DataPower

DataPower

Join this online group to communicate across IBM product users and experts by sharing advice and best practices with peers and staying up to date regarding product enhancements.

 View Only
Back to discussions
Expand all | Collapse all

xml parser of WSP does not use iso-8859-1 characterset when set in content-type results in illegal character

  • 1.  xml parser of WSP does not use iso-8859-1 characterset when set in content-type results in illegal character

    Posted Tue October 22, 2024 03:56 AM

    In our production environment on a wsdl webservice that is used international by EU many countries I ran into the following issue. 

    When parsing a Soap message with in the http header the content-type  set to characterset iso-8859-1, the XML parser tries to parse it using characterset UTF-8 and reports errors for special characters. (illegal character error). Only when including the iso-8859-1 characterset in the xml itself (<?xml version="1.0" encoding="iso-8859-1"?>) the encoding is done using iso-8859-1.  

    I think this is faulty behaviour of the DataPower. If NO characterset is set in xml/soap, but it is set in the http header by means of the content-type, I think the parser should use the indicated characterset of the content-type. 

    What do you think?



    ------------------------------
    John Gouw
    ------------------------------


  • 2.  RE: xml parser of WSP does not use iso-8859-1 characterset when set in content-type results in illegal character

    Posted Tue October 22, 2024 09:09 AM

    These kinds of cross-charset questions date back to as long as I can remember with DataPower.

    You'd have thought, by now, this would have been addressed more elegantly. 

    Yes, I agree.  If you put in an RFE, I'll vote for it.



    ------------------------------
    Joseph Morgan
    CEO - Independent
    Joseph Morgan
    Dallas TX
    ------------------------------

    Original Message


Related Content