W3C home > Mailing lists > Public > www-validator-css@w3.org > December 2001

Ecxeption in parseCLOB procedure

From: Neeraja Divakaruni <neeraja.divakarunI@oracle.com>
Date: Mon, 17 Dec 2001 10:16:08 +0530
Message-ID: <3C1D7890.FC348EB0@oracle.com>
To: www-validator-css@w3.org
Hi ,

I have some problem with the xml parser and please read the following
desciption of the problem.

We are getting an exception while parsing the XML document using Oracle
parser.
The exception is "5-byte UTF-8 encoding not supported".

The detailed scenario is expalined below ::

We have a XML document containing danish character "ø" (as in Søren) and

created in windows 2000 notepad.

<?xml version = '1.0' standalone = "no" ?>
<!-- Oracle eXtensible Markup Language Gateway Server  -->
<!DOCTYPE ECX_MAPS SYSTEM "ecx_mappings.dtd">
<ECX_MAPS>
   <ECX_MAPPINGS>
      <MAP_ID>Søren</MAP_ID>
      <MAP_CODE>A2A_INN</MAP_CODE>
   </ECX_MAPPINGS>
</ECX_MAPS>


Case 1:  It is saved without specifying any encoding format .Then ported

to UNIX in ASCII mode and parsed using   xmlparse.parseCLOB procedure.
  Result : parsing fails with exception "5-byte UTF-8 encoding not
supported".

Case 2:  Saved it in UNICODE format and ported the same to UNIX in ASCII

mode and parsed using  xmlparser.parseCLOB.
  Result : Parsing succeeds.

So the parsing succeeds if the data is unicode format . In our case we
get the data from database tables and teh database default encoding is
ISO-8859-1. we also tried adding an encoding tag as "ISO-8859-1" in the
above document with the idea that parser resolves it. But still it
fails.

How to resolve this. Please reply asap.

--
Neeraja D
Applications Engineer,
Applications Technology Group
Oracle Software India Ltd.

Work : +91 (40) 311 0222   Extn. : 4067
Email  : neeraja.divakaruni@oracle.com
Received on Sunday, 16 December 2001 23:46:18 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 27 June 2012 00:14:05 GMT