Zip Exception Problem using jeode to unzip file

Hi,
I am using a JAva application, that has a client running on the IPAQ-Jeode JVM and the server is on JDK 1.3.
The application is using RMI to communicate between the Server and the PDA. The application code is running on JDK 1.3. Once the RMI communication is established, the server (during deployment) creates a ZIP file The zip file is serialized to the client machine (in this case, the Ipaq). The Client code later, running on Jeode, tries to unzip this file and extract the files.
We observed that the files created by the server (on JDK 1.3) cannot be unzipped by the Jeode JVM and gives an Zip Exception. The same zip file can be successfully uncompressed by machines having Sun's J2SE versions JDK 1.3 and JDK 1.1.8(which is compatible with PJava).
The exception which I receive is
2002-02-28 12:05:37,194 [Thread-0] INFO com.op40.utl.LogStream - java.util.zip.ZipException
2002-02-28 12:05:37,213 [Thread-0] INFO com.op40.utl.LogStream -      at java.util.zip.ZipInputStream.read (bytecode 305)
2002-02-28 12:05:37,234 [Thread-0] INFO com.op40.utl.LogStream -      at com.op40.java.util.InputStreamToOutputStream.copy (bytecode 54)
2002-02-28 12:05:37,879 [Thread-0] INFO com.op40.utl.LogStream -      at com.op40.java.util.FileToStream.copy (bytecode 71)
2002-02-28 12:05:37,895 [Thread-0] INFO com.op40.utl.LogStream -      at com.op40.dis.asset.FleAad.setPayload (bytecode 124)
2002-02-28 12:05:37,913 [Thread-0] INFO com.op40.utl.LogStream -      at com.op40.dis.deployer.Adp.immediateDeploy (bytecode 141)
2002-02-28 12:05:38,000 [Thread-0] INFO com.op40.utl.LogStream -      at com.op40.dis.deployer.Adp.immediateDeploy (bytecode 28)
2002-02-28 12:05:38,020 [Thread-0] INFO com.op40.utl.LogStream -      at com.op40.dis.deployer.Adp.scheduleDeployment (bytecode 91)
2002-02-28 12:05:38,037 [Thread-0] INFO com.op40.utl.LogStream -      at com.op40.dis.deployer.Adp.deployAssets (bytecode 17)
2002-02-28 12:05:38,632 [Thread-0] INFO com.op40.utl.LogStream -      at com.op40.dis.client.CdaImpl.processAssets (bytecode 33)
2002-02-28 12:05:38,652 [Thread-0] INFO com.op40.utl.LogStream -      at com.op40.dis.asttpt.PtpAstRcv.processAssets (bytecode 82)
2002-02-28 12:05:38,669 [Thread-0] INFO com.op40.utl.LogStream -      at com.op40.dis.asttpt.PtpAstRcv.access$900 (bytecode 3)

Is your client running as an applet or application? I have a client applet that tries to connect to my RMI server, but I keep getting the following exception even though java.rmi.registry.LocateRegistry is in the /Windows/lib/core.jar.
java.lang.NoClassDefFoundError: java/rmi/registry/LocateRegistry
at java.rmi.Naming.getRegistry (bytecode 12)
at java.rmi.Naming.list (bytecode 6)
Test.init (bytecode 6)
com.insignia.applet.AppletPanel.run (AppletPanel.java, line 0)
java.lang.Thread.run (bytecode 11)If you have any suggestions, that'd be great. You can contact me at [email protected]
Thanks,
Eric

Similar Messages

  • Problems using CFFile to move files

    Hi there - I am having some problems using CFFile to move
    files from one directory to another. It only seems to only move
    some of the files in the group that I choose to move over, and
    there is nothing specific about which files are moved - different
    files are moved each time. There are no error messages that are
    showing up, so I am totally baffled. Has anyone else experienced
    this or does anyone else have any ideas on how to fix this problem?
    Any help would be much appreciated!! My code is below.

    First, an incidental matter. I don't think there's any need
    for so many try-catch tags. I would expect all the tags <cffile
    action="move"> to throw the same class of exception, and so
    would use one try-catch for them all. Even if you expect different
    exceptions to be thrown, one cftry tag might still be sufficient,
    if implemented as follows
    <cftry>
    <!--- code block 1 --->
    <!--- code block 2 --->
    <!--- ... etc... --->
    <cfcatch type="exceptionType1"></cfcatch>
    <cfcatch type="exceptionType2"></cfcatch>
    <!--- ... etc... --->
    </cftry>
    I suspect the cause of the problems lies in the dynamic
    values that you give the attributes. Perhaps incorrect values are
    passed for the source and/or destination values in certain
    circumstances. As A3gis has said, if that were to happen you
    probably wouldn't notice, because of the try-catches. Find a way to
    ensure that all the generated values for source and destination are
    correct.

  • Problems using codeBase in policy file

    Hi,
    I'm experiencing problems using the codeBase option with the grant option in my .policy file.
    I want to start a 'manager' which has limited access to files. This manager starts workers (threads) who need complete access to... everything. I know I need to use the codebase identifier to separate those two groups of rights, but I don't seem to get the codebase right. The worker threads don't get any rights at all. The manager works fine (without any codebase definition).
    I've used full and relative pathnames, with and without '/-' or '/*'
    The file I use currently is (slightly truncated :), pathnames are package names:
    grant codeBase "file:com/response/timber/broker/*" {
    permission java.security.AllPermission;
    grant {
    permission [..].FilePermission "manager.properties","read";
    permission [..].SocketPermission "localhost","connect,[..]";
    permission [..].FilePermission "com/[..]/jaxb.properties","read";
    According to the various examples on the Internet, it should work, but ofcourse it doesn't. Could the use of threat have any influence on the defined security?

    All I can say is I hope someone gives you
    (and indirectly me) the answer. I've got a
    serious problem granting special permissions
    in Sun ONE to a servlet and it's jars.
    One strange thing (to me), is that the format
    for the .policy file seems quite different between
    say Tomcat and Sun One. Things that work in Tomcat
    don't work for Sun ONE. But I thought the underlying
    vm et al was all the same! So did I... Why does
    this have to be so Opaque? Why different syntax
    platform to platform? Its all Java!

  • Problem using read from spreadsheet file and polar plotting

    Hi to all labview users,
    i am a beginner in labview and i am trying to do a polar plot.
    i read the polar plotting example in labview and it was straightforward.
    I used "write to spreadsheet file" to gather data.
    and they are in the following format
    13  10
    4  20
    8 30
    ....etc
    now. i tried using "read from spreadfile" to get the data into a array, then using "array to cluster" to convert the array into cluster, so i could connect it to the polar plot block
    however, it kept saying i couldnt connect that way, because polar plot uses 1-d array with cluster of 2 element and my source is a cluster of 9 elements....
    but doesnt the "read from spreadfile" block give me a 1-d array? and where does that 9 come from? i only have 3 rows and 2 columns in my data file....
    any guidance would be greatly appreciated.
    thx alot
    Happy guy
    ~ currently final year undergraduate in Electrical Engr. Graduating soon! Yes!
    ~ currently looking for jobs : any position related to engineering, labview, programming, tech support would be great.
    ~ humber learner of LabVIEW lvl: beginner-intermediate

    Helllo,
    I've made an example to try to help you  with that question.
    Notes:
     - the file must have values separeted by tab
     - reading the values from file as you mentioned using "read from spreadfile" you'll get a 2D array and not 1D;
    Software developer
    www.mcm-electronics.com
    PS: Don't forget to rate a good anwser ; )
    Currently using Labview 2011
    PORTUGAL
    Attachments:
    Read Table and plot polar graph.vi ‏26 KB
    teste.txt ‏1 KB

  • Problem using payloadZipBean to unzip a file

    I would like to run OS command to zip large amount of xml into one zip file before XI pick it up, then use unzip option in payloadZipBean to unzip all the xml files so XI can process them one by one from memory, this will reduce lots file I/O time.
    But I am getting this warning message:
    Warning Zip: message is empty or has no payload
    But if I display this message in RWB, the only payload is this zip file.
    Why the payloadZipBean is not unzipping this file as supposed to.

    Did you check this blog
    /people/stefan.grube/blog/2007/02/20/working-with-the-payloadzipbean-module-of-the-xi-adapter-framework
    Sameer

  • Problem using inputFile when a file is not required

    I have a page that has a table component with the following fields
    Title - inputText (required)
    File - inputFile (not required)
    File URL - inputText (not required)
    The code for the input File is this:
    <af:inputFile id="OrigReviewAttachmentsFileDoc"
      label="#{bindings.OrigReviewAttachmentsFileDoc.hints.label}"
      visible="#{bindings.OrigReviewItemsStatName.attributeValue == 'New' &amp;&amp; row.bindings.ClsfnCode.attributeValue != null &amp;&amp; row.bindings.ClsfnCode.attributeValue == 'PEND'}"
      valueChangeListener="#{OrigReviewAttachmentsFileDocFileHandlerInTable.uploadFile}"
      binding="#{OrigReviewAttachmentsFileDocFileHandlerInTable.usesUploadIndicator}"
      shortDesc="#{bindings.OrigReviewAttachmentsFileDoc.hints.tooltip}"
      partialTriggers="OrigReviewAttachmentsClsfnCode"/>The problem I am having is that under certain circumstances (like the file is too large) I want the user to enter a File URL instead of actually uploading a file. But when I save the row with the other 2 fields on the screen filled in and not the inputFile I get the error, "Cannot Find the File Specified." Is there any way that I can tell it, that if it doesn't have a path, to not try to upload a file?
    I am using:
    JDeveloper 11.1.1.4.0
    Oracle 11g with OrdSys.OrdDoc column to hold the file.
    Any help would be appreciated,
    Michelle

    I figured out the problem, and it was caused by JHeadstart. I placed a default display value in JHeadstart for the download link. This caused a default values bean to be created by jheadstart and that bean was what was causing my problem. The download link had a display value, but not a valid path. I removed the default value, re-generated the screen and it is totally fixed. I have pretty much decided that it is never a good idea to use a default value in Jheadstart.
    Thanks for your help,
    Michelle

  • Problem using CLIENT_HOST to FTP files from forms 10g

    Hi,
    I'm trying to use forms 10g ODS & webutil on NT to allow a user to select a file on their machine or the network and transfer the file to a DB server. I want to transfer the file to the file system on the server NOT to a BLOB column on a table. I've used webutil (client_get_file_name) to display the dialog box and allow the user to select a file. This works fine. However, when I try to use CLIENT_HOST to FTP the selected
    to the remote server it fails. I'm issuing the following
    command from within forms:
    CLIENT_HOST('CMD /C ftp -n -s:C:\test_ftp.txt');
    Something tries to kick off and a grey area the size of a dos window flashes on to the screen for a couple of seconds but the FTP does not happen. The test_ftp.txt script is a simple ftp script that transfers one file and I know it works as I can run it successfully from DOS or NT command prompt.
    Questions:
    (1) Does anyone know how to pause the DOS window so you
    can see what is actually happening ?
    (2) Has anyone implemented FTP functionality
    successfully using CLIENT_HOST in the same way that's
    I'm trying ?
    Any help would be appreciated.
    Declan.

    Okay...
    Try this. For testing put your command in a .bat-file. In this .bat-file spool the output to a text-file. I'm not sure about the spooling syntax under Windows, check the Windows documentation for that. Like:
    ftp.bat
    start spooling to ftp.out
    cmd /c .... (your ftp commando line)
    stop spooling
    Perhaps the problem is that it doesn't find the ftp commando? Try hardcode the path like c:\windows\ftp ... etc!

  • Problems using a php include file with an Add-on Domain.

    Hello,
    I am having an issue getting a php include file to work with a new add-on domain I am working on getting up and running.
    This include file is one that supplies the rest of the php code to a contact form page.  It works as it should for my original domain, same file no difference.  I made sure that the files hosted on the remote server had all read, write, execute permissions turned on.
    I have tried putting the include file in several different locations as a test, such as:
    I don't remember the exact name of the include file at the moment, as I'm at work so I will designate it below as 'includefile.php'.
    file path for add-on domain - ../public_html/lorentzpainting/includes/includefile.php
    alternatives I tried moving the file to - ../public_html/lorentzpainting/includefile.php
    When none of those options worked, I tried just pointing the path towards other places on the server that have the include file such as:
    ../public_html/includes/includefile.php
    ../public_html/includefile.php
    What can I do here?  Should I just give up on using the include and put all the code back in the page?  Doesn't seem like it should be this complicated, it works perfectly fine for my other site.. and still does.
    Thanks ahead of time to those who may assist me.

    here is the contents of the include file:
    <?php
    if (isset($_SERVER['SCRIPT_NAME']) && strpos($_SERVER['SCRIPT_NAME'],
    '.inc.php')) exit;
    // remove escape characters from POST array
    if (PHP_VERSION < 6 && get_magic_quotes_gpc()) {
      function stripslashes_deep($value) {
        $value = is_array($value) ? array_map('stripslashes_deep', $value) : stripslashes($value);
        return $value;
      $_POST = array_map('stripslashes_deep', $_POST);
      // assume that there is nothing suspect
        $suspect = false;
        // create a pattern to locate suspect phrases
        $pattern = '/Content-Type:|Bcc:|Cc:/i';
          // function to check for suspect phrases
      function isSuspect($val, $pattern, &$suspect) {
        // if the variable is an array, loop through each element
        // and pass it recursively back to the same function
        if (is_array($val)) {
          foreach ($val as $item) {
            isSuspect($item, $pattern, $suspect);
        else {
          // if one of the suspect phrases is found, set Boolean to true
          if (preg_match($pattern, $val)) {
            $suspect = true;
    //check the $_POST array and any subarrays for suspect content
    isSuspect($_POST, $pattern, $suspect);
        if (!empty($_POST['url'])) {
            $suspect = true;
        if ($suspect) {
            $mailSent = false;
            unset($missing);
        } else {
        //process the $_POST variables
        foreach ($_POST as $key => $value) {
            // assign to temporary variable and strip whitespace if not an array
            $temp = is_array($value) ? $value : trim($value);
            // if empty and required, add to $missing array
            if (empty($temp) && in_array($key, $required)) {
                array_push($missing, $key);
            } elseif (in_array($key, $expected)) {
                // otherwise, assign to a variable of the same name as $key
                ${$key} = $temp;
        //validate the email address
        if (!empty($email)) {
            // regex to identify illegal characters in email address
    $checkEmail = '/^[^@]+@[^\s\r\n\'";,@%]+$/';
    //reject the email address if it doesn't match
    if (!preg_match($checkEmail, $email)) {
        $suspect = true;
        $mailSent = false;
        unset($missing);
    //go ahead only if all required fields OK
        if (!$suspect && empty($missing)) {
    //initialize the $message variable
            $message = '';
    // loop through the $expected array
            foreach($expected as $item) {
    // assign the value of the current item to $val
                if (isset(${$item}) && !empty(${$item})) {
                    $val = ${$item};
                } else {
    // if it has no value, assign 'Not selected'
                    $val = 'Not selected';
    // if an array, expand as comma-sparated string
                if (is_array($val)) {
                    $val = implode(',', $val);
    // add label and value to the message body
      $message .= ucfirst($item).": $val\r\n\r\n";
    //limit line length to 70 characters
        $message = wordwrap($message, 70);
    //create Reply-To header
        if (!empty($email)) {
            $headers .= "\r\nReply-To: $email";
    // send it
        $mailSent = mail($to, $subject, $message, $headers);
        if ($mailSent) {
    // $missing is no longer needed if the email is sent, so unset it
            unset($missing); echo('Thank you for contacting Common Wealth Web Solutions');
    ?>
    <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
    <html xmlns="http://www.w3.org/1999/xhtml">
    <head>
    <meta http-equiv="Content-Type" content="text/html; charset=UTF-8" />
    <title>Untitled Document</title>
    </head>
    <body>
    </body>
    </html>

  • Problem using iPod to backup files

    Sorry to double post, but I just realized that I'd somehow managed to post this on the Mac forum.
    Using WinXP Pro SP2, I connected my 60G iPod and checked Enable disk use. I then tried several times to transfer files to the iPod, but kept getting the following message:
    "The disk in the destination drive is full. Insert a new disk to continue."
    Checking the properties for the drive shows plenty of available space.
    Any ideas as to what I'm doing wrong?
    Also is it possible to use other than the FATS format on an iPod?

    How many things do you have connected to usb?
    Do you have alot of drive letters in "My Computer"
    Windows may be confusing the iPod with a network drive, try changing the drive letter
      Mac OS X (10.4.8)  

  • Problem using scp to copy files between Macs

    I've been trying to copy a folder hierarchy containing a slew of icon files between two of my Macs. Using scp has proven to be futile.
    The command line I'm using is
    scp -r -p -C -E [email protected]:/Icons /Icons
    After a number of files are copied I receive the following error message on a file and all thereafter:
    ./icons/Agrapha-Christmas/Boxes: Too many open files
    More specifically the "too many open files" message.
    I am using Tiger 10.4.7
    What am I doing wrong here?

    One more piece of information. I ran scp with the verbose switch and have attached a portion of the feedback showing successfull copies followed by the error which occurs on all subsequent files:
    Entering directory: D0777 0 Adobe CS Icons
    Sending file modes: C0644 6148 .DS_Store
    .DS_Store 100% 6148 6.0KB/s 00:00
    copyfile(/icons/Adobe CS Icons/..DSStore, /tmp/scp.md.DJyatJ, PACK)
    Sending file modes: C0400 82 ..DSStore
    ..DSStore 100% 82 0.1KB/s 00:00
    copyfile(/icons/icons/Adobe CS Icons/..DSStore.N8P, /icons/icons/Adobe CS Icons/.DS_Store, UNPACK)
    Sending file modes: C0777 68489 Adobe CS Box.icns
    Adobe CS Box.icns 100% 67KB 66.9KB/s 00:01
    copyfile(/icons/Adobe CS Icons/._Adobe CS Box.icns, /tmp/scp.md.00r28e, PACK)
    Sending file modes: C0400 70745 ._Adobe CS Box.icns
    ._Adobe CS Box.icns 100% 69KB 69.1KB/s 00:00
    copyfile(/icons/icons/Adobe CS Icons/._Adobe CS Box.icns.K7Q, /icons/icons/Adobe CS Icons/Adobe CS Box.icns, UNPACK)
    Sending file modes: C0777 56691 Adobe CS Folder 2.icns
    Adobe CS Folder 2.icns 100% 55KB 55.4KB/s 00:00
    copyfile(/icons/Adobe CS Icons/._Adobe CS Folder 2.icns, /tmp/scp.md.EgDDb2, PACK)
    Sending file modes: C0400 58947 ._Adobe CS Folder 2.icns
    ._Adobe CS Folder 2.icns 100% 58KB 57.6KB/s 00:00
    copyfile(/icons/icons/Adobe CS Icons/._Adobe CS Folder 2.icns.hrC, /icons/icons/Adobe CS Icons/Adobe CS Folder 2.icns, UNPACK)
    Sending file modes: C0777 57330 Adobe CS Folder 3.icns
    Adobe CS Folder 3.icns 100% 56KB 56.0KB/s 00:01
    copyfile(/icons/Adobe CS Icons/._Adobe CS Folder 3.icns, /tmp/scp.md.rDiEBy, PACK)
    Sending file modes: C0400 59586 ._Adobe CS Folder 3.icns
    ._Adobe CS Folder 3.icns 100% 58KB 58.2KB/s 00:00
    copyfile(/icons/icons/Adobe CS Icons/._Adobe CS Folder 3.icns.5nC, /icons/icons/Adobe CS Icons/Adobe CS Folder 3.icns, UNPACK)
    Sending file modes: C0777 56289 Adobe CS Folder.icns
    Adobe CS Folder.icns 100% 55KB 55.0KB/s 00:00
    copyfile(/icons/Adobe CS Icons/._Adobe CS Folder.icns, /tmp/scp.md.CVrek6, PACK)
    Sending file modes: C0400 58545 ._Adobe CS Folder.icns
    ._Adobe CS Folder.icns 100% 57KB 57.2KB/s 00:00
    copyfile(/icons/icons/Adobe CS Icons/._Adobe CS Folder.icns.JMp, /icons/icons/Adobe CS Icons/Adobe CS Folder.icns, UNPACK)
    Sending file modes: C0777 54831 Adobe CS Logo.icns
    Adobe CS Logo.icns 100% 54KB 53.6KB/s 00:00
    copyfile(/icons/Adobe CS Icons/._Adobe CS Logo.icns, /tmp/scp.md.keUSYq, PACK)
    Sending file modes: C0400 57087 ._Adobe CS Logo.icns
    ._Adobe CS Logo.icns 100% 56KB 55.8KB/s 00:00
    copyfile(/icons/icons/Adobe CS Icons/._Adobe CS Logo.icns.M39, /icons/icons/Adobe CS Icons/Adobe CS Logo.icns, UNPACK)
    Entering directory: D0777 0 Agrapha-Christmas
    Sending file modes: C0644 12292 .DS_Store
    .DS_Store 100% 12KB 12.0KB/s 00:00
    copyfile(/icons/Agrapha-Christmas/..DSStore, /tmp/scp.md.1s6tDJ, PACK)
    Sending file modes: C0400 82 ..DSStore
    ..DSStore 100% 82 0.1KB/s 00:00
    copyfile(/icons/icons/Agrapha-Christmas/..DSStore.Jmw, /icons/icons/Agrapha-Christmas/.DS_Store, UNPACK)
    Sending file modes: C0777 0 Angel
    Angel 100% 0 0.0KB/s 00:00
    copyfile(/icons/Agrapha-Christmas/._Angel, /tmp/scp.md.hueGHQ, PACK)
    Sending file modes: C0400 44820 ._Angel
    ._Angel 100% 44KB 43.8KB/s 00:00
    copyfile(/icons/icons/Agrapha-Christmas/._Angel.ZDF, /icons/icons/Agrapha-Christmas/Angel, UNPACK)
    Sending file modes: C0777 0 Basket
    Basket 100% 0 0.0KB/s 00:00
    copyfile(/icons/Agrapha-Christmas/._Basket, /tmp/scp.md.8NWfFT, PACK)
    Sending file modes: C0400 49608 ._Basket
    ._Basket 100% 48KB 48.5KB/s 00:00
    copyfile(/icons/icons/Agrapha-Christmas/._Basket.eIY, /icons/icons/Agrapha-Christmas/Basket, UNPACK)
    Sending file modes: C0777 0 Boxes
    Boxes 100% 0 0.0KB/s 00:00
    copyfile(/icons/Agrapha-Christmas/._Boxes, /tmp/scp.md.oWFB8k, PACK)
    Sending file modes: C0400 46314 ._Boxes
    ._Boxes 100% 45KB 45.2KB/s 00:00
    copyfile(/icons/icons/Agrapha-Christmas/._Boxes.GGu, /icons/icons/Agrapha-Christmas/Boxes, UNPACK)
    Sending file modes: C0777 0 Broach
    /icons/icons/Agrapha-Christmas/Broach: Too many open files
    Sending file modes: C0777 0 Candy Cane
    /icons/icons/Agrapha-Christmas/Candy Cane: Too many open files
    Starting with the Broach file, all subsequent files display the error "too many open files".

  • Troubleshooting Network Problems Using Log and Trace Files

    Hi,
    can any one tell me how to generate trace and log files related to network errors.

    start with inspecting listener.log
    Post tailend (last 40 lines here) of listener.log

  • Problems using 1080/50p video files in Premiere CS5 to Encore onto Blu-ray

    Firstly apologies if this has been covered, I've searched various forums but not found anything on 1080/50p.
    I have a Panasonic HDC-TM700 video camera which I only had for a week or so before I went on a weeks holiday in Egypt, so I thought I'd record my memories in the superb 1080/50p mode.
    Upon returning I am now editing my work in Premiere CS5, the resulting video is very jerky (it plays back OK on the camera), rendering a sample in Encore I discover Blu-ray doesn't support 1080/50p so I tried the HDTV 1080i 25 setting, on the resulting Blu-ray the panning scenes are not smooth viewing and the video is still rather jerky.
    Should I have recorded the footage in 1080 interlaced instead?
    Does anybody know if I can salvage my memories?
    Hoping you can help
    Regards
    Kevin

    I have the same camera, only it's NTSC so 1080p60. What I do for now, until there is wider acceptance for this format, is that I export to 720p60. I know I may sound like a party pooper but I prefer progressive before interlaced...  I am sure there will be support for 1080p50 or 1080p60 in the near future, so your memories will not be lost!
    Some Bluray players do support 1080p60/50, but not from a disc but an SD card instead. If that's an option for you, go with that!
    Here's the player I use which plays back 1080p60 beautifully from SD, I'm sure there is an equivalent player for PAL:
    https://panasonic.ca/english/audiovideo/bluray/players/DMPBDT100.asp

  • Problems unzipping files downloaded from Firefox using Windows built-in extractor tool

    Hi,
    I am running Firefox v26 on a Windows 7 (64-bit) PC. I work for a library that is able to reproduce digital audio books that we can download from the Library of Congress's National Library Service (NLS) for the Blind and Physically Handicapped. I use Firefox to download a lot of zipped audio books for my job. The file formats found in one book's zipped folder often include the following file types: .3gp (they're like a more compressed MP3 audio file), .mp3, .ncx, .opf, .pncx, .ppf, .smil, .md5, .dtd, .ent, and .ao.
    If I use Windows' built-in extraction tool to unzip an audio book I have downloaded via Firefox the extraction process ends prematurely and acts as if it was successful, but in reality some files are were never extracted. For example, I downloaded a zipped audio book that contained 21 files and weighed in at 87.6MB. The built-in Windows extraction tool only unzipped 2 files weighing in at 30.3MB.
    I have downloaded this same zipped audio book using Google's Chrome v31 and IE 11. Both downloads unzipped successfully using Windows' extractor. To make things even more bizarre, I *was* able to successfully unzip my audio book that I downloaded using Firefox using a 3rd party extraction tool called Extract Now. Because I tend to use this 3rd party program to extract my audio files I don't know at which version of Firefox this problem started. All I know if that this hasn't always been the case.
    Please help.
    Thanks,
    -Dan M.

    Boot the computer in Windows Safe Mode with network support (press F8 on the boot screen) as a test to see if that helps with extracting files.

  • Want To create Zip file  using java,And Unzip without Java Program

    I want to create a zip text file using java, I know Using ZipOutputStream , we can create a zip file, , But i want to open that zip file without java program. suppose i use ZipOutputStream , then zip file is created But for unZip also difftrent program require. We cant open that zip file without writing diff java program.
    Actually i have one text file of big size want to create zip file using java , and unzip simply without java program.Its Possible??
    Here is one answer But I want to open that file normal way(
    For Exp. using winzip we can create a zip file and also open simply)
    http://forum.java.sun.com/thread.jspa?threadID=5182691&tstart=0

    Thanks for your Reply,
    I m creating a zip file using this program, Zip file Created successfully But when im trying to open .zip file i m getting error like "Canot open a zip file, it does not appear to be valid Archive"
    import java.io.*;
    import java.util.zip.*;
    public class ZipFileCreation
         public static void main (String argv[])
         try {
         FileOutputStream fos = new FileOutputStream ( "c:/a.zip" );
         ZipOutputStream zip = new ZipOutputStream ( fos );
         zip.setLevel( 9 );
         zip.setMethod( ZipOutputStream.DEFLATED );
    //     get the element file we are going to add, using slashes in name.
         String elementName = "c:/kalpesh/GetSigRoleInfo092702828.txt";
         File elementFile = new File ( elementName );
    //     create the entry
         ZipEntry entry = new ZipEntry( elementName );
         entry.setTime( elementFile.lastModified() );
    //     read contents of file we are going to put in the zip
         int fileLength = (int)elementFile.length();
         System.out.println("fileLength = " +fileLength);
         FileInputStream fis = new FileInputStream ( elementFile );
         byte[] wholeFile = new byte [fileLength];
         int bytesRead = fis.read( wholeFile , 0 /* offset */ , fileLength );
    //     checking bytesRead not shown.
         fis.close();
    //     no need to setCRC, or setSize as they are computed automatically.
         zip.putNextEntry( entry );
    //     write the contents into the zip element
         zip.write( wholeFile , 0, fileLength );
         zip.closeEntry(); System.out.println("Completed");
    //     close the entire zip
         catch(Exception e) {
    e.printStackTrace();
    }

  • Cannot use winzip to unzip the zip file zipped by java.util.zip

    Hi all,
    I use the followcode to create a zip file, and i downlaod it and try to use winzip to unzip this file but fail. The path is correct and i got the zip file. but it just cannot unzip.
    pls help
    thanks alot.
    Kin
              int count = 0;
              count = ContentDocuments.size();
              for (int i = 0; i < bb; i++)     {
                   System.out.println(filenames[i] + "");
              // Create a buffer for reading the files
              byte[] buf = new byte[10*1024*1024];
              try {      
                   String outFilename = MyDir + "zipfile/" + getContentID2()+".zip";
                   System.out.println("outFilename = " + outFilename);
                   ZipOutputStream out = new ZipOutputStream(new FileOutputStream(outFilename));
                   for (int i=0; i<filenames.length; i++) {           
                        FileInputStream in = new FileInputStream(filenames);
                        out.putNextEntry(new ZipEntry(filenames[i]));
                        int len;
                        while ((len = in.read(buf)) != -1) {               
                             out.write(buf, 0, len);
                        out.closeEntry();
                        in.close();
                   out.close();
              } catch (IOException e)     {
                   System.err.println("zipprocess " + e.getMessage());

    I've written a replacement zip file creator class. Not much tested but it seems to work, however I've yet to try it with the version of WINZIP that rejected my previous attempts. Oh, and the stored dates are garbage.
    * ZipOutputFile.java
    * Created on 25 March 2004, 13:08
    package zip;
    import java.io.IOException;
    import java.nio.ByteBuffer;
    * <p>Creates a ZIP archive in a file which WINZIP should be able to read.</p>
    * <p>Unfortunately zip archives generated by the standard Java class
    * {@link java.util.zip.ZipOutputStream}, while adhering to PKZIPs format specification,
    * don't appear to be readable by some versions of WinZip and similar utilities. This is
    * probably because they use
    * a format specified for writing to a non-seakable stream, where the length and CRC of
    * a file is writen to a special block following the data. Since the length of the binary
    * date is unknown this makes an archive quite complicated to read, and it looks like
    * WinZip hasn't bothered.</p>
    * <p>All data is Deflated. Close completes the archive, flush terminates the current entry.</p>
    * @see java.util.zip.ZipOutputStream
    * @author  Malcolm McMahon
    public class ZipOutputFile extends java.io.OutputStream {
        byte[] oneByte = new byte[1];
        java.io.RandomAccessFile archive;
        public final static short DEFLATE_METHOD = 8;
        public final static short VERSION_CODE = 20;
        public final static short MIN_VERSION = 10;
        public final static int  ENTRY_HEADER_MAGIC = 0x04034b50;
        public final static int  CATALOG_HEADER_MAGIC = 0x02014b50;
        public final static int  CATALOG_END_MAGIC = 0x06054b50;
        private final static short DISC_NUMBER = 0;
        ByteBuffer entryHeader = ByteBuffer.wrap(new byte[30]);
        ByteBuffer entryLater = ByteBuffer.wrap(new byte[12]);
        java.util.zip.CRC32 crcAcc = new java.util.zip.CRC32();
        java.util.zip.Deflater def = new java.util.zip.Deflater(java.util.zip.Deflater.DEFLATED, true);
        int totalCompressed;
        long MSEPOCH;
        byte [] deflateBuf = new byte[2048];
        public static final long SECONDS_TO_DAYS = 60 * 60 * 24;
         * Entry stores info about each file stored
        private class Entry {
            long offset;        // position of header in file
            byte[] name;
            long crc;
            int compressedSize;
            int uncompressedSize;
            java.util.Date date;
             * Contructor also writes initial header.
             * @param fileName Name under which data is stored.
             * @param date  Date to label the file with
             * @TODO get the date stored properly
            public Entry(String fileName, java.util.Date date) throws IOException {
                name = fileName.getBytes();
                this.date = date == null ? new java.util.Date() : date;
                entryHeader.position(10);
                putDate(entryHeader);
                entryHeader.putShort(26, (short)name.length);
                offset = archive.getFilePointer();
                archive.write(entryHeader.array());
                archive.write(name);
                catalog.add(this);
                crcAcc.reset();
                totalCompressed = 0;
                def.reset();
             * Finish writing entry data. Save the lenghts & crc for catalog
             * and go back and fill them in in the entry header.
            public void close() throws IOException {
                def.finish();
                while(!def.finished())
                    deflate();
                entryLater.position(0);
                crc = crcAcc.getValue();
                compressedSize = totalCompressed;
                uncompressedSize = def.getTotalIn();
                entryLater.putInt((int)crc);
                entryLater.putInt(compressedSize);
                entryLater.putInt(uncompressedSize);
                long eof = archive.getFilePointer();
                archive.seek(offset + 14);
                archive.write(entryLater.array());
                archive.seek(eof);
             * Write the catalog data relevant to this entry. Buffer is
             * preloaded with fixed data.
             * @param buf Buffer to organise fixed lenght part of header
            public void writeCatalog(ByteBuffer buf) throws IOException {
                buf.position(12);
                putDate(buf);
                buf.putInt((int)crc);
                buf.putInt(compressedSize);
                buf.putInt(uncompressedSize);
                buf.putShort((short)name.length);
                buf.putShort((short)0);  // extra field length
                buf.putShort((short)0);  // file comment length
                buf.putShort(DISC_NUMBER);  // disk number
                buf.putShort((short)0); // internal attributes
                buf.putInt(0);      // external file attributes
                buf.putInt((int)offset); // file position
                archive.write(buf.array());
                archive.write(name);
             * This writes the entries date in MSDOS format.
             * @param buf Where to write it
             * @TODO Get this generating sane dates
            public void putDate(ByteBuffer buf) {
                long msTime = (date.getTime() - MSEPOCH) / 1000;
                buf.putShort((short)(msTime % SECONDS_TO_DAYS));
                buf.putShort((short)(msTime / SECONDS_TO_DAYS));
        private Entry entryInProgress = null; // entry currently being written
        private java.util.ArrayList catalog = new java.util.ArrayList(12);  // all entries
         * Start a new output file.
         * @param name The name to store as
         * @param date Date - null indicates current time
        public java.io.OutputStream openEntry(String name, java.util.Date date) throws IOException{
            if(entryInProgress != null)
                entryInProgress.close();
            entryInProgress = new Entry(name, date);
            return this;
         * Creates a new instance of ZipOutputFile
         * @param fd The file to write to
        public ZipOutputFile(java.io.File fd) throws IOException {
            this(new java.io.RandomAccessFile(fd, "rw"));
         * Create new instance of ZipOutputFile from RandomAccessFile
         * @param archive RandomAccessFile
        public ZipOutputFile(java.io.RandomAccessFile archive) {
            this.archive = archive;
            entryHeader.order(java.nio.ByteOrder.LITTLE_ENDIAN);  // create fixed fields of header
            entryLater.order(java.nio.ByteOrder.LITTLE_ENDIAN);
            entryHeader.putInt(ENTRY_HEADER_MAGIC);
            entryHeader.putShort(MIN_VERSION);
            entryHeader.putShort((short)0);  // general purpose flag
            entryHeader.putShort(DEFLATE_METHOD);
            java.util.Calendar cal = java.util.Calendar.getInstance();
            cal.clear();
            cal.set(java.util.Calendar.YEAR, 1950);
            cal.set(java.util.Calendar.DAY_OF_MONTH, 1);
    //        def.setStrategy(Deflater.HUFFMAN_ONLY);
            MSEPOCH = cal.getTimeInMillis();
         * Writes the master catalogue and postamble and closes the archive file.
        public void close() throws IOException{
            if(entryInProgress != null)
                entryInProgress.close();
            ByteBuffer catEntry = ByteBuffer.wrap(new byte[46]);
            catEntry.order(java.nio.ByteOrder.LITTLE_ENDIAN);
            catEntry.putInt(CATALOG_HEADER_MAGIC);
            catEntry.putShort(VERSION_CODE);
            catEntry.putShort(MIN_VERSION);
            catEntry.putShort((short)0);
            catEntry.putShort(DEFLATE_METHOD);
            long catStart = archive.getFilePointer();
            for(java.util.Iterator it = catalog.iterator(); it.hasNext();) {
                ((Entry)it.next()).writeCatalog(catEntry);
            catEntry.position(0);
            catEntry.putInt(CATALOG_END_MAGIC);
            catEntry.putShort(DISC_NUMBER);
            catEntry.putShort(DISC_NUMBER);
            catEntry.putShort((short)catalog.size());
            catEntry.putShort((short)catalog.size());
            catEntry.putInt((int)(archive.getFilePointer() - catStart));
            catEntry.putInt((int)catStart);
            catEntry.putShort((short)0);
            archive.write(catEntry.array(), 0, catEntry.position());
            archive.setLength(archive.getFilePointer());  // truncate if old file
            archive.close();
            def.end();
         * Closes entry in progress.
        public void flush() throws IOException{
            if(entryInProgress == null)
                throw new IllegalStateException("Must call openEntry before writing");
            entryInProgress.close();
            entryInProgress = null;
         * Standard write routine. Defined by {@link java.io.OutputStream}.
         * Can only be used once openEntry has defined the file.
         * @param b  Bytes to write
        public void write(byte[] b) throws IOException{
            if(entryInProgress == null)
                throw new IllegalStateException("Must call openEntry before writing");
            crcAcc.update(b);
            def.setInput(b);
            while(!def.needsInput())
                deflate();
         * Standard write routine. Defined by {@link java.io.OutputStream}.
         * Can only be used once openEntry has defined the file.
         * @param b  Bytes to write
        public void write(int b) throws IOException{
            oneByte[0] = (byte)b;
            crcAcc.update(b);
            write(oneByte, 0, 1);
         *  Standard write routine. Defined by {@link java.io.OutputStream}.
         * Can only be used once openEntry has defined the file.
         * @param b  Bytes to write
         * @param off Start offset
         * @param len Byte count
        public void write(byte[] b, int off, int len) throws IOException{
            if(entryInProgress == null)
                throw new IllegalStateException("Must call openEntry before writing");
            crcAcc.update(b, off, len);
            def.setInput(b, off, len);
            while(!def.needsInput())
                deflate();
        * Gets a buffer full of coded data from the deflater and writes it to archive.
        private void deflate() throws IOException {
            int len = def.deflate(deflateBuf);
            totalCompressed += len;
            if(len > 0)
                archive.write(deflateBuf, 0, len);

Maybe you are looking for

  • Fusion Middleware 11g Weblogic, forms and Reports in Solaris  11 zone

    Hi, Any one Installed Fusion Middleware 11g Weblogic, forms and Reports on Solaris 11 zone ?? So Pls Let me know Weblogic and Fusion Middleware 11g versions which are working I try Weblogic 10.3.3 Installed 64 Bit Went OK Fusion Middleware 11g 11.1.1

  • Oracle SQL Developer - File, Open , File Name Not Accept UNC Path Form

    In Oracle SQL Developer, at least the File, Open (Dialog's), the File Name (if not all File / Directory Name) Text Box(es) is (are) not accepting, at least for me, the UNC Path Form of Folders on Network Servers. The same UNC Path (or UNC Paths in ge

  • VL10B and Delivery button

    Hi, When i am executing the VL10B, the output lists all the deliveries which are due to be delivered. But the issue now is the buttons used to create the deliveries are not appearing in the application bar. If i press "Shift+F7", it creates the deliv

  • Dragon Writer MS-8348B having some bad problems.

    I actually don't have the 8348B, I have a Verbatim but it's the same rebaged drive(BenQ 4816p) and I'm having alot of problems, mostly the same problems that other people are having.  The drive initially would recognize some cheap media at 40x MAX bu

  • Does anyone know what's the difference of storage loation type & bin?

    Hi Experts   Now we use WM. Hope to consult if we could assgin one locaiton for two location type ? Does that mean, for each location type ,it will has its own bin number and physical place? Another thing, we hope when locaiton type A is empty , it c