#51 prolem with handling upper unicode characters

closed
nobody
None
5
2012-10-08
2007-02-15
No

As described in the help forum, saxon seems to have a problem handling entity references that contain upper unicode values (that is, characters with numeric codepoints larger than 65536). Two files are attached to demonstrate the problem.

Discussion

  • Christian Wittern

    xml file that contains a entity reference to demonstrate the problem

     
  • Christian Wittern

    trivial xsl stylesheet that can be used to demonstrate the problem

     
  • Christian Wittern

    Logged In: YES
    user_id=222320
    Originator: YES

    File Added: test.xsl

     
  • Michael Kay

    Michael Kay - 2007-02-15

    Logged In: YES
    user_id=251681
    Originator: NO

    You appear to have found a bug in the Xerces parser, which is the default parser in JDK 1.5. This test case works correctly with the .NET parser and with the Crimson parser (which is the default parser in JDK 1.4), but on JDK 1.5 the parser is feeding incorrect data to Saxon. Please raise the bug against the JDK. In the meantime you can work around it by selecting a different parser - Saxon will work with any SAX2-compliant XML parser. (Having said that, Xerces is usually more reliable than any other).

     


Anonymous

Cancel  Add attachments