RE: [SOAP] bug in client? xml in the soap body
| From: | Andrew S. Nagy | Date: | Wed, 25 Jun 2003 14:11:36 +0000 |
| Subject: | RE: [SOAP] bug in client? xml in the soap body | ||
| References: | 1 | Groups: | php.soap |
| Request: | Send a blank email to php-soap+get-730@lists.php.net to get a copy of this message | ||
Does anyone else have an opinion on this?
> -----Original Message-----
> From: Jonathan Schilling [mailto:jls@sco.com]
> Sent: Monday, June 23, 2003 8:49 AM
> To: Andrew S. Nagy; soap@lists.php.net; Shane Caraveo
> Subject: Re: [SOAP] bug in client? xml in the soap body
>
>
> > From asnagy@lsb.syr.edu Fri Jun 20 15:28:26 2003
> >
> > I am trying to consume a soap response where the results are listed as a
> > sting, but the string is an actual xml document with all of the brackets
> > turned into > and <
> >
> > However the soap client throws a soap_fault with the message:
> > XML error on line 9414 col 33 byte 999069 not well-formed
> (invalid token)
> >
> > According to the group offering the web service, it is a working and
> > well-formed xml document. Even though I told them that
> publishing an XML
> > document via soap is kinda pointless, it should still work
> though, right?
> >
> > Has anyone ever tried to consume xml through SOAP before? Will
> PEAR::SOAP
> > allow this?
>
> We have web services that take and return strings as parameters,
> where the strings are XML documents. These services have been consumed
> successfully by several client SOAP implementations including PEAR::SOAP.
>
Good to hear, thanks!
> However this service does not do the >/< replacement. Don't see
> why that would be necessary, since the XML isn't going to be interpreted
> as HTML by a browser.
>
> Why do you say that "publishing an XML document via soap is kinda
> pointless"?
Publishing XML via SOAP seems pointless to me, because all it is is an extra
layer. Whats the difference between publishing an XML document as a link on
the web as aposed to offering through a web service. It just seems like an
extra layer that is not needed.
>
> Jonathan Schilling
>