JNA - problems with pointers

Hi!
At the moment I'm experimenting with the Java Native Access (JNA) library. I have a DLL file written in C and I want to create its wrapper class in Java. Here is the simplified version of the C code:
typedef struct {
} my_struct;
my_struct my_struct_array[ITEM_COUNT];
int my_func(my_struct **items) {
    *items = my_struct_array;
    return 0;
}I don't know how to make the Java representation of the my_func() function. I've tried to translate the argument type to Java in the following ways:
1. my_struct **items -> MyStruct[]
2. my_struct **items -> MyStruct.ByReference[]
3. my_struct **items -> MyStruct.ByValue[]
None of them works...
Thank you for your help!
Edited by: Tomi1226 on Apr 3, 2010 6:38 AM

JNA = java native access
[https://jna.dev.java.net/]
It is a wrapper over JNI intended to make JNI "easier". Myself I doubt that assertion since one would need a firm understanding of the basics of pointers and data types in C/C++ in the first place and then one must also understand how JNA wraps those. And on top of that one must also contend with the often expected gain of JNI in performance versus the dynamic thunking in JNA.
At any rate since it is a wrapper and not part of the Java API it has its own forum.
[https://jna.dev.java.net/#community]
It is there one is most likely to find answers about it. If any help existed on this site it would be most likely to show up in the JNI forum. From my experience it isn't likely.
None of them works...You have a pointer to an array. However that work in JNA.

Similar Messages

  • Problem with JNA+DLL

    hello every one. I want to use a DLL in mi project.To reach this aim i had started to use Java Native Access..
    This is a desctiption of one of the most functions in my DLL:
    long TRANS2QUIK_CONNECT (LPCSTR lpcstrConnectionParamsString, long* pnExtendedErrorCode, LPSTR lpstrErrorMessage, DWORD dwErrorMessageSize)
    Where
    lpcstrConnectionParamsString-string
    pnExtendedErrorCode-pointer on a variable LONG
    lpstrErrorMessage-pointer on a string
    dwErrorMessageSize-long..
    I tried to use this code for connect , but it isdo not work:
    public interface TransToQuik  extends Library{
       public long TRANS2QUIK_CONNECT (String lpcstrConnectionParamsString, long pnExtendedErrorCode, String lpstrErrorMessage, long dwErrorMessageSize);
        private void jButton1ActionPerformed(java.awt.event.ActionEvent evt) {                                        
        String Error;
        long len;
        long FunctionResult=0;
        long pnExtendedErrorCode=0;
        String lpstrErrorMessage="";
        long dwErrorMessageSize=0;
        TransToQuik lib = (TransToQuik) Native.loadLibrary ("D:\\java project\\Robot\\TRANS2QUIK.dll",TransToQuik.class);
        FunctionResult=lib.TRANS2QUIK_CONNECT("D:\\QUIK5.15\\info.exe",pnExtendedErrorCode,lpstrErrorMessage,dwErrorMessageSize);
        }And this error occuredException in thread "AWT-EventQueue-0" java.lang.UnsatisfiedLinkError: Error looking up function 'TRANS2QUIK_CONNECT': Íå íàéäåíà óêàçàííàÿ ïðîöåäóðà.
    at com.sun.jna.Function.<init>(Function.java:179)
    at com.sun.jna.NativeLibrary.getFunction(NativeLibrar y.java:345)
    at com.sun.jna.NativeLibrary.getFunction(NativeLibrar y.java:325)
    at com.sun.jna.Library$Handler.invoke(Library.java:20 3)
    at $Proxy0.TRANS2QUIK_CONNECT(Unknown Source)
    at robot.Robot.jButton1ActionPerformed(Robot.java:106 )
    at robot.Robot.access$000(Robot.java:8)
    at robot.Robot$1.actionPerformed(Robot.java:32)
    at javax.swing.AbstractButton.fireActionPerformed(Abs tractButton.java:2015)
    at javax.swing.AbstractButton$Handler.actionPerformed (AbstractButton.java:2338)
    at javax.swing.DefaultButtonModel.fireActionPerformed (DefaultButtonModel.java:402)
    at javax.swing.DefaultButtonModel.setPressed(DefaultB uttonModel.java:259)
    at javax.swing.plaf.basic.BasicButtonListener.mouseRe leased(BasicButtonListener.java:253)
    at java.awt.Component.processMouseEvent(Component.jav a:6279)
    at javax.swing.JComponent.processMouseEvent(JComponen t.java:3311)
    at java.awt.Component.processEvent(Component.java:604 4)
    at java.awt.Container.processEvent(Container.java:208 4)
    at java.awt.Component.dispatchEventImpl(Component.jav a:4639)
    at java.awt.Container.dispatchEventImpl(Container.jav a:2142)
    at java.awt.Component.dispatchEvent(Component.java:44 65)
    at java.awt.LightweightDispatcher.retargetMouseEvent( Container.java:4664)
    at java.awt.LightweightDispatcher.processMouseEvent(C ontainer.java:4327)
    at java.awt.LightweightDispatcher.dispatchEvent(Conta iner.java:4257)
    at java.awt.Container.dispatchEventImpl(Container.jav a:2128)
    at java.awt.Window.dispatchEventImpl(Window.java:2612 )
    at java.awt.Component.dispatchEvent(Component.java:44 65)
    at java.awt.EventQueue.dispatchEvent(EventQueue.java: 618)
    at java.awt.EventDispatchThread.pumpOneEventForFilter s(EventDispatchThread.java:286)
    at java.awt.EventDispatchThread.pumpEventsForFilter(E ventDispatchThread.java:201)
    at java.awt.EventDispatchThread.pumpEventsForHierarch y(EventDispatchThread.java:191)
    at java.awt.EventDispatchThread.pumpEvents(EventDispa tchThread.java:186)
    at java.awt.EventDispatchThread.pumpEvents(EventDispa tchThread.java:178)
    at java.awt.EventDispatchThread.run(EventDispatchThre ad.java:139)Thank to all for help!
    From Russia with Love!

    jschell wrote:
    qpile wrote:
    But no pointers are exist in JAVA?! GIve me some example please for this instantMany C APIs deal with pointers. And JNI and thus JNA deal with them. The fact that java doesn't do pointers doesn't mean that you get to ignore that when you use JNA. There must be some difference between an API that takes a pointer and one that doesn't.
    And the JNA docs and forums are sure to address it.
    >
    >But no pointers are exist in JAVA?! GIve me some example please for this instant
    Many C APIs deal with pointers. And JNI and thus JNA deal with them. The fact that java doesn't do pointers doesn't mean that you get to ignore that when you use JNA. There must be some difference between an API that takes a pointer and one that doesn't.
    And the JNA docs and forums are sure to address it.
    Firts problem that the instruction for this DLL was created for C ++ or C SHARP. And those languages use different name of function... The real name of function is TRANS2QUIKCONNECT@16...
    And ishould to write this
    public int _TRANS2QUIK_CONNECT@16(String lpcstrConnectionParamsString, int pnExtendedErrorCode, String lpstrErrorMessage, int dwErrorMessageSize);
        private void jButton1ActionPerformed(java.awt.event.ActionEvent evt) {                                        
        String Error;
        int len;
        int FunctionResult=0;
        int pnExtendedErrorCode=0;
        String lpstrErrorMessage="";
        int dwErrorMessageSize=0;
        TransToQuik lib = (TransToQuik) Native.loadLibrary ("D:\\java project\\Robot\\TRANS2QUIK.dll",TransToQuik.class);
       // FunctionResult=lib._TRANS2QUIK_CONNECT@16("D:\\QUIK5.15\\",pnExtendedErrorCode,lpstrErrorMessage,dwErrorMessageSize);
       // jTextArea1.setText(long(FunctionResult));
        }   But another problem appear
    :run:
    Exception in thread "AWT-EventQueue-0" java.lang.ClassFormatError: Duplicate field name&signature in class file robot/Robot$TransToQuik
            at java.lang.ClassLoader.defineClass1(Native Method)
            at java.lang.ClassLoader.defineClass(ClassLoader.java:764)
            at java.security.SecureClassLoader.defineClass(SecureClassLoader.java:144)
            at java.net.URLClassLoader.defineClass(URLClassLoader.java:382)
            at java.net.URLClassLoader.access$100(URLClassLoader.java:75)
            at java.net.URLClassLoader$1.run(URLClassLoader.java:294)
            at java.net.URLClassLoader$1.run(URLClassLoader.java:288)
            at java.security.AccessController.doPrivileged(Native Method)
            at java.net.URLClassLoader.findClass(URLClassLoader.java:287)
            at java.lang.ClassLoader.loadClass(ClassLoader.java:399)
            at sun.misc.Launcher$AppClassLoader.loadClass(Launcher.java:334)
            at java.lang.ClassLoader.loadClass(ClassLoader.java:332)
            at robot.Robot.jButton1ActionPerformed(Robot.java:100)
            at robot.Robot.access$000(Robot.java:8)
            at robot.Robot$1.actionPerformed(Robot.java:36)
            at javax.swing.AbstractButton.fireActionPerformed(AbstractButton.java:2015)
            at javax.swing.AbstractButton$Handler.actionPerformed(AbstractButton.java:2338)
            at javax.swing.DefaultButtonModel.fireActionPerformed(DefaultButtonModel.java:402)
            at javax.swing.DefaultButtonModel.setPressed(DefaultButtonModel.java:259)
            at javax.swing.plaf.basic.BasicButtonListener.mouseReleased(BasicButtonListener.java:253)
            at java.awt.Component.processMouseEvent(Component.java:6279)
            at javax.swing.JComponent.processMouseEvent(JComponent.java:3311)
            at java.awt.Component.processEvent(Component.java:6044)
            at java.awt.Container.processEvent(Container.java:2084)
            at java.awt.Component.dispatchEventImpl(Component.java:4639)
            at java.awt.Container.dispatchEventImpl(Container.java:2142)
            at java.awt.Component.dispatchEvent(Component.java:4465)
            at java.awt.LightweightDispatcher.retargetMouseEvent(Container.java:4664)
            at java.awt.LightweightDispatcher.processMouseEvent(Container.java:4327)
            at java.awt.LightweightDispatcher.dispatchEvent(Container.java:4257)
            at java.awt.Container.dispatchEventImpl(Container.java:2128)
            at java.awt.Window.dispatchEventImpl(Window.java:2612)
            at java.awt.Component.dispatchEvent(Component.java:4465)
            at java.awt.EventQueue.dispatchEvent(EventQueue.java:618)
            at java.awt.EventDispatchThread.pumpOneEventForFilters(EventDispatchThread.java:286)
            at java.awt.EventDispatchThread.pumpEventsForFilter(EventDispatchThread.java:201)
            at java.awt.EventDispatchThread.pumpEventsForHierarchy(EventDispatchThread.java:191)
            at java.awt.EventDispatchThread.pumpEvents(EventDispatchThread.java:186)
            at java.awt.EventDispatchThread.pumpEvents(EventDispatchThread.java:178)
            at java.awt.EventDispatchThread.run(EventDispatchThread.java:139)
    BUILD SUCCESSFUL (total time: 5 seconds)Edited by: qpile on Jan 28, 2010 10:28 PM

  • Memory problems with PreparedStatements

    Driver: 9.0.1 JDBC Thin
    I am having memory problems using "PreparedStatement" via jdbc.
    After profiling our application, we found that a large number oracle.jdbc.ttc7.TTCItem objects were being created, but not released, even though we were "closing" the ResultSets of a prepared statements.
    Tracing through the application, it appears that most of these TTCItem objects are created when the statement is executed (not when prepared), therefore I would have assumed that they would be released when the ResultSet is close, but this does not seem to be the case.
    We tend to have a large number of PreparedStatement objects in use (over 100, most with closed ResultSets) and find that our application is using huge amounts of memory when compared to using the same code, but closing the PreparedStatement at the same time as closing the ResultSet.
    Has anyone else found similar problems? If so, does anyone have a work-around or know if this is something that Oracle is looking at fixing?
    Thanks
    Bruce Crosgrove

    From your mail, it is not very clear:
    a) whether your session is an HTTPSession or an application defined
    session.
    b) What is meant by saying: JSP/Servlet is growing.
    However, some pointers:
    a) Are there any timeouts associated with session.
    b) Try to profile your code to see what is causing the memory leak.
    c) Are there references to stale data in your application code.
    Marilla Bax wrote:
    hi,
    we have some memory - problems with the WebLogic Application Server
    4.5.1 on Sun Solaris
    In our Customer Projects we are working with EJB's. for each customer
    transaction we create a session to the weblogic application server.
    now there are some urgent problems with the java process on the server.
    for each session there were allocated 200 - 500 kb memory, within a day
    the JSP process on our server is growing for each session and don't
    reallocate the reserved memory for the old session. as a work around we
    now restart the server every night.
    How can we solve this problem ?? Is it a problem with the operating
    system or the application server or the EJB's ?? Do you have problems
    like this before ?
    greetings from germany,

  • Is there a known problem with USB microphones?

    Background
    OS X 10.5.8
    Premiere Elements 9
    Hardware Intel iMac
    I use a USB mic with all my audio applications.
    However, it appears that PrE doesn't see the mic for narration.  I've set up the audio hardware preferences to use the Default System Input/Output but no recognition of the mic.
    The mic works fine on all my other applications and is selected in system preferences to be the input of choice, and tests indicate it is recognized.
    I changed the audio hardware selection in PrE to use the built in mic and it works as expected.
    So I'm led to wonder if there is a problem with using a USB mic?
    The documentation, referencing windows only, mentions plugging in a microphone to the microphone socket on the PC, not mention of a USB mic?
    Any pointers to where I might find a solution?
    Thanks

    Thanks Neale, in particular for posting the images:
    nealeh wrote:
    To use my USB microphone (built in to my webcam) I go to Preferences> Audio Hardware and select "Premiere Elements Windows Sound" as the Default Device.
    Then click ASIO settings and ensure your microphone is selected.
    I believe I'm making a similar selection
    I've tried all the selections. The built in system works but my external mic, when selected in system preferences as the default, is not recognized.

  • Problem with SMS_NOTIFICATION_SERVER

    Hallo Guys and gals
    I am having some problems with SMS_NOTIFICATION_SERVER, and i would like to know if anyone has any pointers onto where to look for a solution.
    This problem isn't one i have seen before.
    What i have done so far in regards to troubleshooting:
    If i look into the Monitoring log in "System Status" - "Component status", i get these errors:
    Notification Server on "Servername" failed to initialize. The operating system reported error 2147943458: The service cannot be started, either because it is disabled or because it has no enabled devices associated with it.
    Possible cause: Failed to initialize with error code -2147023838.
    To help identify the problem, see the bgbserver.log on the management point "Servername".
    the error: -2147023838 Means:
    The service cannot be started, either because it is disabled or because it has no enabled devices associated with it.
    If i look into the "bgbserver.log" i get these errors:
    Error: Failed to create BgbServerController instance 80070422 SMS_NOTIFICATION_SERVER 1/14/2014 1:16:44 PM 4436 (0x1154)
    Failed to initialize BGB server. Wait 300 seconds to do initialization again SMS_NOTIFICATION_SERVER 1/14/2014 1:16:44 PM 4436 (0x1154)
    It doesn't look like i get any errors in the bgbmgr.log, this log has:
    Wait for events timed out after 60 seconds. SMS_NOTIFICATION_MANAGER 1/14/2014 1:21:51 PM 4516 (0x11A4)
    BgbManager is waiting for file and registry change notification or timeout after 60 seconds SMS_NOTIFICATION_MANAGER 1/14/2014 1:21:51 PM 4516 (0x11A4)
    in the logfile so far.
    if i look at the bgbsetup.log in the logfiles on the MP i don't see anything than, installation was successful, also Successfully configured BGB server applikation.
    I have no clue onto where to start my troubleshooting, also not if i can disregard this problem, as to my knowledge the BGB Server is related to push of the native policies functionality which was added in SP1.
    I am looking forward hearing from you Guys.
    Kind Regards
    Morten Leth

    Hi,
    I have the exact same problem as you Morten. I have searched for an answear, but come up empty. Anyone got any idea of what's wrong would be much appreciated. 
    Did you ever manage to solve the problem Morten?
    The log in the console says this:
    "Notification Server on xxx-sccm.xx.local failed to initialize. The operating system reported error 2147943458: The service cannot be started, either because it is disabled or because it has no enabled devices associated with it. 
    Possible cause: Failed to initialize with error code -2147023838.
    To help identify the problem, see the bgbserver.log on the management point hfs-sccm.hf.local."
    And when looking in the bgbserver.log, the following shows -  over and over again:
    "Error: Failed to create BgbServerController instance 80070422
    Failed to initialize BGB server. Wait 300 seconds to do initialization again"
    This is on a newly installed SCCM 2012 R2, on a Server 2012 R2.
    Kind Regards
    Mathias

  • Mac Pro (First Generation) problem with 1TB Seagate Barracuda

    Hi Macianer,
    I have a first generation Mac Pro which was sold with max. 750GB Hard drives for each bay. At this time I am grabbing for straws and maybe someone has any info on this problem.
    I purchased 4 x 1TB Seagate Barracuda hard drives ( Model: ST31000340AS / Firmware: SD15 ), did a regular format after I purchased them ( GUID Partition Table, Journaled HFS+ File System). I am using 2 of them internally (they get quite hot and i didn't want to put all 4 of them in there) and two of them externally (backup purposes). The other two internal bays are covered by 2x 500GB Barracuda drives.
    I put all four 1 TB hard drived through a 1 week stress test to make sure I am not dealing with bad apples. So far this practice served me well.
    Now 2 weeks later one completely crashed (internal boot drive). It sounded to spin up alright but was not accessible at all. I went to BestBuy and GeekSquad said this drive is crashed and I returned it.
    Since I had a backup I didn't worry too much. I put the back up hard drive in as the new boot drive (since it was a SuperDuper duplicat) it worked like a charm. Now the second hard drive starts temporarily freezing the system for 2-4 minutes and seems to get stuck showing the great colored ball of doom. After the freeze it continues to operate normal again and I am able to continue my work.
    I ran the Disk Utility and I got the following message:
    INVALID KEY LENGTH
    VOLUME CHECK FAILED.
    ERROR: FILESYSTEM VERIFY OR REPAIR FAILED.
    It just seems weird, I never had problems with hard drives and now I get these problems with 3-4 week old hard drives.
    I was in a chat with seagate support. First question to trouble shoot: Do you have access to a PC. My answer was no since Microsoft Vista I donated all of them to the Salvation Army. He asked if I formatted the hard drives low level and MBR with HFS+. I said since I do not have access to PC nor do I plan on using my hard drives for PC I formatted it regular with GUID and HFS+. I asked if there is any history of these drives that would help explain my experience. I was hoping for a recall or something of that sort or something that a firmware upgrade would remedy.
    I guess my question is, is the first generation Mac Pro, even though it seems to format and run them fine, not able to handle 1TB hard drives on a daily basis. Or am I dealing with hard drive problems? The Seagate guy didn't seem to be of any help.
    Any help is appreciated or even pointers that would help me in finding the problem.
    Cheers,
    PixelGrinch
    Message was edited by: PixelGrinch

    Excellent questions, I too am looking at SSD.
    Some things I know about SSD:
    1. Intel recently released new firmware for their SSD drives to improve long term performance including Trim/Garbage collection.
    2. SSD drives can be setup as SLC or MLC - SLC is faster read 2X, write 4X over MLC, however SLC is smaller capacity. SLC also has a longer life span.
    3. Intel plans to have an 320GB X25-M (MLC) and a 128GB X25-E (SLC) out soon.
    OSX does not fully support Trim and I'm not aware of any 3rd party software for OSX that helps in the garbage collection area. Diskeeper 2010 for Windows 7 provides full support for all variants of SSD, but nothing on the Mac side.
    I'm probably going to wait for 2010 to see what comes out, SSD appears to be good news, but it still also seems to have issues. Given the cost of these drivers and the frequent firmware updates, a wait and see is my choice.
    I've thought about going with a SAS RAID controller and 15K RAID drivers, but the cost/performance ratio comes out about the same and the 15K drives are noisey and power hungry but they're a "known" entity.
    Rob

  • Problem with FIle Adapter

    Hi
    We r facing typical problem with file adapter.
    Lets say there are 30 files to be processed  by placing an copy in arichive folder by file adapter  based on pooling 
    parameters.
    All the 30  files are picked up and we can see the copies in archive folder.
    But when we go to RWB and check the audit log,
    there we could see only 25 files , the 5 files are missing.
    no trace found therre about the 5 files.
    This is happening only some times and not to all the files placed in the FTP. Later when the same file is placed back from the Archive folder to the Source folder, the file is being picked and processed succesfully.
    unfortunately this is happening on Production server
    Any pointers which could trouble shoot this issue.
    Regards
    shekar chandra

    Hi,
    Couple of thing which you can try....
    1. Use EOIO QoS and check if you still face the issue.
    Is some program or application writing files to the directory? If yes I would recommend the use of a temporary name to write the files and then rename them.
    Sometimes when an external application connects via FTP and transfers the files...
    The file could be picked up by the adapter poll even before the transfer of data is done.
    Which is why you will find that the file is archived but there is no xml message created for the same in the Integration engine.
    Try increasing the poll interval. This helps sometimes.
    Regards,
    Sidharth

  • HR Programming :  Problem with IDOC HRMD_A

    Hello ,
    We use Idoc HRMD_A for transferring PA records from
    one R3 system (ECC5) to another R3 system (46d).
    We have a problem - The Idoc doesn't check the data
    inserted against the database check tables and
    inserts wrong values into the target system database.
    (At least that what has happened when we transferred
    wrong value of MASSN in 0000 records).
    Is there something we can do ?
    Maybe - Some flag we can sign in order a check against check tables will be done ?
    Is it a problem with this IDOC ?
    Thanks .
    I appreciate any help.
    Nitzan S.

    Consultancy note 134085 :
    "Ultimately you are responsible for consistency of this data. SAP saves this data directly on the database without checking the single fields and consistency of the data records (time constraints check, writing change pointers, value check) because the consistency for an R/3 - R/3 lingage is of course ensured."
    The IDOCs were intended for R3->R3 data replication, and not for data loads/interfacing. Since you're having problems, I'm guessing in addition they require the same SAP version as well.

  • Problem with Table display in Adobe Form of Web Dynpro ABAP

    Hi Team,
    I'm trying to display a table in Adobe Print Form (not interactive). The table is bound to the context node in the Web Dynpro ABAP where the cardinality is 0:N. I get a short dump. Don't know what to do here. I'm able to display individual fields from my context successfully. Looks like something is missing when I try to display a table. Any pointers to display a data table in Adobe? Any step by step example you know of where I can find out what I did wrong? Many thanks for your time
    We are on support pack SP06 on 2004S release.
    DUMP INFORMATION:
    Short text                                                               
        Access via 'NULL' object reference not possible.                 
    What happened?                                                           
        Error in the ABAP Application Program  The current ABAP program "CL_WD_ADOBE_SERVICES==========CP" had to be terminated because it has come across a statement that unfortunately cannot be executed.
    An exception occurred that is explained in detail below.                    
    The exception, which is assigned to class 'CX_SY_REF_IS_INITIAL', was not caught in procedure "CREATE_PDF" "(METHOD)", nor was it propagated by a RAISING clause.

    Hi Thomas,
    We upgraded our system(2004S) to SP10 so that we can use the ADOBE form printing in Web Dynpro ABAP but could not apply note 1034425 because of the validity constraints in SNOTE. All the pre-requisite notes are already in the system apart from one note 1029721 which talks about supporting input helps. We don't want to use input helps in our system and cannot upgrade to SP11 because of problems with Business functionality.
    Questions:
    1) Is note 1034425 absolutely dependent on 1029721?
    2) If NOT, can someone please correct the validity so that we can implement using SNOTE.
    3) Would it be OK if we manually apply the changes specified in 1034425 without implementing 1029721?
    4) If all of the above is not applicable, is there any other note which fixes the above mentioned short dump.
    Your help in this matter will greatly help us in using the ADOBE services provided by SAP WDA.
    Thanks much

  • Bridge 6 will not open common file types - e.g., jpeg, but CS6 has no problem with them

    I had had CS5.1 and CS6 on this PC. I uninstalled CS5.1; now, I can open common file types (eg.g, *.jpf, *.bmp) directly from CS6. However, Bridge 6 triggers an error message to the effect that "Windows cannot find (....\CS5.1) - in other words, Windows still wants to associate file types in Bridge *^* to CS 5.1 - not to CS6. As a result, bridge is effectively dead in the water.

    Yet another case of what happens when you uninstall an old version with a newer one present.
    The old version takes a lot of registry items with it on the way out, leaving empty pointers to itself rather than the new version still present. It's as if the old version uninstaller can't imagine the possibility of anything newer, so it just says to the OS "Photoshop is no longer installed". It's not just file associations, Lightroom users have reported similar problems with "Edit in Photoshop".
    Sometimes these pointers can be easily re-established, but more often a full uninstall/reinstall is needed, maybe with a run of the CC cleaner tool tossed in for good measure.
    This happens all the time, posts like this trickle through the forum at a steady state. I can't imagine why this has never been acknowledged, and why Adobe officials keep saying that it's perfectly safe to uninstall previous versions. It's not. Yes, it could well be that the bug, or whatever you prefer to call it, is ultimately in the Windows registry. But this happens with such solid regularity that some official recognition would be in order.
    The safe procedure is to always uninstall in strict reverse version order, then reinstall.

  • Downloading a trial - problem with the Adobe Download Assistant

    Q: What is the Adobe Download Assistant?
    A: Trial versions of Adobe Creative Suite, Photoshop Elements, and Adobe Premiere Elements are downloaded using the Adobe Download Assistant — a small application that is meant to improve download reliability and avoid issues that commonly occur during large downloads, such as frequent disconnects. After the trial product is downloaded, Adobe Download Assistant should automatically start installing the product.
    Q: I installed the Adobe Download Assistant, but my trial didn't download?
    A: Under certain conditions, the Adobe Download Assistant installs but does not immediately start downloading the trial product. If this happens:
    1. Quit the Adobe Download Assistant.
    2. In your browser, return to the product trial download page on Adobe.com. Refresh the page, then click the Download now button again and follow the prompts.
    Important: When you start the download, be sure to select your User folder or Desktop folder. A common cause of errors are due to attempts to download the Adobe Download Assistant to the Program Files/Applications folder. The Adobe Download Assistant only downloads the trial application files. Once the download is complete it will  launch the application installer; when the installer  is launched you can then choose to install the actual trial application into the Program Files/Application folder.
    For more info, see
    Troubleshoot Adobe Download Assistant
    or
    Adobe Download Assistant FAQ
    Q: I've have a problem on Windows 7. The download manager just sits there, and does nothing, even after I navigate back to the trial page. It turns out that you can't install Flash in the Programs folder. You have to pick a different folder for ADA to actually work?
    A: You are correct, the Adobe Download Assistant does not have sufficient permission to save files to the Program Files folder. If you run the Adobe Download Assistant as an administrator, or save to a different location (for example, in your Users folder), this should resolve the issue.
    Q: When I try and download the Creative Suite Design Premium the Download Assistant opens and I log in. On the top it says "Welcome, User" but on the bottom right below the action bar for the download it says "Sign in ith your Adobe ID to continue" and the action/download bar just sits there. I have quit and restarted numerous times and even left it running all night. Nothing is working though. I have Windows 7 Home Premium and see no error messages.
    A: Try opening your browser by right-clicking in the start-programs menu and select "Run as Administrator".
    Q: I am trying to download the Premiere trial version onto Windows. I had no issues installing Adobe Download Assistant, however when I tried to log in, it came up with error 100, claiming it could not communicate with the internet.
    A: Many of the errors which forum users are experiencing with the download process is due to trying to download to the Program Files/Applications folder. This seems to cause Error 100, 101, 107, and even problems with being able to extract the install files after the download completes. The Adobe Download Assistant is only downloading the install files.  Once the download is complete it will then launch the installer.  Once the installer itself is launched you can then install to the Program Files/Application folder.
    Q: I've tried everything and nothing works. I'm really frustrated and I don't know what else to do. How can I download a trial?
    A: Another way to get trials of Adobe software is to sign-up for a Free Creative Cloud Membership. With Creative Cloud Membership you can download full, downloadable 30-day trials of all our CS6 Master Collection straight to your desktop, and store up to 2GB of files. Alternately, you can review http://forums.adobe.com/thread/981369 which provides information on how to obtain direct downloads.

    I have suddenly become computer illiterate... that's the only conclusion I can draw from my total failure to install the Premiere Pro CS6 trial version for Mac.
    Step1: I downloaded the Adobe Downloader Assistant installer and installed ADA in the default location (/Applications), which then was followed by said ADA launch. I logged in, selected PP and started a 2 hrs download of a 1.7 GB file.
    Step2: I come back after a while and find a nice DVD icon on my Desktop, reading PP CS6: Great! I quit ADA and open that folder, to find another folder in it, with 3 subfolders and an "Install" application. Click on it. Nada. Reclick on it. Same + an error -1704 "can't open application".
    Step3-30: I am cutting a long story short, but after reading all the support threads I could find, trying to reinstall ADA on my desktop, relog on the website, restart and cancel PP CS6 download a few times (without ADA noticing that there was already a 1.7 GB corresponding to that same file), and repeatedly failed to launch the install.app, I decided to act stupidely and remove all files and go to:
    Step 31: reinstall ADA, re-download PP CS6.
    Step 32: I waited after the install, nothing happened so I went to the new PP CS6 "DVD" and found the "install" app, which reliably did not start anything.
    Step 33: I rebooted. After the reboot, the DVD was gone, replaced by an empty folder of the same name...
    Step 34: me typing all this... I am using a 2011 iMac running Lion and Creative Suite CS 5.5. Thanks for any pointers...

  • Problem with reading PNG metadata due to unicode strings

    Being new to the XMP SDK I have problems trying to dump the XMP data from a .png file because the data contains UTF-8 data.
    I have searched the forum for answers, but the answers I found do not help.
    The answer I found was to replace std::string with std::wstring. but that causes problems with the compile - btw, I am using MSVC 2010 C++ Express
    and XMP SDK-CC201306
    Changing the std:: line to:
    #define TXMP_STRING_TYPE std::wstring
    and all string definitions in my code to wstring, gives me a bunch of compile errors, such as:
    f:\pkg\c++\xmp-toolkit-sdk-cc201306\public\include\client-glue/TXMPMeta.incl_cpp(74): error C2664: 'std::basic_string<_Elem,_Traits,_Ax> &std::basic_string<_Elem,_Traits,_Ax>::assign(const _Elem *,unsigned int)' : cannot convert parameter 1 from 'XMP_StringPtr' to 'const wchar_t *'
    1>          with
    1>          [
    1>              _Elem=wchar_t,
    1>              _Traits=std::char_traits<wchar_t>,
    1>              _Ax=std::allocator<wchar_t>
    1>          ]
    1>          Types pointed to are unrelated; conversion requires reinterpret_cast, C-style cast or function-style cast
    1>          f:\pkg\c++\xmp-toolkit-sdk-cc201306\public\include\client-glue/TXMPMeta.incl_cpp(72) : while compiling class template member function 'void TXMPMeta<tStringObj>::SetClientString(void *,XMP_StringPtr,XMP_StringLen)'
    1>          with
    1>          [
    1>              tStringObj=std::wstring
    1>          ]
    1>          f:\pkg\c++\XMP-Toolkit-SDK-cc201306\public\include\XMP.incl_cpp(55) : see reference to class template instantiation 'TXMPMeta<tStringObj>' being compiled
    1>          with
    1>          [
    1>              tStringObj=std::wstring
    1>          ]
    Any help or pointers will be most welcome.
    TIA,
    DayWalker

    OK, I now understand the point you were driving at.
    To check out this idea, I have modified the code in the routine which dumps the output from the callback to a file as follows:
    (Though if you were to dump the file using the std:: stream - .i.e. using the utility as supplied, you should see the same problem - it is not at all dependent on my using the wxWidgets toolkit)
    XMP_Status DumpXMPToFile(void * WXUNUSED(refCon), XMP_StringPtr buffer, XMP_StringLen bufferSize )
        XMP_Status status = 0;
        wxString wsT;
        try
        wxString wsText( buffer, wxConvUTF8 );
        if( g_wfDumpFile.IsOpened() )
          wsT.Printf( _T("<%s> size(%d)"), wsText, bufferSize );
          g_wfDumpFile.SeekEnd();
          g_wfDumpFile.Write( wsT, wsT.Length() );
        else
          return -1;
        catch(XMP_Error & e)
    The output for the one line in question:
    <Iptc4xmpCore:CiAdrExtadr> size(24)< = "> size(4)<Leo-Saurer-Straße> size(15)<> size(1)<C3> size(2)<> size(1)<9F> size(2)<>> size(1)<e> size(1)<"> size(1)<
    > size(1)<   > size(3)<   > size(3)<   > size(3)
    The hex contents of the buffer I get in the callback is
    0x003908D0  4c 65 6f 2d 53 61 75 72 65 72 2d 53 74 72 61 c3 9f 65 00 cd cd cd cd  Leo-Saurer-Straße.ÍÍÍÍ
    0x003908E7  cd cd cd cd cd cd cd cd cd fd fd fd fd ab ab ab ab ab ab ab ab ee fe  ÍÍÍÍÍÍÍÍÍýýýý««««««««îþ
    As you can see, my conversion routines takes all the buffer contains and converts it to the appropriate UTF-8 string, even though the count passed to me is short and represents only the count up to the first non-ASCII character and all of the subsequent strings  - interpreted as hex values by the caller of the call back would not give me the correct characters, athough they are passed to me for display.
    The two fragments of one byte length are the ones I could not find in the actual output.
    The call stack at that point:
    >
    wxMeta.exe!DumpXMPToFile(void * __formal, const char * buffer, unsigned long bufferSize)  Line 110
    C++
    wxMeta.exe!TextOutputProcWrapper(void * refCon, const char * buffer, unsigned long bufferSize)  Line 60 + 0x18 bytes
    C++
    wxMeta.exe!DumpClearString(const std::basic_string<char,std::char_traits<char>,std::allocator<char> > & value, long (void *, const char *, unsigned long)* outProc, void * refCon)  Line 409 + 0x1c bytes
    C++
    wxMeta.exe!DumpPropertyTree(const XMP_Node * currNode, int indent, unsigned int itemIndex, long (void *, const char *, unsigned long)* outProc, void * refCon)  Line 161 + 0x14 bytes
    C++
    wxMeta.exe!DumpPropertyTree(const XMP_Node * currNode, int indent, unsigned int itemIndex, long (void *, const char *, unsigned long)* outProc, void * refCon)  Line 222 + 0x1f bytes
    C++
    wxMeta.exe!XMPMeta::DumpObject(long (void *, const char *, unsigned long)* outProc, void * refCon)  Line 1100 + 0x23 bytes
    C++
    wxMeta.exe!WXMPMeta_DumpObject_1(__XMPMeta__ * xmpObjRef, long (void *, const char *, unsigned long)* outProc, void * refCon, WXMP_Result * wResult)  Line 968
    C++
    wxMeta.exe!TXMPMeta<std::basic_string<char,std::char_traits<char>,std::allocator<char> > >::DumpObject(long (void *, const char *, unsigned long)* outProc, void * refCon)  Line 833 + 0x21 bytes
    C++
    wxMeta.exe!MyFrame::DisplayMetaData(wxString & wsrFilePathAndName)  Line 484
    C++
    Hoping this will help resolve the issue :-)

  • Problems with Hindi language translation into DW8

    Hi,
    I have a whole truck load of problems ...but two have so far
    beat me!
    I have a Hindi text translation in MS Word ( yes I know copy
    and paste is
    evil...but (I have no other option)
    I have installed "Shusha" font, as advised by the translators
    who created
    the document. I have installed and applied the ( Win XP)
    Hindi language bar
    and keyboard.
    I am using the DW 8, P7 Ennergi Pack.- 3 col, left menu
    verion
    Because I do not read, write or even understand Hindi, I
    simply over paste
    the English with the Hindi version. In the "main body" centre
    column I dont
    have any real problems with this. BUT the P7 menu does not
    like being messed
    about!
    If I overcopy text while in theP7 "modify" interface...the
    Hindi become
    gobbleDegook....when I close the box , right click and select
    the text..then
    apply the shusha font...its seems OK....but when I re-open
    the "modify"
    interface....all the links I converted...dont show. They are
    there in the
    code and show up on the page..some links work - most dont.
    I have tried being a complete moron and simply pasting hindi
    text over the
    english in the normal page work mode.....useless!
    I have completed an "Arabic" version without any of these
    problems...well
    they started as above but I resolved them....the Hindi just
    wont have it!
    If by chance there is a Hindi capable web guy or
    guyess..with a bit of
    spare time who can give me
    some pointers...I will be more than grateful.
    ( Previously posted on the Project Seven newsgroup &
    hoping for advice.
    I have been advised of WEFT the MS embedded fonts utility for
    web pages -
    spo far not changed anything for t he better - but still
    working with it)
    TIA
    SB

    I think the problem has been identified - <u>should</u> be possible to solve in short-term (no guarantee )...
    Frank
    Edited by: Frank Moebius on Sep 22, 2009 11:39 AM
    The bug - which caused most of the trouble - had been fixed in 2006 - connection time has gone down to about 1/50!
    However, the design (loading translations when logging on) has not been changed so far. Subsequently using a large number of translations may still cause connection times which may be considered "too long".

  • Problem with clientgen task in weblogic 9.0

    hi,
    While developing webservices for our project, we used the clientgen ant task for generating the artifacts on the client side. I'm getting trouble in returning a collection of objects (list, vector) from the server to the client. When i try to do so, the clientgen task tries to create a java.util package on the client side in which it creates a List class which basically contains an object array with getter-setter. What i want is for the sun java collections classes to be used instead. The above problem applies to both user-defined as well as built-in objects i.e. say a list of objects of type T or a list of Strings. If someone, has any pointers regarding the same plz help...
    tia,
    cyril

              Try limiting the length of the session id.
              This is in either the weblogic.xml/web.xml
              files. You can get the dtd in the 6.0 doc.
              "Thierry Cools" <[email protected]> wrote:
              >Hi I have a problem with the session id that is generated in tha address bar
              >of my brower
              >When I start my application the following link appears
              >
              >http://127.0.0.1:7001/ebpp_beans/enterLogonInformation.do;jsessionid=Olw9VMc
              >0I1Z52YMgycyD23c8L4Ch2FurHLwbV2WhZw731dR3mBtJ|8186718416288373359/-140823374
              >1/7001/7002
              >
              >It seems that Weblogic 6.0 generates '/' characters in the session id.
              >This has for consequences that the servlet engine is taking the session id
              >in its context path, so that none of my links are working anymore.
              >
              >Could someone give me a clue to solve this problem
              >Thanks,
              >Thierry
              >
              >--
              >
              >Thierry Cools
              >
              >Senior Java Developer
              >S1 Brussels
              >Kleine Kloosterstraat, 23
              >1932 st. Stevens-Woluwe
              >Belgium
              >Tel : +32 2 200 43 82
              >Email : [email protected]
              >
              >
              >
              

  • Problem with HR inbound IDoc

    Hi,
    I have a problem with processing inbound IDocs from an external payroll system to infotype 0008.  The process is quite simple; salary changes from the payroll system is recorded to the corresponding employee's infotype 0008.  When testing I noticed that that the new record is recorded as is.  For example if there is already an existing record with the validity dates 01.01.2010 - 31.12.9999 and the new record in the IDoc is 01.01.2011 - 31.12.9999, the new record is written as is.  I would expect the system to delimit the old record correctly - similar to what would happen if you maintain records online.  There is no error checking or any kind of processing done by the system. 
    I did some debugging and found out that the function module linked to process code HRMD writes the information directly to the database.  If this is a standard way of doing this, it is really unusual.
    Has anybody else encountered this?  Any pointers?
    Thanks.
    Edited by: Theo Droste on Jan 20, 2011 11:56 AM

    Hi
    This is correct, the ale programme writes the data directly to the database, I have faced this isssue on occassion. I once raised an oss message on this as well, and sap confirms this is what happens. If it is a sap to sap ale, it somehow seems to work, - this could be because the outgoing idocs are created by sap itself  but if it is a non sap to sap ale the onus is on us to ensure that the external system sends the correct data to sap in the way we intend it to be displayed.
    that is how it has been in my experience. It is unusual, but apparently not impossible. Please let us know if you find out anything different.

Maybe you are looking for