Character reordering problem in swing text components on macintosh

I have created a swing application that handles devanagari script.(script for Hindi language).The application works perfectly in windows but in apple macintosh
there is a problem with the reordering of characters in text componets. Reordering means some characters needs to be displayed not in the exact order in which they are stored.I was wondering how this can happen as the application works fine on windows.Also other applications like text editor displays devanagari script correctly on macintosh.
I tried with several fonts like Arial Unicode MS,Devanagari MT etc/.java

Very odd. Post your Reordering code.

Similar Messages

Arabic numerals not displaying in Swing text components

On a test system setup with Windows NT Arabic version, we had problems entering Arabic numerics into Swing text components.
Other Arabic characters were displayed correctly, but numerals were all represented by Western characters ('0 to 9'). Even when using the "Alt-" method to enter the Unicode character codes via. the keypad, we still got incorrect characters.
When using the keypad to enter the code '\u0660' (Arabic "1"), the code sent to the text component was '\u2022' (Bullet).
We tried this on an English system and the same code was passed to the text component.
Does anyone have any suggestions as to what might be the cause of this problem ?
Thanks,
Brian...

hello!
I am developing an arabic editor but i am unable to display the arbic numeric in swing text .Plz help me on this topic and send some code example on this topic.
I have another problem about events i.e how can i handle the beckspace event, enter event, and spaceevent?
plz send me mail .
Thanx for advance.

Text components

Hello,
I thought copy/paste functions (ctrl v, ctrl c) were automatically included in Swing Text components but when they are on a JApplet, they don't always work. The keyword is always which is weird because on some computers it works, but others it only will copy/paste text internally in the applet (not from external programs). This is not a signed applet...but the copy and paste from external sources DO work on some computers, so I'm not sure if this is a permissions issue or not. Does anyone know any solutions?
Thanks.

Thank you camickr.
addHighlight() highlights a range of characters and this is exactly what I needed. What about highlighting lines. To highlight a line, I will have to count all characters before that line to get to line start, and then count all characters before next "\n" to get to line end. Then repeat this process for all lines that I need to highlight. Is there an easier way to highlight lines ?

Text input in text components in swing?

Hi,
is there any UML diagram, or any site on how text input is carried out in text components? Is it just a simple matter of JTextComponent adding itself as a KeyListener to itself, and then updating the document with each key press? or is there a little more too it?
thanks,
J

justinlawler wrote:
or is there a little more too it?There's a lot more to it. I think good beginning reading is the summary at the top of the [Javadoc for JTextComponent|http://java.sun.com/javase/6/docs/api/javax/swing/text/JTextComponent.html] . It is a convoluted API that has many "kludges" that can't be completely removed because of the need for backwards compatibility.
I'd also recommend reading up on the classes EditorKit, InputMap, ActionMap, Keymap, and (Basic)TextUI. Good luck!

[svn:fx-trunk] 10214: This fixes the problem that if two text components share the same textFlow there is an infinite loop involving updateDisplayList - damageHandler - invalidateDisplaylist - back to updateDisplayList.

Revision: 10214
Author:   [email protected]
Date:     2009-09-13 07:33:58 -0700 (Sun, 13 Sep 2009)
Log Message:
This fixes the problem that if two text components share the same textFlow there is an infinite loop involving updateDisplayList -> damageHandler -> invalidateDisplaylist -> back to updateDisplayList. The bug file was for TextArea which is RET but the same bug was in RichText as well.
This example with a renderer exposed it because the typicalItem that is composed to figure out sizes and the actual first item in the list share the same textFlow. It actually has nothing to do with useVirtualDisplay other than it was sharing a textFlow.
It turns out that the TextFlowFactory dispatches damage events every time the textFlow is composed. Unlike when the flowComposer is used, it always considers the flow damaged. It was exacerbated by each of the two components having a damage handler for the same textFlow.
The solution is to use the textFlow generation number. In the damageHandler if the generation is the last known generation number, assume no changes, and return immediately from the damage handler.
QE notes: There are 1 TextArea, 6 TextInput and 2 NumericStepper failuers, with or without my changes. The common link seems to be DispatchKeyEvent. Most were testing maxChar, displayAsPassword and restrict. I tested these and they seem to be working correctly.
Doc notes:
Bugs: SDK-23002
Reviewer: Gordon
Tests run: checkintests, TextArea, TextInput and NumericStepper
Is noteworthy for integration: no
Ticket Links:
    http://bugs.adobe.com/jira/browse/SDK-23002
Modified Paths:
    flex/sdk/trunk/frameworks/projects/spark/src/spark/components/RichEditableText.as
    flex/sdk/trunk/frameworks/projects/spark/src/spark/components/RichText.as
    flex/sdk/trunk/frameworks/projects/spark/src/spark/components/supportClasses/RichEditable TextContainerManager.as

Revision: 10214
Author:   [email protected]
Date:     2009-09-13 07:33:58 -0700 (Sun, 13 Sep 2009)
Log Message:
This fixes the problem that if two text components share the same textFlow there is an infinite loop involving updateDisplayList -> damageHandler -> invalidateDisplaylist -> back to updateDisplayList. The bug file was for TextArea which is RET but the same bug was in RichText as well.
This example with a renderer exposed it because the typicalItem that is composed to figure out sizes and the actual first item in the list share the same textFlow. It actually has nothing to do with useVirtualDisplay other than it was sharing a textFlow.
It turns out that the TextFlowFactory dispatches damage events every time the textFlow is composed. Unlike when the flowComposer is used, it always considers the flow damaged. It was exacerbated by each of the two components having a damage handler for the same textFlow.
The solution is to use the textFlow generation number. In the damageHandler if the generation is the last known generation number, assume no changes, and return immediately from the damage handler.
QE notes: There are 1 TextArea, 6 TextInput and 2 NumericStepper failuers, with or without my changes. The common link seems to be DispatchKeyEvent. Most were testing maxChar, displayAsPassword and restrict. I tested these and they seem to be working correctly.
Doc notes:
Bugs: SDK-23002
Reviewer: Gordon
Tests run: checkintests, TextArea, TextInput and NumericStepper
Is noteworthy for integration: no
Ticket Links:
    http://bugs.adobe.com/jira/browse/SDK-23002
Modified Paths:
    flex/sdk/trunk/frameworks/projects/spark/src/spark/components/RichEditableText.as
    flex/sdk/trunk/frameworks/projects/spark/src/spark/components/RichText.as
    flex/sdk/trunk/frameworks/projects/spark/src/spark/components/supportClasses/RichEditable TextContainerManager.as

Problem with adding text to a syntax document

Hello,
I am trying to write an editor with syntax highlighting option.
The syntax document works fine but on a slow computer
after loading a huge document it is impossible to write fluently.
I think the method highlightKeyword() is not the problem because it highlights still only one line.
The problem seems to be the rootElement.Has anybody an idea how to make the rootElement faster
working?
I also tried to use threads (such as http://ostermiller.org/syntax/editor.html) for inserting text, but then the document is highlighted very slowly.
Can anybody help me?
import javax.swing.*;
import javax.swing.event.*;
import javax.swing.text.*;
import java.util.*;
import java.awt.*;
import java.awt.event.*;
public class SyntaxTest extends DefaultStyledDocument
     Element rootElement;
     String wordSeparators = ",:;.()[]{}+-/*%=<>!&|~^@? ";
     Vector keywords = new Vector();;
     SimpleAttributeSet keyword;
     public SyntaxTest()
          rootElement= this.getDefaultRootElement();
          keyword = new SimpleAttributeSet();
          StyleConstants.setForeground(keyword, Color.blue);
          keywords.add( "abstract" );
          keywords.add( "boolean" );
          keywords.add( "break" );
          keywords.add( "byte" );
          keywords.add( "case" );
          keywords.add( "catch" );
          keywords.add( "char" );
          keywords.add( "class" );
          keywords.add( "continue" );
          keywords.add( "default" );
          keywords.add( "do" );
          keywords.add( "double" );
          keywords.add( "else" );
          keywords.add( "extends" );
          keywords.add( "false" );
          keywords.add( "final" );
          keywords.add( "finally" );
          keywords.add( "float" );
          keywords.add( "for" );
          keywords.add( "if" );
          keywords.add( "implements" );
          keywords.add( "import" );
          keywords.add( "instanceof" );
          keywords.add( "int" );
          keywords.add( "interface" );
          keywords.add( "long" );
          keywords.add( "native" );
          keywords.add( "new" );
          keywords.add( "null" );
          keywords.add( "package" );
          keywords.add( "private" );
          keywords.add( "protected" );
          keywords.add( "public" );
          keywords.add( "return" );
          keywords.add( "short" );
          keywords.add( "static" );
          keywords.add( "super" );
          keywords.add( "switch" );
          keywords.add( "synchronized" );
          keywords.add( "this" );
          keywords.add( "throw" );
          keywords.add( "throws" );
          keywords.add( "true" );
          keywords.add( "try" );
          keywords.add( "void" );
          keywords.add( "volatile" );
          keywords.add( "while" );
     public void insertString(int offset, String str, AttributeSet a) throws BadLocationException
          super.insertString(offset, str, a);
          int startOfLine = rootElement.getElement(rootElement.getElementIndex(offset)).getStartOffset();
          int endOfLine = rootElement.getElement(rootElement.getElementIndex(offset+str.length())).getEndOffset() -1;
          //highlight the changed line
          highlightKeyword(this.getText(0,this.getLength()), startOfLine, endOfLine);
     public void remove(int offset, int length) throws BadLocationException
          super.remove(offset, length);
          int startOfLine = rootElement.getElement(rootElement.getElementIndex(offset)).getStartOffset();
          int endOfLine = rootElement.getElement(rootElement.getElementIndex(offset+length)).getEndOffset() -1;
          //highlight the changed line
          highlightKeyword(this.getText(0,this.getLength()), startOfLine, endOfLine);
     public void highlightKeyword(String content,int startOffset, int endOffset)
          char character;
          int tokenEnd;
          while (startOffset < endOffset)
               tokenEnd = startOffset;
               //find next wordSeparator
               while(tokenEnd < endOffset)
               character = content.charAt(tokenEnd);
                    if(wordSeparators.indexOf(character) > -1)
                         break;
                    else
                         tokenEnd++;
               //check for keyword
     String token = content.substring(startOffset,tokenEnd).trim();
                    if(keywords.contains(token))
               this.setCharacterAttributes(startOffset, token.length(), keyword, true);
     startOffset = tokenEnd+1;
     public static void main(String[] args)
          JFrame f = new JFrame();
          JTextPane pane = new JTextPane(new SyntaxTest());
          JScrollPane scrollPane = new JScrollPane(pane);
          f.setContentPane(scrollPane);
          f.setSize(600,400);
          f.setVisible(true);

I don't think the root element is the problem. Syntax highlighting is just plain slow for a large file. Here is an example, you can use for comparison purposes, that does [url http://www.discoverteenergy.com/files/SyntaxDocument.java]syntax highlighting. It is very slow when loading large files into the text pane, but works reasonably when modifying the contents of the text pane.

Hindi characters' display in swing components on macintosh

I have created a swing application that handles devanagari script.(script for Hindi language).The application works perfectly in windows but in apple macintosh
there is a problem with the reordering of characters in text componets. Reordering means some characters needs to be displayed not in the exact order in which they are stored.I was wondering how this can happen as the application works fine on windows.Also other applications like text editor displays devanagari script correctly on macintosh.
I tried with several fonts like Arial Unicode MS,Devanagari MT etc/.java

Try the method setFont(java.awt.Font f) of the class java.awt.Container (which java.applet.Applet extends) with suitable fonts. This doesn't work for 1.1. Only logical fonts name are allowed here.
You need either to edit font.properties file for corresponding logical font or for MS JVM use com.ms.awt.FontX (get it using Java reflections if you don't have MS SDK) as described at http://www.onjava.com/pub/a/onjava/2001/04/12/internationalization.html?page=1
or http://www.jadcentral.com/newscentral/feature.jsp?feature_ID=18

Devanagari characters' display in swing components on macintosh.

I have created a swing application that handles devanagari script.(script for Hindi language).The application works perfectly in windows but in apple macintosh
there is a problem with the reordering of characters in text componets. Reordering means some characters needs to be displayed not in the exact order in which they are stored.I was wondering how this can happen as the application works fine on windows.Also other applications like text editor displays devanagari script correctly on macintosh.
I tried with several fonts like Arial Unicode MS,Devanagari MT etc/.java

Hi Ken,
You'd be reading the HTML into your Java program using an InputStream of some description.
You can set the encoding of this stream by using an InputStreamReader to pull everything in through the stream. e.g.
InputStreamReader isr = new InputStreamReader(myInputStream, "UTF-8");
The first parameter is your input stream, the second is the encoding that input stream uses.
The HTML may be in an encoding called "ISO-8859-1", a Windows favourite, or it may be in a Chinese encoding like those I've mentioned (I can't recall the exact encoding strings you'd use in the call for the Chinese ones, but I believe GBK's is "GB-2312" or something similar).
From there, it's easy to use your HTML encoding specific InputStreamReader to get a bunch of chars or bytes. These will be in the same encoding again. Then, it's just a matter of creating a new String with a better encoding (one that Java prefers, and one that Chinese can display in).
The easiest way is to do the following:
String myConvertedString = new String(myHTMLBytes, "UTF-8");
... where myHTMLBytes is an array of bytes (byte[]) that have been pulled in through the InputStreamReader. The String should be nice and easy to use from there. A good test would be to create a JLabel with the string as its text. Remember to set the JLabel's font to ArialUnicodeMS, and it should work OK. If it doesn't first up, try "UTF-16" in place of "UTF-8" in the new string declaration.
In addition, Sun was actually shipping its own Java web browser classes with the JDK for a while. They were unofficial JDK code (i.e. they existed in a com.sun.* package), but they did get shipped. I haven't seen them in a JDK of late, but I know it was an early Unix version, so it might be worth looking for. AFAIK, the classes worked pretty well.
Hope that helps!
Martin Hughes

Limit the number of characters in Text Components

I'm trying to limit the number of characters in text components. So I use the follow KeyListener:
private class KeyHandler extends KeyAdapter{
String oldText;
private KeyHandler(){ }
public void keyTyped(KeyEvent ev){
String newText = textField.getText();
System.out.println("Old text: " + this.oldText); //debug
System.out.println("New text: " + newText); //debug
if(newText.length() < limit){
this.oldText = newText + ev.getKeyChar();
}else{
textField.setText(this.oldText);
if(textField.getCaretPosition() + 1 > limit)
textField.setCaretPosition(limit - 1);
But the problem is that after that, the last key is typed in the component and the last key always appear.
�Does anyone know the way for consume the key presses?
Thank you

(Belongs in the Swing forum.)
The callback method keyTyped comes after keyPressed and keyReleased.
The event.consume() should consume the event.
Consider cut/copy/paste.
Maybe a addPropertyChangeListener would be better?

Problem in creating text file from report file

Hello Everybody...
I have problem in creating text file.
I had set System Parameter as below
DESTYPE : File
MODE : Character
DESNAME : gayu.txt
And ruler setting
Units : Character Cells
Character Cell Size : Horiziontal = 7
Vertiacle =12
GridSpacing : 1
no of snap points per grid spacing : 1
but when i run report it will give error
"REP_1219 M_2 or R_2 has no size -- length or width zero"
so is there any solution of that
or another way to convert report into text file?....

Hi Folks,
Please don't go after that Error as its a very deceptive one. Please don't go for altering your design Or changing the size of items in your layout.
Similar errors I have come across in reports. I think its due to file transfer type. You might have transferred the file in Binary and your File Mode I am seeing it as CHAR.
This normally happens while making the Report Format as "Text". Either you change that format which I think is possible in your case as your basic requirement is making a report output of CHAR type.
Please check for this error in Metalink and proceed accordingly. I could fix similar errors when I changed Format type to PDF(from text).
Regards
Shibu

Javax.swing.text.TableView bug in 1.4 beta 2 ?

Hi,
I derived a class from TableView, which worked fine with 1.3, but now causes the following exception as soon as being layouted:
java.lang.Error: should not happen: class com.lexetius.swing.text.LexTableView
at javax.swing.text.BoxView.baselineLayout(Unknown Source)
at javax.swing.text.ParagraphView$Row.layoutMinorAxis(Unknown Source)
at javax.swing.text.BoxView.setSpanOnAxis(Unknown Source)
at javax.swing.text.BoxView.layout(Unknown Source)
at javax.swing.text.BoxView.setSize(Unknown Source)
at javax.swing.text.BoxView.updateChildSizes(Unknown Source)
at javax.swing.text.BoxView.setSpanOnAxis(Unknown Source)
at javax.swing.text.BoxView.layout(Unknown Source)
at javax.swing.text.FlowView.layout(Unknown Source)
at javax.swing.text.BoxView.setSize(Unknown Source)
at javax.swing.text.BoxView.updateChildSizes(Unknown Source)
at javax.swing.text.BoxView.setSpanOnAxis(Unknown Source)
at javax.swing.text.BoxView.layout(Unknown Source)
at javax.swing.text.BoxView.setSize(Unknown Source)
at javax.swing.plaf.basic.BasicTextUI$RootView.setSize(Unknown Source)
at javax.swing.plaf.basic.BasicTextUI.getPreferredSize(Unknown Source)
at javax.swing.JComponent.getPreferredSize(Unknown Source)
at javax.swing.JEditorPane.getPreferredSize(Unknown Source)
at javax.swing.ScrollPaneLayout.layoutContainer(Unknown Source)
at java.awt.Container.layout(Unknown Source)
at java.awt.Container.doLayout(Unknown Source)
at java.awt.Container.validateTree(Unknown Source)
at java.awt.Container.validate(Unknown Source)
at javax.swing.RepaintManager.validateInvalidComponents(Unknown Source)
at javax.swing.SystemEventQueueUtilities$ComponentWorkRequest.run(Unknow
n Source)
at java.awt.event.InvocationEvent.dispatch(Unknown Source)
at java.awt.EventQueue.dispatchEvent(Unknown Source)
at java.awt.EventDispatchThread.pumpOneEventForHierarchy(Unknown Source)
at java.awt.EventDispatchThread.pumpEventsForHierarchy(Unknown Source)
at java.awt.EventDispatchThread.pumpEvents(Unknown Source)
at java.awt.EventDispatchThread.pumpEvents(Unknown Source)
at java.awt.EventDispatchThread.run(Unknown Source)
any ideas???

I wrote the com.lexetius.swing.text.LexTableView component myself. It is derived directly from javax.swing.text.TableView and it adds virtually no functionality. I'm quite sure, the problem comes from the TableView class, but I can't verify that directly, since this class is abstract (why is that, anyway??).
It would certainly help me very much, if anybody had a component, which is also derived from javax.swing.text.TableView and works with the 1.4 beta.

Character display problem

Hi All,
When ever capital 'w' (i.e W, plain,Dialog and size 11) is first character in a text field left upper portion is not visible.But if I change any of these properties( i.e plain,Dialog, size 12) there is no problem.
Can any body suggest me how can I display 'W' correctly with these propeties(i.e W, plain,Dialog and size 11).
Thanks in advance

I think after all it is a problem with the either:
a. OS installer, which should really try to delete fonts which are duplicated by new system fonts
b. OS which should give precedence to new system fonts over old system fonts where there are two the same, especially when the system fonts are in /System/Library/Fonts and the old ones are in /Library/Fonts
The major annoyance is that this problem appeared with Leopard and took me hours of searching to figure out its origin - when you don't know if it is a character entry problem (which I thought) or a character display problem (which it turned out to be). In the end the fix was simple - delete the superceded fonts.

Character Converting Problem at File Adapter

I am having a character conversion problem bringing in a XML file into XI.
It is converting the data wrong.
I have a file adapter picking up Audit.xml encoding is utf-8, use binary mode to pick it up.
XI takes the file and converts it into a text file.
I compared the hex of Audit.xml source file and the Audit.txt file and the characters are different for a specific few.
See attached image for details
Please let me know if any additional information will help
http://i9.photobucket.com/albums/a68/tkc204/SAP%20XI/Audit.png

Actually for this case Christian is right the source xml that I am receiving wasn't valid UTF-8. The program generating the xml was just putting the encoding - utf-8 tag on top because that's what the code told it to do.
The actual data is Windows-1252 in xml. So when XI reads it, it uses the utf-8 tag on top and is expecting UTF-8, but it was getting something else causing the unsupported characters to get converted.
The HEX was different on the Source XML comparing it to what XI was generating after File Adapter.
After making the output change to Windows 1252. It changes back to the expect HEX.
Source - is at Windows 1252
XI - it will use UTF-8 (no way to change it)
Target - forced to create in Windows 1252
Other helpful tools I used: W3C Validator - upload the file and found that I wasn't receiving UTF-8
I hope this is clear and helps someone else in the future.
Thanks For everyone's input.
Chirag

Using of javax.swing.text package

Where can I find any resources devoted to using javax.swing.text package?

At the end of the API documentation for that package you may notice this:
<quote>
Related Documentation
For overviews, tutorials, examples, guides, and tool documentation, please see:
Using Text Components, a section in The Java Tutorial.
</quote>

Swing text wrapping

Hello,
I have an app thats makes use of an extended WrappedPlainView to add syntax highlighting. Everything works great except when I use bold fonts in random spots. When WrappedPlainView is asked to draw the text
swing does not allow enough room for the line with bold words.With bold words the line needs to be a bit
wider at times to not cut off any letters. Any ideas on how I can fix this ? On a side note , works fine
with fixed width fonts..
Thanks,
jd

I've actually done this before. Instead of subclassing WrappedPlainView though, I subclassed BoxView (WrappedPlainView's parent class). This had to be done because WrappedPlainView has too much data hidden via package private/private methods, to the point where you cannot modify it for your multi-font needs. Besides, WrappedPlainView states in its documentation that it only supports rendering a single font. Hacking it to support multiple fonts (plain, bold, italic) may have unforseen consequences. Subclass BoxView instead.
It's a lot of work to do what you want to do, probably more than you realize. You have to write code to calculate where the wrapped lines should be broken apart (must be fast too), override modelToView/viewToModel to take into account different fonts, etc. Use WrappedPlainView as a guide, but change the parts that do things such as compute break locations, model-to-view translations, etc. to handle your font size data.
But in all seriousness, you probably should use JEditorPane for this. Here's my attempt to talk you into using one :)
mohadib_ wrote:
Also JEditorPaneWhat are you using to render a multi-font view, if not JEditorPane? JTextArea? You do realize that JTextArea and its associated views are designed to render only 1 font? While this can be done, it's not simple, as several parts of JTextArea and its associated views make assumptions about using a single font for performance reasons. You'll have to seriously study the Swing text package code, and make your own View subclasses for JTextArea that handle rendering multiple fonts, computing text widths, etc.
is horribly slowOf course it's slower than rendering a single font with JTextArea. Again, that's because JTextArea can take advantage of the fact that it's using a single font, which allows it to greatly speed up modelToView/viewToModel calculations, rendering, etc. JEditorPane is not slow because it's poorly designed or inefficient; rather, it's slow because of the extra work it's doing. It's mature enough to the point where it's probably difficult to make it faster without a redesign. Do you honestly think you can create a Swing text component that has all of the features of JEditorPane, but performs better?
and eats memeory.It consumes more memory than, say, JTextArea, because it's caching the fonts/colors used to render its document. If you want to support multiple fonts and syntax highlighting, you'll have to do the same thing. Can you cache this information in a more efficient way than JEditorPane does? Have you tried using JEditorPane and run into memory problems?
Again, this can be done, but really shouldn't. The perceived benefits don't really pan out, especially when weighed against the development and maintenance costs. More code => more bugs. Why not use the support already provided to you in the JDK?

Character reordering problem in swing text components on macintosh

Similar Messages

Maybe you are looking for