Re: XML Parsing with Encoded Characters

[Scott Klement <sk@xxxxxxxxxxxxxxxx>]

Hi Alan,

To add some context to the message you quoted.  The person who asked the 
question was retrieving their SOAP message with http_url_post_xml().

It was the same symptom as you describe... but when http_url_post_xml() 
calls the end-element-handler for the <soap:Body> element, it provides a 
'value' that contains the entire XML document, with the &lt;, &gt;, etc 
already decoded.

If you then take the value from soap:Body, treat it as a new XML 
document, and decode it a second time with http_xml_parse_stmf() or 
http_xml_parse_string() you'll get the answer you're looking for.

There's an example of this here:
http://www.scottklement.com/httpapi/bankrout.txt

Good luck


On 3/1/2011 9:15 AM, Alan Smith wrote:
>
>     I'm downloading a SOAP message from a webservice where the body
>     contains an xml document.  The xml tags (<  >) within the body of the
>     message are being escaped.  This is causing a problem with the xml
>     parser.  While searching for an answer as to how to handle this, I
>     came across another thread on this mailing list where Scott explained
>     how to handle the situation.  This is from the other thread:
> <soap:Envelope whatever-whatever>
>     <soap:Body whatever>
>       &lt;ADStart&gt;&lt;error code="0" text="NO_ERROR"
>       /&gt;&lt;ticket&gt;fe81cb7e-fd9e-411b-b3fd-b9c5729728f7
>       &lt;/ticket&gt;&lt;/ADStart&gt;
>     </soap:Body>
> </soap:Envelope>
>
> What I mean is... are the<,>  and&  characters escaped so that they
> won't be treated as XML tags, as they are in the above example?  If so,
> that's a very common issue.  The solution is to have the contents of the
> 'Body' element saved to a file (or a string if it's small enough) and
> then parse that file/string separately using http_parse_xml_stmf() or
> http_parse_xml_string()
>
>     I've taken the body portion of the message and placed it into a
>     separate streamfile and am using http_parse_xml_stmf() to parse that
>     file.   However, this still doesn't work.  Should the xml tags be
>     un-escaped in order for the parser work?  Or is there another
>     solution?
> -----------------------------------------------------------------------
> This is the FTPAPI mailing list.  To unsubscribe, please go to:
> http://www.scottklement.com/mailman/listinfo/ftpapi
> -----------------------------------------------------------------------
>

-----------------------------------------------------------------------
This is the FTPAPI mailing list.  To unsubscribe, please go to:
http://www.scottklement.com/mailman/listinfo/ftpapi
-----------------------------------------------------------------------