I think I need the help of outside eyes, because I must be not seeing the forest for the trees. Under Tomcat 4.0.6 I was serving out XML which had Unicode code points for Arabic characters and in all decent browsers (Mozilla, Phoenix, Chimera, etc.) the Arabic displayed beautifully. Yesterday we updated to Tomcat 4.1.17: none of the data files changed, none of the XSL files changed, none of our server settings changed. Under the new Tomcat everything seemed to be as before except ... the beautiful Arabic has gone and in its place is that gobbledegook of squares, "at-signs", etc. that browsers give when they don't know what else to do.
When I use Mozilla to look at the XML file that the JSP is importing I can see that the Unicode is definitely there and displays fine, so the problem must be occurring after Tomcat gets hold of it. After reading Bill Barker's response to someone else's query yesterday I specifically added "charset=UTF-8" to the <%@ page contentType="text/xml" %> that I previously had, but still no Arabic. I couldn't find anything in the RELEASE-NOTES-4.1.txt that looked relevant. Has something changed between 4.0.6 and 4.1.17 that would explain my vanished Arabic? Thanks, Paul. -- To unsubscribe, e-mail: <mailto:[EMAIL PROTECTED]> For additional commands, e-mail: <mailto:[EMAIL PROTECTED]>