Re: mathml xsl &#symbols

[Joseph Kesselman <ke...@us.ibm.com>]

The "?" in place of ࠎ  may actually not be a problem. Some tools 
can't display all the Unicode characters, and show the ones they don't 
understand as question marks. So the first thing to check is to load the 
file into something that will let you see the actual bytes and check 
whether the problem is in the file or in the tool you're using to view it.

If it really isn't showing up correctly: Are you using Xalan's serializers 
as your output, or are you taking SAX or DOM output? Our serializers 
should be scanning for codes that can't be expressed in the document's 
encoding (as set in xsl:output) and converting them to numeric character 
references. If you don't specify an encoding, we should be producing 
UTF-8, which can represent all of Unicode but which may require 2 or 3 
bytes to represent the higher-numbered characters; this will only displsy 
properly in a tool that understands UTF-8.

If you *ARE* using SAX or DOM output, the text will be UTF-16 (since 
that's how those APIs represent characters) and it's the responsibilty of 
your next stage of code to be able to deal with that.

______________________________________
Joe Kesselman  / IBM Research