had been dealing with them in the past and I can say that their service are quite gd and fast. If you wanted a refund, I tink you need to contact the store that sells you the product. Dun tink creative will refund you though.

  • Lync 2013 - re-connecting is un-muting microphone, is that a bug or a feature?

    Every time Lync 2013 re-connects (very often sadly) the microphone is un-muted.
    Is that a bug or a feature?
    I'm using a Plantronics Voyager UC headset (Lync certified), doesn't know if it happens with others too.
    Please advise if there's a solution, I have to disconnect from Lync every time I'm in a conf call using another software like Cisco Webex.
    Quite annoying and very embarassing sometimes...
    Thanks

    Hi,
    When Lync change the status of Mute/Un-Mute, it change the Audio value with the following path: HKLM\SOFTWARE\Microsoft\Windows\CurrentVersion\MMDevices\Audio
    Please check if Cisco Webex also use the same parameter for Mute\Un-Mute. If the two use the same parameter, it can cause the issue.
    Also, if the re-connection issue only happen to you, please try to update Lync to the latest version and then test again.
    If the re-connection issue happen for multiple users, please check the Lync Server performance and network situation as well.
    Best Regards,
    Eaosn Huang
    Eason Huang
    TechNet Community Support

  • Bug,exploit or feature?

    One thing I have notice is that when I lock my phone by pressing power button the screen turns off but the touch still works for couple of seconds. (e.g. Press power button and swipe from top to bottom and press power button again and you will see that notification panel is active, same goes with camera and call feature in lockscreen). I am not sure if this is a bug or a feature available, but i do believe is when the screen is off the touch shouldn't be active.

    Quester wrote:
    Couldn't replicate this on my Z3 Dual running Lollipop, most probably it's a bug. A PCC phone repair should fix it.I tried repair as well as flashing rom with flash tool but its still there. Its not a big problem though. 

  • Is this a bug or a feature in JDeveloper 11.1.1.4?

    Hello everybody,
    when I try to create a new "Reusable Process" with the composite editor in Jdeveloper, nothing happens on in the composite design view! I can see the process in the file structure and I can also edit the process in the bpm editor, but I can't see it on the composite.
    This looks like a bug to me. Can anybody confirm this?
    Best regards, Chris
    Windows 7 (64 bit)
    JDK 1.6.0_24 (64 bit)
    JDeveloper Build JDEVADF_11.1.1.4.0_GENERIC_101227.1736.5923
    BPMN Editor 11.1.1.4.0.5682
    SOA Composite Editor 11.1.1.4.0.56.82

    it's a feature of cource. all bugs in 11g are features!

  • ServerSocket bind, bug or windows feature

    I wanted to check if a TCP socket was already bound before attempting to use it.
    I had my own ideas, but looked for other ideas as well.
    While testing these other ideas I stumbled on the following issue.
    On Windows XP(SP3), its the work platform of choice, I noted the following from a netstat:
    $ netstat -an|grep 139
    TCP 10.11.13.98:139 0.0.0.0:0 LISTENING
    TCP 192.168.56.1:139 0.0.0.0:0 LISTENING
    TCP 192.168.113.1:139 0.0.0.0:0 LISTENING
    TCP 192.168.231.1:139 0.0.0.0:0 LISTENING
    Note that this is the Windows 'NETBIOS Session Service' of course and that it is bound to all my machines interfaces except for the loopback.
    If I execute the following call:
    ServerSocket anyAddressSocket = new ServerSocket(139, 2, null);
    I will not get java.net.BindException thrown.
    I will see a new listener bound to port 139 and the 'any address' as follows:
    $ netstat -an|grep TCP|grep 139
    TCP 0.0.0.0:139 0.0.0.0:0 LISTENING
    TCP 10.11.13.98:139 0.0.0.0:0 LISTENING
    TCP 192.168.56.1:139 0.0.0.0:0 LISTENING
    TCP 192.168.113.1:139 0.0.0.0:0 LISTENING
    TCP 192.168.231.1:139 0.0.0.0:0 LISTENING
    But if I replicate the situation using Java on a different arbitrary port as follows:
    List<ServerSocket> sockets = new ArrayList<ServerSocket>();
    Enumeration<NetworkInterface> enumOfNetworkInterfaces =
    NetworkInterface.getNetworkInterfaces();
    while(enumOfNetworkInterfaces.hasMoreElements())
    NetworkInterface networkInterface = enumOfNetworkInterfaces.nextElement();
    List<InterfaceAddress> addrs = networkInterface.getInterfaceAddresses();
    for(int idx = 0; idx < addrs.size(); idx++)
    InterfaceAddress intFace = addrs.get(idx);
    if (!networkInterface.isLoopback())
    ServerSocket socket = new ServerSocket(7777, 2, intFace.getAddress());
    sockets.add(socket);
    // At this point, netstat shows the same output as the first port 139 output above.
    ServerSocket brokenSocketNo = new ServerSocket(7777, 2, null);
    I will get the expected exception:
    "java.net.BindException: Address already in use: JVM_Bind"
    I am testing this with Java 6u6 which I know is old, but is what is required to be used for this product.
    Is this a bug in the JVM implementation? I cannot find any reference to such in the bugs DB.
    Is this a feature of Windows TCP stack?
    I have not tried this on a linux platform because I am really trying to understand the issue as exhibited on the MS platform.
    Any insight into this would be helpful.
    Cheers Mark

    Hello EJP.
    Thanks for that pointer.
    I tried it as you indicated and inside the loop using the following:
    ServerSocket socket = new ServerSocket();
    socket.setReuseAddress(true);
    socket.bind(new InetSocketAddress(intFace.getAddress(), 7777), 2);
    Each ServerSocket is now listening on a specific interface using the 'setReuseAddress(true);' call.
    This by itself did not change the effect I see when I then make the call 'new ServerSocket(7777, 2, null)' which succeeded with port 139 but not with my port 7777 test.
    If I changed the call for the for the 'wild card' address to also use 'setReuseAddress(true);' then all succeeds, but this is not the situation that succeeded with the port 139 test.
    I think I am still attempting to solve the issue as a bug or a 'windows XP feature', but I am leaning more about the intricacies of sockets.
    I will follow your suggestion and have a look at 'Stevens' 3rd Edition.
    Cheers Mark

  • Bug or Lost Feature in new Quicktime?

    Not sure if the following is a bug but we have lost an important feature with the new updated release of QuickTime Pro.
    I draw your attention to the three buttons at the top left hand side of a program window - RED Yellow GREEN.
    With the latest QT update in place select any compatible movie or video to play. It opens in original size. Prior to this update you could expand it easily to fit your desktop by clicking on the green button. Now the green button is disabled and does not work upon original opening of a media file.
    Go to the Menu and select "Fit to screen" or hit Control 3 and the video will expand to fit the screen. Hit the green button and it will return to its normal size. Hit the green button again and it will expand to fit the screen.
    Prior to this update once you opened a media file with QuickTime Pro and clicked the green button it would automatically expand to fit the screen. That no longer happens. Apple has either disabled it or it is a missed bug.
    All other OSX media players (RealTime, VLC, NicePlayer, etc) will immediately expand the video window to fit the screen when their green button is selected.
    Apple please fix this right away. Without this feature many who use that green button (especially those of us with 30" Cinema monitors) will avoid QuickTime Pro and opt for one of the other more compatible/feature rich media players.
    Tim...

    Well, for now I'll drop QuickTime Pro and make other video players the default. If I have no alternative I'll convert the MP4 or MOV to another format like MPG or AVI and play it under VLC or RealPlayer.
    I can't be the only user who is going to be doing this - dropping use of QTPro in favor of other more reliable and feature-rich Mac video players.
    Tim...
    3Ghz Mac Pro w/8GRam, 3TeraHD, 2 SupDrives; 13" BlackBook; 80G iPod, 8G iPhone Mac OS X (10.4.9) 30" Cinema, HPColorLJ Printer, LaCie 600Gig External & LaCie DVDRW/CDRW Drives

  • Using TFS and email for bug reports or feature requests

    We've just rolled out a beta of our website and are soliciting feedback from external customers.  We set up a distribution list in Exchange where those emails all come to, but it's not practical.  It's difficult to track, it's non threaded and
    we have different people responding and testing the same items.
    We recently installed TFS 2010 to give it a trial run (nobody has ever used TFS before at all nor is anyone really familiar with it) and I have to imagine that you can use TFS to provide this sort of functionality.  Is there a paper or website out there
    that discussion doing this via TFS?  Basically what I'm looking for is:
    1. User is presented with a webform, fills it out, clicks submit which generates an email sent to us (done)
    1b. An auto response message is sent to the user who submits the form (not done as Exhchange distro wont do this without going the public folder route)
    2. Instead of the email coming to us, it would go to TFS which would auto-open a ticket
    3. All of us would have access to the ticket queue and be able to do the following
    a. ask further questions to the end user, have it sent to them automatically, update the ticket with the emails so we can all follow the thread
    b. able to reassign the ticket to a backlog, or as a bug to a current iteration or simply close it
    I'm an IT guy/project manager rather than a developer but I'm pretty good at following documentation if I can find it for what I'm trying to do.  Thanks for any pointers.

    Ahh, thanks for the extra details and clarification.  The features that you are asking for are definitely not provided out of the box, but are implementable using the TFS SDK.  Inlining a suggestion of one way your developers could
    implement this:
    1. User is presented with a webform, fills it out, clicks submit which generates an email sent to us (done)
    Have the code for the webform make a call to TFS to add a new work item with a special iteration path, e.g. "UnconfirmedCustomerRequests" or a special field like "ActiveCustomerRequest" set to true.  You can add a potentially add customer
    email address field to the work item template to keep track of the email address.  Then send the email out, and include the TFS bug id (and a link to the bug in TFS web access, if desired) in the email.
    1b. An auto response message is sent to the user who submits the form (not done as Exhchange distro wont do this without going the public folder route)
    Why can't the code for your web form take care of that?
    2. Instead of the email coming to us, it would go to TFS which would auto-open a ticket
    Taken care of in step 1
    3. All of us would have access to the ticket queue and be able to do the following
    a. ask further questions to the end user, have it sent to them automatically, update the ticket with the emails so we can all follow the thread
    You can write a web service that listens for notifications from TFS when certain work items are changed.  Updating the work item from customer replies may be trickier without writing another web form that talks to TFS or an email
    parser.
    b. able to reassign the ticket to a backlog, or as a bug to a current iteration or simply close it.

  • JavaFX KeyEvent: a bug or a feature?

    Hello Everyone,
    please take a look at the following toString()-Representation of KeyEvents. I was expecting '$' as text value, because Shift was pressed.
    Unfortunately, KeyTyped is not fired in this case, so I am not able to get the information from this event.
    How can I get the actually typed '$' from KeyPressed?
    Regards,
    YDzh
    KeyEvent [source = BorderPane[id=cnt_borderPane], target = Button@1a3fc21a[styleClass=button]'', eventType = KEY_PRESSED, consumed = false, character =  , text = , code = SHIFT, shiftDown]
    KeyEvent [source = BorderPane[id=cnt_borderPane], target = Button@1a3fc21a[styleClass=button]'', eventType = KEY_PRESSED, consumed = false, character =  , text = 4, code = DIGIT4, shiftDown]
    KeyEvent [source = BorderPane[id=cnt_borderPane], target = Button@1a3fc21a[styleClass=button]'', eventType = KEY_RELEASED, consumed = false, character =  , text = , code = SHIFT]
    KeyEvent [source = BorderPane[id=cnt_borderPane], target = Button@1a3fc21a[styleClass=button]'', eventType = KEY_RELEASED, consumed = false, character =  , text = 4, code = DIGIT4]

    I guess you should be monitoring KEY_TYPED events, not presses or releases.
    I think you have discovered a feature, not a bug, but probably misunderstood the feature, which is, I think, understandable.
    The confusing thing is that the KeyCode represents just the primary unmodified key value, e.g. there is a KeyCode.DOLLAR, however you would actually need a keyboard which produced a $ without any modifiers (like shift) to get that code (and I haven't seen a keyboard like that).  I think the press and release key events are a bit too low level to be generally useful for most tasks and you can pretty much ignore them and just rely on the typed events and the character associated with such typed events.
    If you need further help than this answer, you can provide an mcve that produces your output.
    Here is a small program I wrote:
    import javafx.application.Application;
    import javafx.scene.Scene;
    import javafx.scene.control.TextField;
    import javafx.stage.Stage;
    public class KeyTypeListener extends Application {
        public static void main(String[] args) {
            launch(args);
        @Override public void start(Stage stage) throws Exception {
            System.out.print(
                    "Java Runtime:   " + System.getProperty("java.runtime.version") + "\n" +
                    "JavaFX Version: " + System.getProperty("javafx.version") + "\n" +
                    "OS:             " + System.getProperty("os.name") + "\n"
            TextField field = new TextField();
            field.setOnKeyTyped(System.out::println);
            stage.setScene(new Scene(field));
            stage.show();
    On typing j$4 into the text field, the output of this program is:
    Java Runtime:  1.8.0_40-ea-b10
    JavaFX Version: 8.0.40-ea
    OS:            Windows 7
    KeyEvent [source = TextField@722b0015[styleClass=root text-input text-field], target = TextField@722b0015[styleClass=root text-input text-field], eventType = KEY_TYPED, consumed = false, character = j, text = , code = UNDEFINED]
    KeyEvent [source = TextField@722b0015[styleClass=root text-input text-field], target = TextField@722b0015[styleClass=root text-input text-field], eventType = KEY_TYPED, consumed = false, character = $, text = , code = UNDEFINED, shiftDown]
    KeyEvent [source = TextField@722b0015[styleClass=root text-input text-field], target = TextField@722b0015[styleClass=root text-input text-field], eventType = KEY_TYPED, consumed = false, character = 4, text = , code = UNDEFINED]
    So the character seems to represent the key typed and for $, the character is a $ (at least when I am typing on a US keyboard, though, from the documentation, the keyboard layout shouldn't really matter).
    Here is some info from the KeyEvent doc:
    "Key typed" events are higher-level and generally do not depend on the platform or keyboard layout. They are generated when a Unicode character is entered, and are the preferred way to find out about character input. In the simplest case, a key typed event is produced by a single key press (e.g., 'a'). Often, however, characters are produced by series of key presses (e.g., SHIFT + 'a'), and the mapping from key pressed events to key typed events may be many-to-one or many-to-many. Key releases are not usually necessary to generate a key typed event, but there are some cases where the key typed event is not generated until a key is released (e.g., entering ASCII sequences via the Alt-Numpad method in Windows). No key typed events are generated for keys that don't generate Unicode characters (e.g., action keys, modifier keys, etc.).
    The character variable always contains a valid Unicode character(s) or CHAR_UNDEFINED. Character input is reported by key typed events; key pressed and key released events are not necessarily associated with character input. Therefore, the character variable is guaranteed to be meaningful only for key typed events.
    For key pressed and key released events, the code variable contains the event's key code. For key typed events, the code variable always contains KeyCode.UNDEFINED.
    "Key pressed" and "key released" events are lower-level and depend on the platform and keyboard layout. They are generated whenever a key is pressed or released, and are the only way to find out about keys that don't generate character input (e.g., action keys, modifier keys, etc.). The key being pressed or released is indicated by the code variable, which contains a virtual key code.

  • A bug, or a feature removed from iTunes 11? Every playlist folder shows entire library

    Hi folks - wondering if anyone else has come across this issue (have searched, can't find references...) In older versions of iTunes, if you selected a folder containing playlists, the main window would show all tracks contained in those playlists - for example, if I have a Jazz folder containing 10 different playlists, each with 10 tracks, I could click on the folder and see (and play) just those 100 tracks.
    I just got a new Mac Mini and haved moved all my stuff to the new machine, all working fine with iTunes 11 except now when I click on any of my playlist folders, it shows me every sinlge track in my entire library. If I click on any of the playlists inside the folder, they're fine, just as they should be.
    So, does anyone know - is this a bug, did something get messed up with my transfer from the old machine, or did Apple just remove a handy feature?
    Thanks!

    Well wouldn't ya know it - a few hours later I check again and it's acting like it always used to, now showing the correct number of tracks. I didn't quit iTunes or reboot, relaunch, sleep, or anything - so I guess it was just taking its sweet time (like, 4 hours) to sort itself out after the move!

  • Is this a bug or a feature with audio regions?

    ok so I have an audio part in a project then copy/paste in a later part of the arrangement and after that change the size of the new copied part with the mouse(dragging from the end of it)it changes and messes up the original version as well. I just recently switched to Logic from using Cubase so is this some feature of having the regions linked together somehow or is it just working wrong?

    I found the part that talks about how copy/paste is supposed to work, and it turns out that the manual describes how people want it to work instead of how the app actually works. It says that copy/paste makes a copy (a new region) and the two can be edited independently. Copy/paste is NOT supposed to create a clone.
    Unfortunately, the app doesn't work the way the manual describes. Here's the relevant part from the manual:
    "To copy a region (method 2):
    1 Select the region, then choose Edit > Copy (or use the Copy key command,
    default: Command-C).
    This copies the selected region to the Clipboard.
    2 Set the desired playhead position by clicking on the Bar ruler.
    3 Choose Edit > Paste (Command-V) to paste the contents of the Clipboard into the
    Arrange. The paste will occur on the selected track at the playhead position.
    If several regions are selected, their relative time and track positions are retained. "
    "*Copied regions are actually true, independent replicas of original regions. Changes*
    *made to a parent region will not affect child copies.*
    If you want this type of behavior—where changes to the parent region affect child
    regions—make use of the Loop function (see “Looping Regions” on page 329), or
    create cloned or alias regions (see “Repeating Regions” on page 324).
    *When you copy an audio region, a new region is automatically created in the Audio Bin.*
    *The new region will retain the name of the original, with a sequential number added.*
    As an example, kickingloop.1 will be the name of a region copy, if the original region is
    called kickingloop. A second copy will be named kickingloop.2, a third kickingloop.3,
    and so on."
    Looks like either the manual or the app is wrong. Hopefully apple will actually read the bug reports and fix the app.

  • IOS 6 Bug? VIP feature doesn't work w/ ActiveSync mail

    I noticed that the new VIP feature doesn't work against ActiveSync mail. The feature works as advertised with my Hotmail account, however.
    I am able to select VIP users from within my AS account but their emails are not being filtered into the VIP folder. I do get an audio alert however when their emails come in.
    is this by design or is it a bug?

    I use Active sync on Exchange 2007.
    I have 1 user, my boss, currently set to VIP and it is working ok.

  • Is this a bug or a feature? Failed to access remote jar

    I tried this thing:
    I have this aaa.jar which contains aaa.class and Data.class. In aaa.class, I use URLClassLoader to load some class from a remote jar (http://xxx/xxx/bbb.jar. In bbb.jar there is a class bbb.class and bbb.class has a method to use Data. Note, Data.class in already in aaa.jar.
    I run this aaa.jar without trouble locally, i.e. by
    java -jar aaa.jar.
    However, if I put aaa.jar into my server and write one jnlp file. Then I run aaa.jar from Web start from another machine. I got error:
    NoClassFound Data
    Note, bbb.jar is definitely downloaded from server to another computer. I also signed the jar, signed jnlp. I also set allpermission in jnlp.
    I am quite puzzled by this behavior. Can someone from Sun explain this for us. I personally think it is a bug of web start. However, it could be a hidden feature. If so, what is the way to overcome such difficulties.
    Of course, I can put Data.class into bbb.jar. But, that is not the point. I suppose not to do so since in aaa.jar Data.class is already there.

    Another thing you could do is to set your URLClassLoader as currentThread's contextClassLoader at the begining of your application:
    public static void main(String[] args) {
    ClassLoader cl = YourApplication.class.getClassLoader();
    URLClassLoader ucl = new URLClassLoader(urls, cl);
    try {
    EventQueue eq = Toolkit.getDefaultToolkit().getSystemEventQueue();
    eq.invokeAndWait(new Runnable() {
    public void run() {
         Thread.currentThread().setContextClassLoader(ucl);
    } catch (Exception e) {
    e.printStackTrace();
    masa

  • Annoying bug/ badly implemented feature

    On the torch, you could set the alarm and turn off your mobile, knowing the alarm setting would turn on the mobile and the alarm would go off.
    On the passport this is not possible!
    If you set an alarm and turn it off, it stays off.
    You have no choice but use bedside mode, which does eat battery power.
    Speaking of which, if you are charging your mobile and put into bedside mode (because you have to to get the alarm to go off) this stops the battery charging.
    On the torch, you could set the alarm, shut it off, leave it charging overnight and it didn't stop charging and the alarm went off in the morning.
    Let's see the lovely passport match this simple and elegant feature!

    I searched some more. I didn't find anyone else with the issue.
    So I'm guessing you are on one of the versions of 10.3.1. Maybe you know the first push caused a lot of issues for people that did it both OTA and with Link. Many of the issues if not most are being corrected with subsequent updates. Given the variety and inconsistencies of the issues, it isn't unreasonable to suspect your beside mode charging is related.
    My best ideas is upgrade if you can, reinstall if you can't. Or live with it until the next upgrade. I really do believe it is a software hiccup.
    - Ira

  • UIX radioSet: a bug or a feature ?

    I have a quite simple radio set:
    <radioSet name="radioSet_1">
      <contents>
       <option text="label_1" value="first"/>
       <option text="label_2" value="second"/>
    </contents>
    <primaryClientAction>
       <fireAction formSubmitted="true"/>
    </primaryClientAction>
    </radioSet>It looks like a set of circles and labels near them :)
    If an option is selected it submits a form.
    When I click on a circle it acts as I expect. But when I click on the label the form is submitted but none of the values is selected. So the page reloads but nothing changes.
    I think this isn't a critical bug but such behavior may confuse some of the users.
    PS: I'm one of those users who are confused by that :)

    This was a known bug; I agree it's confusing, and it has since been fixed.

Maybe you are looking for