The project now I am working on is the internal project at ACE group of Sun
Microsystems Inc. Canada, and it needs to use Star Office 5.2 API which I got
from the "star office development kit v1.0(sdk 1.0)", to communicate with the
Star Office documents. 

I am currently experiencing some problems with the soffice 5.2 API. Mr. Michael
Honnig refers me to report bugs here.

I am writing a java program that need to use star office v5.2 API to search and
replace texts in the text document. In the text document, it contains 20 OLE
objects, and each OLE object is Spreadsheet document. I also need to search and
replace texts in the OLE objects (Spreadsheets). The keys and values for
replacement are more than one thousand pairs.
     PROBLEMS:
     1. The java program searchs and replaces all texts in the text 
        document first, and then starts searching and replacing texts in the 
        OLE objects. After the program searchs and replaces 4th or 5th OLE 
        objects, the star office always crashes.  Data is lost. There is no 
        any error message and the java program is still running. It is always 
        nomal, when the text document contains less than 4 OLE objects.

     2. In order to avoid the above problem, I do the task in two 
	programs. At first, I run one program for replacing all texts in the 
	 text document, then close the first program. Then, I run the second 
	 program for replacing all texts in OLE objects in the text document. 
	 This ways seems OK, when the document contains less than 4 OLE 
	 objects. When it has more than 4 OLE objects, there is some problems. 
	 I will always get the error message, " An unrecoverable error has 
	 occurred. All modified files have been saved and can probably be 
	 recovered at program restart.", whenever I  dispose the doc frame in my 
	 java programs or I close the doc manually.


I attach the way I grab the OLE objects and replace texts in those objects.

	TextEmbeddedObjectsSupplier = (XTextEmbeddedObjectsSupplier) 
	 UnoRuntime.queryInterfac( XTextEmbeddedObjectsSupplier.class, 
 xTextDocument );
	     xNameAccess = xTextEmbeddedObjectsSupplier.getEmbeddedObjects

	     String OLENames[] = xNameAccess.getElementNames();

	     XTextEmbeddedObject xOLE = 
	 (XTextEmbeddedObject)xNameAccess.getByName(OLENames[index]);
	     XEmbeddedObjectSupplier oEOS = (XEmbeddedObjectSupplier) 
	 UnoRuntime.queryInterface(XEmbeddedObjectSupplier.class, xOLE);
	     XInterface oInt = oEOS.getEmbeddedObject();
	     XSpreadsheetDocument SDoc = 
(XSpreadsheetDocument)UnoRuntime.queryInterface(XSpreadsheetDocument.class,oInt);
	     XSpreadsheets oSheets = SDoc.getSheets() ;
	     XIndexAccess oIndexSheets = (XIndexAccess) 		
UnoRuntime.queryInterface(XIndexAccess.class, oSheets);
	     int totalSheets = oIndexSheets.getCount();
	     for(int j = 0; j < totalSheets; j++){
	        oSheet = (XSpreadsheet) oIndexSheets.getByIndex(j);
	         xReplaceable = ( XReplaceable ) UnoRuntime.queryInterface( 
XReplaceable.class, oSheet);
  
	        xReplaceDescriptor = ( XReplaceDescriptor)
xReplaceable.createReplaceDescriptor();
        	Iterator iter = tagMap.keySet().iterator();
	        while(iter.hasNext()) {
        	    String tagStr = (String)iter.next();
	            xReplaceDescriptor.setSearchString(tagStr);
	        xReplaceDescriptor.setReplaceString((String)tagMap.get(tagStr));
   	 xReplaceable.replaceAll(xReplaceDescriptor);
	        }//end while
	    }//end inner for

Also, I would like to attach the example text document for you. But, there is no
way to attach the file. 

Thanks

Candy Chen