[XERCESC-959] Surrogate characters mishandled by SAXPrint and SAX2Print.

• Type: Bug
• Status: Closed
• Resolution: Fixed
• Affects Version/s: Nightly build (please specify the date)
• Component/s: Samples/Tests
• Environment: Operating System: All; Platform: All
• Bugzilla Id: 21780
• Assignee: Unassigned
• Reporter: Michael Glavassevich
• Created: 22/Jul/03 03:18
• Updated: 16/Feb/11 23:37
• Resolved: 09/Apr/04 16:40

Description

From local CYGWIN build from CVS head (July 21, 2003):

The SAXPrint and SAX2Print samples write supplemental characters as character
references of their high and low surrogates. It looks like the problem might be
in framework/XMLFormatter, as I don't see any code in there that checks for
surrogates. If this is where the problem is, I would guess that the DOMWriter
exhibits the same behaviour.

Here's an example...

Input to SAXPrint:

<?xml version="1.0" encoding="UTF-8"?>
<root>𐀀􏿿</root>

Output from SAXPrint:

<?xml version="1.0" encoding="LATIN1"?>
<root>����</root>

The surrogate characters (xD800-xDFFF) are not part of Char, and thus those
char refs are illegal.