can a valid xml body have escaped characters for the '<' and '>' around the element names

Question

My web service is receiving xml from a third party that looks like this:

<SOAP-ENV:Envelope xmlns:SOAP-ENV="http://schemas.xmlsoap.org/soap/envelope/"&gt;
    <SOAP-ENV:Body>
&lt;Foo&gt;bar&lt;/Foo&gt;
    </SOAP-ENV:Body>
</SOAP-ENV:Envelope>

My jaxws web service rejects this with a parsing error. Also if I try to validate this xml using soapui it says Body with element-only content type cannot have text element.

My question is, is that xml valid? Or is the client supposed to send me something without escaping the < and >.

Any references to xml standards or rules are appreciated.

Answer 1

Alternatively, you may escape the full string by enclosing it with the CDATA (character data) section. A CDATA section begins with the nine-character delimiter ![CDATA[ and ends with the delimiter ]]:

  <![CDATA["Me, Myself & <I>"]]>

take a look here http://xmmssc-www.star.le.ac.uk/SAS/xmmsas_20070308_1802/doc/param/node24.html

Answer 2

Well, that's certainly valid XML. However your question seems to be rather whether it conforms to the SOAP protocol, which is a more restrictive question.

In the example, your SOAP body is just a text element. I'm no SOAP expert but in all of the SOAP examples I see, there are element nodes in the body and not just text. Also the SOAP Recommendation (http://www.w3.org/TR/2000/NOTE-SOAP-20000508) does say

"A body entry is identified by its fully qualified element name"

which does suggest a requirement for elements rather than text.