We had been using an older version of Crimson and Xalan in our Java App. Our
java app has an XmlReader that "reads" the data from our objects and uses an
XSL Stylesheet to transform it to HTML. The following example shows xml
elements that are created inside a root element.
public final void xmlContent(org.xml.sax.ContentHandler ch) throws
org.xml.sax.SAXException{
String name, token;
Object value;
TableRow row;
int rows = getRowCount();
int cols = getColumnCount();
for(int iRow = 0; iRow < rows; iRow++){
row = (TableRow)getValueAt(iRow);
ch.startElement("","","row",dpak.xml.AbstractXMLReader.EMPTY_ATTR);
for (int iCol = 0; iCol < cols; iCol++ ){
name = getColumnName(iCol);
ch.startElement("","",name,dpak.xml.AbstractXMLReader.EMPTY_ATTR);
value = row.getValueAt(columnIndex[iCol]);
if (value == null){
ch.characters(NULL_ARRAY,0,4);
}
else{
token = value.toString();
ch.characters(token.toCharArray(),0,token.length());
}
ch.endElement("","",name);
}
ch.endElement("","","row");
}
}
We wanted to try Saxon. When we use Saxon with the identity stylesheet we
get output that looks like:
<?xml version="1.0"
encoding="utf-8"?><><><>75-8730</><>2,040.00</></><><>94-5781</><>26.80</></
><><>104-0214</><>94.07</></></>
where all of the tag names are obviously empty. Using crimson the tag names
show up OK.
What are we missing?
|