Question about 25 additional points

Hi all,
  I hope this is the right place for this question...
Today I've found a nice gift from SDN: 25 additional points with the comment: "Joined the SDN World"
I'm wondering what I've done special in these days (since I joined SDN some years ago)...
Maybe it's a mistake...
Did it happened to anyone else ?
Have a nice day,
Manuel

hi Manuel,
Its not for joining SDN but joining SDN World.
Check this out.
http://sdn.idizaai.be/sdn_world/sdn_world.html
/people/eddy.declercq/blog/2006/05/24/we-are-the-world
Good day
~Ak

Similar Messages

  • A question about character addition.

    Hi,
    i have a question about character. seet the following code.
    DATA: lv_entryid TYPE char3.
    lv_entryid = '001'.
    WRITE : / lv_entryid.
    lv_entryid = lv_entryid + 1.
    WRITE : / lv_entryid.
    the  answer is :
    001
    2
    but the result i expect is
    001
    002.
    how to process, Could you please help me?
    Thanks in advance.

    DATA: lv_entryid TYPE char3.
    lv_entryid = '001'.
    WRITE : / lv_entryid.
    lv_entryid = lv_entryid + 1.
    CALL FUNCTION 'CONVERSION_EXIT_ALPHA_INPUT'
         EXPORTING
              input  = lv_entryid
         IMPORTING
              output = lv_entryid.
    WRITE : / lv_entryid.
    Try this code and see
    Regards
    Gopi

  • Question about OTL additional hour pay if employee didn't submit meal break

    I have a question
    Does OTL capable of adding the standard rule where the employer is required to pay an employee 1 hour of additional pay if they did not have their 30 minute meal period after working 6 hours? In those rare cases where employees are not doing what they are supposed to do, it should be an automatic process.

    Dharma,
    I guess you are talking about California Meal penalty rule. Well, it is not supported in OTL. On the other hand, Oracle has the hour deduction policy setup which actually deducts hours for a given duration. I haven't tested this, but just try using that feature with a negative deduction.
    --Shiv                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                       

  • Various questions about System Restore Point

    How can I delete old System Restore Points?
    Is it possible to move the folder that holds them to a different drive then C:?
    Can I limit the amount of data System Restore is allowed to have?
    I have downloaded the TreeSize program to see what is using up so much space on my SSD and a folder named "System Volume Information" has 17GB of storage. Now I have found out that this is the folder that keeps System Restore Points and I checked
    on my restore points. There are too many of them and I was wondering if I could delete some of the old ones? I read on another post that someone said that you could limit the amount of data System Restore is allowed to use. Is this possible?
    Best Regards.

    Hello HurricaneHojax,
    Here I list the requirement, and please correct me if I have misunderstanding:
    1. Delete some System restore point
    2. Move the folder that hold system restore data to a different drive then C:.
    3. Limit the amount of system restore data.
    1, We can delete some system restore point.
    For more information, please take a look at the following article.
    http://windows.microsoft.com/en-hk/windows/delete-restore-point#1TC=windows-7
    2,We are not able to move the system restore folder to other Drive.
    3, We can limit the amount of system restore data.
    For more information, please take a look at the following article about managing the disk space that is used by System Restore.
    http://windows.microsoft.com/en-hk/windows7/how-much-disk-space-does-system-restore-require
    Best regards,
    Fangzhou CHEN
    Fangzhou CHEN
    TechNet Community Support

  • One more question about "snap to" points in audio objects

    sorry, one more question people.............. is there a way to set a "snap to" point in an audio region? Can I have logic set a point, or an "anchor" within an audio region that would coincide with the cursor. Once I get the cursor to a bar or a beat, is there a command that would allow a "snap to" point to be created in an audio object? That way I could clean up any "noise" or "chatter" at the beginning of an audio bounce or any audio object, and still have no problem pasting it all over my arrangement. I'm sure it's in here somewhere, but can't find it in the manual..
    thanks again!!!

    The anchor point in an audio region can be moved to wherever you want it to be. Let's say you recorded a loud piano bass note and afterwards you trimmed the file so that the attack of this sound happens right at the beginning of the audio region. Your anchor point will now coincide with the begnning of that region. This means that you can place that audio at, say, bar 37 and it's guaranteed that the attack of that sound will play exactly at bar 37. In this case anchor point = position of audio region in the Arrange Window.
    Even though you haven't played a MIDI note into the song in order to place this audio region, an "event" is generated for this sound which can be viewed in a special version of the Event Editor which shows the position of all audio regions in the arrangement. Access this by simply clicking on a blank area of the Arrange window and then opening the Event Editor.
    So let's say this sound is positioned at bar 37. Open the Event Editor (as described above) and you'll see an event at 37 1 1 1 and the name of that audio region to its right. In this case, the event's position = the start of the sound file = the position of the anchor within that sound file. If you changed the position of that event to 38, or 109, or 3, that audio region will be "triggered" at any of those positions. That sound will always "attack" on a downbeat.
    Now let's say you reversed that pinao note in order to create a dramatic swell into a section of your song. (Open the sample editor and use the Reverse function). At this point, the anchor point is still positioned at the beginning of the file! Play back the sequence (from our bar 37) and the audio will play back from bar 37, only backwards this time; the peak of that piano note will be at some point later than 37.
    So let's say you wanted to position this sound so that it hits exactly on the downbeat of bar 37. How do you do it? Well, you could always slide the audio file in the Arrange Window and try to line up the attack of the audio waveform with the downbeat of bar 37, but there's an easier way...
    Open the sample editor and move the anchor point to the end of the file, coinciding with the peak of the waveform (the piano's attack is now at the end because we reversed the audio). Next, close the sample editor window, click on a blank area in the Arrange Window and open the Event Editor. You'll see that this audio region is still positioned at bar 37, but the actual beginning of the sound file now starts before bar 37. Now when you play back, Logic will start playing that sound file before bar 37, but the audio's peak will be lined up at bar 37.

  • A question about transactions and point-of-time

    We have an operation which we want to serialize, since it can be called concurrently by same user from different web servers which would cause duplicates
    We have a Table 'Search'
    UserID
    <other fields>
    the Search table is updated in stored proc (INUserID, INSearchString) where INSearchString is a complete SELECT statement that can encompass many tables. Currently this is what the proc does:
    DELETE FROM Search WHERE UserID = INUserID;
    EXECUTE IMMEDIATE 'INSERT INTO Search SELECT INUserID,RowNum FROM (' || INSearchString || ')';
    I want to add a new table SearchLock (UserID)
    and add a SELECT * FROM SearchLock WHERE UserID = INUserID FOR UPDATE
    to this stored proc
    my question is, where exactly should I put the transaction BEGIN and COMMIT to insure that a second call, after waiting if needed, will see the contents of Search AFTER the first call has commited? would the following sequence work every time?
    BEGIN TRANSACTION ...
    SELECT ... FOR UPDATE
    DELETE...
    EXECUTE ...
    COMMIT
    (obviously will add en EXCEPTION handling)
    my worry, and what i'm trying to prevent, is that the second call, after waiting, will see the contents of Search BEFORE the first call has commited it

    To avoid this, just create a primary key for your
    table. If a row is currently inserted in a
    non-committed transaction, all other transactions
    that want to insert the same primary key in the same
    table will wait on the first transaction to complete.
    If the first transaction commits, all other
    transaction will get ORA-0001 error: unique contraint
    ... violated.
    If the first transaction rollbacks, one will be able
    to insert the primary key and the same process
    applies with other transactions.
    Oracle automatically locks the row with the primary
    key to be inserted for you.intriguing, but I am trying to protect several statements, I don't want the second sessin any chance to be able to run a DELETE.
    I am thinking the SELECT ... FOR UPDATE will have a NOWAIT option so that the second session will get immediate error notification.

  • W530, questions about adding additional RAM

    My current setup came with the factory optional 2x 8gb Samsung DDR3 1600mhz RAM. I recently purchased gskill ripjaws 8gb stick ( 1.35V model). My computer recognizes the ram is installed however mentions that the new ram is unusable.  I have the i7-3840QM 3rd Gen processor.
    Is the lower voltage causing the ram the culprit? Where else should i begin looking to allow my machine to be fully capable of using the full 24gb now installed?
    -Thanks
    -Concerned
    W530 i7-3840Qm. Quad K1000M. 16GB RAM. 16GB mSSD(expresscache). 500gb HDD

    I never buy memory that isn't guaranteed compatible with the machine by the vendor, which is why I always buy memory from Crucial.
    I also don't buy single-sticks to upgrade the total memory, but rather buy them in pairs (i.e. "kits") to guarantee dual-mode performance with identical matching size sticks in each pair of DIMM sockets.
    The Crucial site shows all 1.35v memory for the W530, so that's fine.  Memory latencies available from CL=9 to CL=11.
    Of course mixing latency memory speeds will just force the machine to run at the SLOWEST speed which seems undesirable, which again is another reason you should always match new memory latency with your older memory latency, or instead opt to upgrade all of your memory to newer faster "kits" of guaranteed matching identical performance pairs of cards.
    The Crucial site shows both PC3-12800 and PC3-14900 memory as compatible with the quad-core W530, ranging from CL=9 to CL=10 to CL=11.  So you have your choice, and price range.
    If your new single 8GB stick of memory has different latency characteristics from your existing 2x8GB factory memory, I guess the results should really just be that things run at the slower speed.  But to actually not even see the new single 8GB stick, well I don't know.
    Can you return the new memory and buy a replacement "kit" of 2 sticks (either 2x4GB or go with 2x8GB) which match the performance specs of your existing memory?

  • Question About Adding Ram from other laptop to T-series Thinkpad

    Hi,
    I have a question about adding additional RAM. Right now I have a 4gb ram stick in the t510. I have an acer laptop which I do not need and there is a 4GB ram stick onboard (maybe two 2gb ram sticks) there in that older laptop. Would it be possible to add one stick of 2gb ram (or 4gb i need to check if it is 2 or 1) from the acer to the lenovo thinkpad making the total ram in the thinkpad over 4gb????
    Thanks in advance

    Hi Richk,
    Yes, I am using a 64-bit operating system. I am running on windows 7. And as for reported incompatibilities....technically taking a RAM from another laptop and placing it in the laptop should be the same as purchasing a RAM card from ebay or something and putting into the laptop right?

  • Question about selecting vector paths in Photoshop

    Hi, I have a question about selecting path points in photoshop.
    This is the setup. I have a vector shape layer (the redi circle). but that's placed inside a group with a vector mask (the black pentagon). What I want to do is select parts of the red circle using click and drag) so I can modify portions of the circle).
    This is Photoshop's behavior as I've noticed it. Photoshop likes to select the shapes for you, and will target layers on its own when you use the Path Selection Tool. It tends to want to select the top most shape. If I target the shape layer first, I can use the Direct Selection tool to select individual points. I can add selected points bu shift+clicking.This is all well and good.
    If I try to select multiple points using click+drag though, it gets a bit weird.
    If I try to click+drag to select multiple points from within the vector mask (from the black), it will fail and select the vector mask. If i try to click+drag to select multiple points from outside the vector mask (from the white), then it will allow me to select points from my vector shape like i intended.
    I could also shift click to hide the vector mask, and it will work the way i want.
    Is there an easy way to tell photoshop not to auto target vector shapes? or a better way to select paths inside vector masks?
    Thanx for any input

    canvai wrote:
    Is there an easy way to tell photoshop not to auto target vector shapes? or a better way to select paths inside vector masks?
    Sure.
    Simply click once on the vector mask for the shape layer (red circle) to highlight that vector mask in the Layers Panel. Then when you click-drag the anchors of that path will be selected, not the group's path.
    If you have no mask highlighted, Photoshop assumes it should select the first path it comes in contact with.

  • HT4972 Good afternoon! I addressed in the AT&T company with a question about untying of a sim card of AT&T from my iPhone, all of them made Iphone now is unlocked, I executed all points which were written: To complete the unlock, simply  1 . Open iTunes o

    Good afternoon! I addressed in the AT&T company with a question about untying of a sim card of AT&T from my iPhone, all of them made Iphone now is unlocked, I executed all points which were written:
    To complete the unlock, simply
    1 . Open iTunes on your Mac or PC and verify that you have Internet connectivity.
    2 . Ensure a SIM card is inserted into your iPhone.
    3 . Connect your iPhone using the dock connector to USB cable that came with your iPhone.
    4 . Backup and restore your iPhone using iTunes. For information on backup and restore, please visit http://support.apple.com/kb/HT1414.
    5 . After restoring, your iPhone will be unlocked.
    Additional information on unlocking can be found at http://support.apple.com/kb/TS3198
    I made everything as well as was written, but on completion of the ios 6.1.3 installation the iPhone asks to be activated through wifi or through a wire to the computer.
    I connected Iphone to the computer and the text takes off: (Were sorry, we are unable to continue with your activation at this time. Please TR again later, or contact customer care! )
    how to me to activate my iPhone?
    I hope for your help....

    Hey APPLELovestory,
    We have an article that goes over troubleshooting iPhone activation issues here:
    iPhone: Troubleshooting activation issues
    http://support.apple.com/kb/TS3424
    Perform the following steps:
    Restart the iPhone.
    Try another means of reaching the activation server and attempt to activate.
    Try connecting to Wi-Fi if you're unable to activate using a cellular data connection.
    Try connecting to iTunes if you're unable to activate using Wi-Fi.
    Restore the iPhone.
    If you receive an alert message when you attempt to activate your iPhone, try to place the iPhone in recovery mode and perform a restore. If you're still unable to complete the setup assistant due to an activation error, contact Apple for assistance.
    Hope that helps,
    David

  • Question about points earned on gaming consoles

    I am confused about how the gamer's club unlocked works.  The Best Buy site says that my Best Buy members get 1 point per dollar spent.  The Gamer's Club Unlocked details says you get points on gaming consoles.
    I purchased an Xbox One Assassin's Creed bundle for $299.99 on 11/8/14 and 299 points are pending in my My Best Buy account.  That doesn't seem to me like I am getting any points for the console purchase; it seems like I am only getting points for the dollar amount spent in the store.  If this is the case, why even mention that the Gamer's Club Unlocked membership gives you points on gaming consoles?
    I purchased a cell phone and received the Gamer's Club Unlocked membership on 11/8/14.  I was told the membership was effective immediately.  This seems to be the case, since I received the 10% bonus discount when I traded in a video game during the same visit.  I purchased the Xbox One Assassin's Creed Video game bundle with a free Xbox live gold membership on 11/8/14 during the same store visit.  I actually made three transactions: one to buy the phone/gamer's club unlocked membership, one to trade in the video game, and one to buy the xbox one console bundle.  The cell phone/gamer's club membership transaction happened first.
    Should I have received additional points for buying a gaming console?

    Hi kre1997,
    Regrettably, no -- Gamers' Club members do not earn any additional points for video game consoles. My Best Buy Elite and Elite Plus members still receive their respective bonus points for these items though, as do My Best Buy Credit Card holders for purchasing them using their card. For more information though I'd encourage you to check out the two pages:
    BestBuy.com > My Best Buy
    http://bbyurl.us/MyBBY
    BestBuy.com > My Best Buy > Gamers' Club
    http://bbyurl.us/MyBBYGC
    I hope this helps.
    Aaron|Social Media Specialist | Best Buy® Corporate
     Private Message

  • Questions about Contracts and New Phones

    I've had some questions about how upgrading and adding new lines to a contract works and I haven't been able to find the answers through Google. First some background information: We started this most recent contract in September of 2011 and as such the current contract will end September of 2013. According to the phone information portal, all three of the devices on my account will be eligible for an upgrade this Saturday (May 4th, 2013). My daughters birthday just so happens to coincide with that date and as such we were planning on surprising her by taking her to buy a new phone and we are going to allow our son to upgrade his phone as he has been asking to for a while.
    Is there any way of locking/limiting the amount of data allowed for each phone on a shared plan? I'm worried that my daughter will go way over our limit and not pay attention to the point where she will end up costing us a very large amount of money
    Does adding a new line require the start of a new contract?
    My son was interested in purchasing the "Samsung Galaxy S4" and has noticed that pre-orders for that phone are currently available. Is it possible to pre-order the phone on that date using the upgrade so that the price will be reduced?
    If the answer to number 2 is a no, does adding the new line take away the upgrades from other phones or would he be able to go to the store past the release date and upgrade then?
    Thanks for taking the time to read through all of this. Any help that I receive will be extremely appreciated.

    1. Yes. There is an additional fee per line you wish to do this for. When you add the line you can choose this option and set the data amount it is allowed to use.
    2. Yes. Each line is a separate contract with its own expiration date.
    3. The contract starts when you sign up for the contract. The amount will be pro-rated on the plan if it is not started until part-way through your billing cycle that is what you are asking. Doing the pre-order just guarantees you the phone in case they sell out.
    4. No it will not take any upgrades away as each line has its own termination date. You can move upgrades around between lines though.

  • Questions about buying a new Mac Pro for 4k video editing.

    Hi everyone,
    I'm currently looking into buying a new mac pro and I have a few questions. I'm a filmmaker/freelance editor looking to get a system that can handle any/at least most 4k formats that I might throw at it, and will hopefully last me around 7 years or so, like my last mac pro has. I've saved up about $5,300 and am becoming more obsessed with getting it asap, but am willing to wait a bit and save up more if necessary. I also play the occasional elder scrolls or civilization game, and might run windows on the new system as well. So here are my questions:
    1. I've read rumors that a newer build could be released this year, with newer processors and graphics cards. Is there anything to point to when? I tend to buy things a month before a newer version is released, and I'd like to prevent myself from doing it this time around..
    Here's the Build I'm looking at:
    6 core
    2 D700s
    base ram to be upgraded myself to 32Gb (2x16Gb cards leaving 2 slots empty to expand to 64Gb later)
    512Gb-1Tb internal hd
    2. Should I be considering the 8 core? I'm not too excited about the additional $1500, but I want a system that will last.
    3. Is getting the two 16Gb chips of ram and leaving two slots empty a bad idea?
    4. I currently work with FCP studio 2 and love it. Not sure whether to go with FCP X, or adobe. Any thoughts?
    5. I'm not finding many deals for cheaper ram and hard drives. OWC's prices seem to be comparable to Mac's. I want to do the ram so I have room to upgrade to 64Gb later, but are there any hard drives out there that would make it worth upgrading it myself?
    I appreciate any insights you might have. I plan on getting a decent raid and 4k monitor in the next year or so, but for now just want a base system that will keep me editing and will be ready for 4k when I take that next step.

    The late 2013 Mc Pro uses Intel Xeon ECC processors (error correction), and as far as I know Intel has not announced any newer Intel Xeon processors than those in the late 2013 Mac Pro.  I would not expect to see an update to the 2013 Mac Pro until the end of 2015 at the earliest and probably later than that.
    If time is not an issue, then you should be quite happy with the 6 core 2013 Mac Pro.  It will do an excellent job with 4K video footage. And, yes, I would suggest getting the best raid system you can afford.  That is actually more important than processor speed since I/O is frequently the bottleneck when doing multi camera video or 4K video.
    I have the latest version of Adobe's Premiere Pro 2014  CC installed on my late 2013 Mac Pro and i have used it a bit without problems.  However, I find it much much slower to edit with than FCP X.  Also be advised that if you Google you will find several individuals on the Adobe Forums who purchased the late 2013 Mac Pro and have not been able to use it with Premiere Pro CC because of either a hardware incompatibility or software issues between Premiere Pro CC and BMD's Resolve.  It is quite possible that I have not experienced these problems because I have not made very demanding projects with Premiere Pro CC on my 2013 Mac Pro.
    I strongly recommend FCP X.  Apple released FCP X before it was ready, and many early users were unwilling to take the time to learn how to use this very different NLE which is not track based.  Apple has over the last 3 years since FCP X was released, issued more than 10 updates (all free), and the program is stable and blazingly fast.  I urge you to check out the FCP X training offered by Ripple Training and/or Larry Jordan. Both are inexpensive, and worth every cent.  Watch their training videos and you will be up to speed in FCP X in no time at all, and you will wish you had switched a long time ago.
    If you can afford the 1 TB of PCie internal flash storage on your Mac Pro, then by all means get it.  For me 1TB is well worth the cost.
    As far as editing 4K video, the format of the video will be important to the ease of editing.  For example, I am able to edit in its native format (XVAC S) several streams of 4K video form my Sony FDR-AX100 with no problems.  If I were editing Sony's XVAC format used in their professional 4K cameras, that might pose a problem that would require transcoding.  Similarly for other 4K formats. XVAC S is an easy format to edit natively because it is essentially a high bit rate h.264 format.
    Best of luck on whatever you decide to do, and happy editing.
    Tom

  • Basic questions about programing for J9 VM

    I need to create a java application to run on a Pocket PC 2003 OS and I'm limited to using IBM's Websphere J9 Virtual Machine. I'm new to using java in this capacity and therefore have lots of questions which I'm guessing are pretty basic. Despite my best efforts, I've found very little to help me online. Below are a couple questions I have that I'm hoping someone can help me with. Even better, if you know of any resources that could help me with these and other questions, I'd love to have them. Thanks in advance.
    1. I'm using standard AWT for the GUI but I'm having problems getting frames, dialogs, etc., to come up in anything less then full screen. setSize() and resize() have no effect, even if I extend the frame and override the setMinimumSize, setPreferredSize, setMaximumSize methods. What do I need to do to get a child window to appear in something less than full screen?
    2. I'd like to be able to add menu's to the buttom toolbar (with the keyboard) but have no idea how I would add something to it. Can someone point me in the right direction?
    3. When my application gets an iconified event I go ahead and call a System.exit() to exit the application. This doesn't kill the java VM though, which continues to run in the background (taking up plenty of memory). If I run my application using the J9w.exe so that it runs without the console, then not only does the VM continue to run, but I have no way to stop it (the running program list can't see it). Since each time I run my application it starts a new VM, it only takes a couple of times before I'm out of memory and have to do a soft reset of the PDA to make things right. Can I kill the VM when I kill my application?
    4. I learn best by looking at example code, but have been unable to find any code people are running on J9. I have the GolfScoreTracker installed and running on my PDA, but the jar file contains only the classes. Since it's supposed to be a demo, I had hoped that the source code would be included, is that hoping too much or is the source code available somewhere? In addition, if you know of any applications out there with available source code that are known to run well on a PocketPC J9 environment I'd appreciate the link.

    Hi there, I saw your questions, im currently developing a java pocket pc application as well and here's what i go so far:
    AWT seem to be hard to use on pocket pc, instead i recommend that you use SWT, which is a technology developed by eclipse. It's already installed on the eclipse plugins, you just need to download the jar and dll files for the pocket pc. They are available at: http://www.eclipse.org/downloads/index.php
    I assume that you already have the J9 working on your device. So you only need to copy the SWT.jar file and the swt-win32-3064.dll to the folder containing the .class files on your device.
    Now, on your java IDE(im using eclipse) Add the SWT.jar library to your project which should be on C:/eclipse/plugins/org.eclipse.swt.win32_3.0.2/ws/win32/swt.jar or something like taht depending on your eclipse root.
    Now you are ready to code some SWT, here is a sample program from Carolyn MacLeod, i modified a few things so it could run on the pocket pc but it's almost the same. Once you coded in eclipse run it, then copy the class file from your desktop to your device(There should be like 4 or 5 classes like sample.class, sample$1.class etc. copy all of them) where you put the jar and dll files(If program does not run you may try to rename the dll file for swt-win32-3050.dll instead of 3064.dll). Now after you have everything set up you need to create the lnk file in order to run the program. On your desktop create a new textfile and write the following on it:
    255#"\Archivos de Programa\J9\PPRO10\bin\j9w.exe" "-jcl:PPRO10" "-cp" "\My
    Documents\Java\swt.jar" ;\My Documents\Java" sample
    Paths depend on your install, and the place where you put your classes and jar file. path for j9w is usually "\Program Files\J9\PPRO10\bin\j9w.exe", and the last word should be the classname.
    Currently this form only displays an image on a canvas, click on the browse button and choose a pic and it will display on the canvas. Im trying to connect the form to an oracle database but no success yet on the device.
    Hope it helps, if you have any questions about the code or something, and I know the answer, please feel free to ask.
    See ya
    import org.eclipse.swt.*;
    import org.eclipse.swt.widgets.*;
    import org.eclipse.swt.layout.*;
    import org.eclipse.swt.events.*;
    import org.eclipse.swt.graphics.*;
    public class sample {
    static Shell shell;
    static Display display;
    static Text dogName;
    static Text textdb;
    static Combo dogBreed;
    static Canvas dogPhoto;
    static Image dogImage;
    static List categories;
    static Text ownerName;
    static Text ownerPhone;
    static String[] FILTER_EXTS = {"*.jpg"};
    public static void main(String[] args) {
    display = new Display();
    shell = new Shell(display, SWT.RESIZE | SWT.CLOSE);
    Menu menu = new Menu(shell, SWT.BAR);
    shell.setText("Dog ShowS Entry");
    shell.setMenuBar(menu);
    GridLayout gridLayout = new GridLayout();
    gridLayout.numColumns = 3;
    shell.setLayout(gridLayout);
    new Label(shell, SWT.NONE).setText("Dog's NameS:");
    dogName = new Text(shell, SWT.SINGLE | SWT.BORDER);
    GridData gridData = new GridData(GridData.HORIZONTAL_ALIGN_FILL);
    gridData.horizontalSpan = 2;
    gridData.widthHint = 60;
    dogName.setLayoutData(gridData);
    new Label(shell, SWT.NONE).setText("Breed:");
    dogBreed = new Combo(shell, SWT.NONE);
    dogBreed.setItems(new String [] {"Collie", "Pitbull", "Poodle", "Scottie"});
    dogBreed.setLayoutData(new GridData(GridData.HORIZONTAL_ALIGN_FILL));
    Label label = new Label(shell, SWT.NONE);
    label.setText("Categories");
    label.setLayoutData(new GridData(GridData.HORIZONTAL_ALIGN_FILL));
    new Label(shell, SWT.NONE).setText("Photo:");
    dogPhoto = new Canvas(shell, SWT.BORDER);
    gridData = new GridData(GridData.FILL_BOTH);
    gridData.widthHint = 60;
    gridData.verticalSpan = 3;
    dogPhoto.setLayoutData(gridData);
    dogPhoto.addPaintListener(new PaintListener() {
    public void paintControl(final PaintEvent event) {
    if (dogImage != null) {
    event.gc.drawImage(dogImage, 0, 0);
    categories = new List(shell, SWT.MULTI | SWT.BORDER | SWT.V_SCROLL);
    categories.setItems(new String [] {
    "Best of Breed", "Prettiest Female", "Handsomest Male",
    "Best Dressed", "Fluffiest Ears", "Most Colors",
    "Best Performer", "Loudest Bark", "Best Behaved",
    "Prettiest Eyes", "Most Hair", "Longest Tail",
    "Cutest Trick"});
    gridData = new GridData(GridData.HORIZONTAL_ALIGN_FILL | GridData.VERTICAL_ALIGN_FILL);
    gridData.widthHint = 60;
    gridData.verticalSpan = 4;
    int listHeight = categories.getItemHeight() * 12;
    Rectangle trim = categories.computeTrim(0, 0, 0, listHeight);
    gridData.heightHint = trim.height;
    categories.setLayoutData(gridData);
    Button browse = new Button(shell, SWT.PUSH);
    browse.setText("BrowsePic");
    gridData = new GridData(GridData.HORIZONTAL_ALIGN_FILL);
    gridData.widthHint = 60;
    browse.setLayoutData(gridData);
    browse.addSelectionListener(new SelectionAdapter() {
         public void widgetSelected(SelectionEvent event) {
              FileDialog dialog = new FileDialog(shell, SWT.OPEN);
              dialog.setFilterExtensions(new String[] {"*.*"});
              String fileName = dialog.open();
    if (fileName != null) {
    dogImage = new Image(display, fileName);
    shell.redraw();
    Button delete = new Button(shell, SWT.PUSH);
    delete.setText("No action");
    gridData = new GridData(GridData.HORIZONTAL_ALIGN_FILL | GridData.VERTICAL_ALIGN_BEGINNING);
    gridData.widthHint = 60;
    delete.setLayoutData(gridData);
    delete.addSelectionListener(new SelectionAdapter() {
    public void widgetSelected(SelectionEvent event) {
    Button enter = new Button(shell, SWT.PUSH);
    enter.setText("No action");
    gridData = new GridData(GridData.HORIZONTAL_ALIGN_END);
    gridData.horizontalSpan = 3;
    enter.setLayoutData(gridData);
    enter.addSelectionListener(new SelectionAdapter() {
    public void widgetSelected(SelectionEvent event) {
         shell.open();
         while (!shell.isDisposed()) {
              if (!display.readAndDispatch())
                   display.sleep();
         display.dispose();
    if (dogImage != null) {
    dogImage.dispose();
    }

  • Question about servlet architecture

    Hello,
    I have a question about Web Application / servlet architecture and development, and haven't had any luck finding an answer so far. My plan must be either unusual, or wrong, or both, so I'm hoping the community here could point me in the right direction.
    I have quite a bit of background in OSGi, but very little in servlets. I know my architecture would work with something such as sever-side Equinox, but for the open source project I'm contemplating, I think that the more mainstream Tomcat/servler route would be better -- assuming it can be done. Here is a rather simplified scenario:
    There is a class, Manager. It could be a Java class that implements Runnable, or a non-HTTP servlet (GenericServlet?), but somehow it should be configured such that it runs as soon as Tomcat starts.
    When Manager runs, it instantiates its primary member variable: Hashtable msgHash; mshHash will map Strings (usernames) to ArrayList<String> (list of messages awaiting that user). Manager also has two primary public methods: addMsg() and getMsgs().
    The pseudocode for addMsg is:
    void addMsg(String username, String msg) {
    if username=="*" { for each msgList in msgHash { msgList.add(msg) } return;
    if username not in msgHash { msgHash.add(username, new ArrayList<String>) }
    msgHash[username].add(msg)
    And for getMsgs():
    String getMsgs(username) {
    String s = "";
    for each msg in msgHash[username] { s = s + msg; }
    return s;
    Once Manager has started, it starts two HTTP Servlets: Sender and Receiver, passing a reference back to itself as it creates the servlets.
    The HTML associated with Sender contains two text inputs (username and message) and a button (send). When the send button is pressed and data is POSTed to Sender, Sender takes it and calls Manager.addMsg(username.value, message.value);
    Receiver servlet is display only. It looks in the request for HTTP GET parameter "username". When the request comes in, Receiver calls Manager.getMsgs(username) and dumps the resulting string to the browser.
    In addition to starting servlets Sender and Receiver, the Manager object is also spawning other threads. For example, it might spawn a SystemResourceMonitor thread. This thread periodically wakes up, reads the available disk space and average CPU load, calls Manager.addMsg("*", diskAndLoadMsg), and goes back to sleep for a while.
    I've ignored the synchronization issues in this example, but during its lifecycle Manager should be idle until it has a task to do, using for example wait() and notify(). If ever addMsg() is called where the message string is "shutdown", Manager's main loop wakes up, and Manager shuts down the two servlets and any threads that it started. Manager itself may then quit, or remain resident, awaiting a command to turn things back on.
    Is any of this appropriate for servlets/Tomcat? Is there a way to do it as I outlined, or with minor tweaks? Or is this kind of application just not suited to servlets, and I should go with server-side Equinox OSGi?
    Thanks very much for any guidance!
    -Mike

    FrolfFla wrote:
    Hello,
    I have a question about Web Application / servlet architecture and development, and haven't had any luck finding an answer so far. My plan must be either unusual, or wrong, or both, so I'm hoping the community here could point me in the right direction.
    I have quite a bit of background in OSGi, but very little in servlets. I know my architecture would work with something such as sever-side Equinox, but for the open source project I'm contemplating, I think that the more mainstream Tomcat/servler route would be better -- assuming it can be done. Here is a rather simplified scenario:
    There is a class, Manager. It could be a Java class that implements Runnable, or a non-HTTP servlet (GenericServlet?), but somehow it should be configured such that it runs as soon as Tomcat starts.You can configure a servlet to be invoked as soon as Tomcat starts. Check the tomcat documentation for more information on that (the web.xml contains such servlets already though). From this servlet you can start your Manager.
    >
    When Manager runs, it instantiates its primary member variable: Hashtable msgHash; mshHash will map Strings (usernames) to ArrayList<String> (list of messages awaiting that user). Manager also has two primary public methods: addMsg() and getMsgs().
    The pseudocode for addMsg is:
    void addMsg(String username, String msg) {
    if username=="*" { for each msgList in msgHash { msgList.add(msg) } return;
    if username not in msgHash { msgHash.add(username, new ArrayList<String>) }
    msgHash[username].add(msg)
    And for getMsgs():
    String getMsgs(username) {
    String s = "";
    for each msg in msgHash[username] { s = s + msg; }
    return s;
    Once Manager has started, it starts two HTTP Servlets: Sender and Receiver, passing a reference back to itself as it creates the servlets. Not needed. You can simply create your two servlets and have them invoked through web calls, Tomcat is the only one that can manage servlets in any case. Servlets have a very short lifecycle - they stay in memory from the first time they are invoked, but the only time that they actually do something is when they are called through the webserver, generally as the result of somebody running the servlet through a webbrowser, either by navigating to it or by posting data to it from another webpage. The servlet call ends as soon as the request is completed.
    >
    The HTML associated with Sender contains two text inputs (username and message) and a button (send). When the send button is pressed and data is POSTed to Sender, Sender takes it and calls Manager.addMsg(username.value, message.value);
    Receiver servlet is display only. It looks in the request for HTTP GET parameter "username". When the request comes in, Receiver calls Manager.getMsgs(username) and dumps the resulting string to the browser.This states already that you are going to run the servlets through a browser. You run "Sender" by navigating to it in the browser. You run "Receiver" when you post your data to it. Manager has nothing to do with that.
    >
    In addition to starting servlets Sender and Receiver, the Manager object is also spawning other threads. For example, it might spawn a SystemResourceMonitor thread. This thread periodically wakes up, reads the available disk space and average CPU load, calls Manager.addMsg("*", diskAndLoadMsg), and goes back to sleep for a while.Since you want to periodically run the SystemResourceMonitor, it is better to use a Timer for that in stead of your own thread.
    >
    I've ignored the synchronization issues in this example, but during its lifecycle Manager should be idle until it has a task to do, using for example wait() and notify(). If ever addMsg() is called where the message string is "shutdown", Manager's main loop wakes up, and Manager shuts down the two servlets and any threads that it started. Manager itself may then quit, or remain resident, awaiting a command to turn things back on.
    Is any of this appropriate for servlets/Tomcat? Is there a way to do it as I outlined, or with minor tweaks? Or is this kind of application just not suited to servlets, and I should go with server-side Equinox OSGi?
    Thanks very much for any guidance!
    -MikeI don't see anything that cannot be done in a normal java web environment, as long as you are aware of how servlets operate.

Maybe you are looking for