Re: Problem parsing <

[David N Bertoni/Cambridge/IBM]

Hi Steve,

 I doubt there's a bug in Xerces -- I suspect the bug is in your code.

SAX makes no guarantee that all of the content of an element will be
delivered in one call to characters().  Since Xerces-C breaks up calls to
characters when dealing with entities, I'll bet your code assumes the
content _will_ be delivered in one call, so you're creating multiple child
elements because there are multiple characters() calls.  When you get a
characters() call, you should save the characters in a buffer, then process
the buffer once you get a call to something other than characters().

Hope that helps...

Dave



                                                                                       
                                                
                      "Steve Barge"                                                    
                                                
                      <[EMAIL PROTECTED]         To:      <[EMAIL PROTECTED]>  
                                                
                      m>                       cc:      (bcc: David N 
Bertoni/Cambridge/IBM)                                           
                                               Subject: Problem parsing &lt;           
                                                
                      10/07/2002 10:07                                                 
                                                
                      AM                                                               
                                                
                      Please respond                                                   
                                                
                      to xerces-c-dev                                                  
                                                
                                                                                       
                                                
                                                                                       
                                                




Hi,

I'm using version 1.7 of xerces-c++.  I'm having problems parsing the
following:

<smd:attr_li name="ancestor">
<smd:li>Contrib-Tools.PIMTOOL,r&lt;B.11.11,l=/</smd:li>
<smd:li>OS-Core.UX-CORE,r&lt;B.11.11.%,l=/</smd:li>
</smd:attr_li>

The "r&lt;B.11.11" should translate to "r<B.11.11".  I have derived a class
from the SAX DefaultHandler and in the ::characters method, the "&lt;" is
converted to '<' but for some reason it thinks that the '<' is then the
beginning of the end tag "</smd:li>" or it thinks it is a new list element
"<smd:li>".  So it converts the above to:

             <smd:attr_li name="ancestor">
                         <smd:li>Contrib-Tools.PIMTOOL,r</smd:li>
                         <smd:li><B.11.11,l=/</smd:li>
                         <smd:li>OS-Core.UX-CORE,r</smd:li>
                         <smd:li><B.11.11.%,l=/</smd:li>
             </smd:attr_li>

Does anybody know why this is happening?  Is this a bug in this version of
xerces-c++ or is this a problem whith the way the '<' is escaped in the
XML?

Thanks in advance!

Steve Barge


---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]





---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]