(I'm switching this thread to the axis-c-dev list, because that seems
more appropriate for the level of detail I'm asking about)

So if axutil_stream_write actually writes the xml out, why does the
generated foo_serialize() function return a node, and why does that get
passed to axis2_svc_client_send_receive_with_op_qname?  Is that actually
necessary for something, or would it work to pass a NULL there, or some
other dummy node?

eric

>-----Original Message-----
>From: Dimuthu Gamage [mailto:dimut...@gmail.com] 
>Sent: Saturday, January 31, 2009 2:36 AM
>To: Apache AXIS C User List
>Subject: Re: special character escaping in adb requests
>
>Hi Eric,
>
>Yea we have the stream write function which directly write the 
>xml in the generated code. So fixing it axiom_node_serialize 
>is not enough, we have to change the generated code a little 
>bit. But for your custom need it is better setting data after 
>escaping special characters.
>
>
>
>On Sat, Jan 31, 2009 at 3:25 AM, Haszlakiewicz, Eric 
><ehas...@transunion.com> wrote:
>
>
>
>       hmm... I had assumed that the adb code would just generate an
>       appropriately structured axiom_node_t, and it seems 
>like it does return
>       a node, but it's also generating xml directly and 
>writing it out (with
>       axutil_stream_write).  huh?  I think I'm missing 
>something with how this
>       works.
>       
>       I tried to read through the 
>axis2_svc_client_send_receive_with_op_qname
>       call that uses the axiom_node_t (which is called from 
>the stub "myOp"
>       call), hoping that that might provide some insight, but 
>quickly got
>       lost.  It seems the figuring out what this code does 
>will take more time
>       than I have available right now. :(
>       
>
>       >-----Original Message-----
>       >From: Haszlakiewicz, Eric [mailto:ehas...@transunion.com]
>       >Sent: Friday, January 30, 2009 3:04 PM
>       >To: Apache AXIS C User List
>       
>       >Subject: RE: special character escaping in adb requests
>       >
>       >
>       >Until you try to send some data that contains "]]>".  There is
>       >no way to
>       >avoid to the need to escape the data.  This should be 
>handled by the
>       >serialization methods.  I'm very surprised that it 
>isn't.  I haven't
>       >looked at the code that does the serialization yet, 
>but if there's a
>       >general axion serialize-an-element routine, this 
>probably won't be too
>       >hard to fix.
>       >
>       >>-----Original Message-----
>       >>From: Dumindu Pallewela [mailto:pallew...@gmail.com]
>       >>Sent: Friday, January 30, 2009 12:27 PM
>       >>To: Apache AXIS C User List
>       >>Subject: Re: special character escaping in adb requests
>       >>
>       >>Does AXIOM support CDATA sections?
>       >>
>       >>If so I think this can simply be solved by using a 
>CDATA section.
>       >>
>       >>Thanks,
>       >>Dumindu.
>       >>
>       >>On Fri, Jan 30, 2009 at 9:20 AM, Dimuthu Gamage
>       >><dimut...@gmail.com> wrote:
>       >>> Unfortunately we don't have such a function in the axutil
>       >>yet. But it is
>       >>> trival to implement.
>       >>> I just did a google and found this, [1]. We can use the
>       >>algorithm for the
>       >>> parse data.
>       >>>
>       >>>
>       >>> [1] http://www.hdfgroup.org/HDF5/XML/xml_escape_chars.htm
>       >>>
>       >>> Thanks
>       >>> Dimuthu
>       >>>
>       >>> On Thu, Jan 29, 2009 at 10:48 PM, Haszlakiewicz, Eric
>       >>> <ehas...@transunion.com> wrote:
>       >>>>
>       >>>> hmm... well that sucks.  Is there an axis function to do
>       >>this easily, or
>       >>>> do I need to find one somewhere else?
>       >>>>
>       >>>> Also, how do I know what kind of escaping to do?  It's
>       >>possible to have
>       >>>> axis send the request as a soap doc, or as a json message,
>       >>so I can't
>       >>>> use the same escaping for each, right?
>       >>>>
>       >>>> eric
>       >>>>
>       >>>> >-----Original Message-----
>       >>>> >From: Dimuthu Gamage [mailto:dimut...@gmail.com]
>       >>>> >Sent: Wednesday, January 28, 2009 9:01 PM
>       >>>> >To: Apache AXIS C User List
>       >>>> >Subject: Re: special character escaping in adb requests
>       >>>> >
>       >>>> >Hi Eric,
>       >>>> >Unfortunately You have to set the texts after 
>escaping these
>       >>>> >characters at least for now. It is apparently a bug. (we
>       >>>> >should do it inside). Can you please raise a JIRA 
>issue on that.
>       >>>> >
>       >>>> >Thanks
>       >>>> >Dimuthu
>       >>>> >
>       >>>> >
>       >>>> >On Thu, Jan 29, 2009 at 12:48 AM, Haszlakiewicz, Eric
>       >>>> ><ehas...@transunion.com> wrote:
>       >>>> >
>       >>>> >
>       >>>> >
>       >>>> >       I have an app that uses adb to construct 
>messages to
>       >>>> >send to a web
>       >>>> >       service.  Everything works fine, unless my data
>       >>>> >contains characters like
>       >>>> >       '<', in which case axis fails to escape 
>the characters
>       >>>> >and generates
>       >>>> >       invalid xml (I'm using xml messages).
>       >>>> >
>       >>>> >       I'm doing things like:
>       >>>> >
>       >>>> >       axis2_stub_t *my_service =
>       >>>> >axis2_stub_create_MyService(env, client_home,
>       >>>> >       url);
>       >>>> >       adb_foo_t *myreq = adb_foo_create(env);
>       >>>> >       adb_foo_set_myfield(myreq, env, "abc<xyz");
>       >>>> >       adb_fooresp_t *response =
>       >>>> >axis2_stub_op_MyService_myOp(an_service, env,
>       >>>> >       myreq);
>       >>>> >
>       >>>> >       The adb code was generated from a wsdl 
>(and xsd) that
>       >>>> >has elements
>       >>>> >       defined like:
>       >>>> >         <xs:element name="foo" type="xs:string"/>
>       >>>> >       and the command used was:
>       >>>> >         axis2.sh org.apache.axis2.wsdl.WSDL2C -uri
>       >>>> >myservice.wsdl -d adb -u
>       >>>> >       -o generated
>       >>>> >
>       >>>> >       Is there some extra step I need to take to 
>make this
>       >>>> >work right?  Or, is
>       >>>> >       there some bug that might be fixed in a more
>       >recent version?
>       >>>> >
>       >>>> >       eric
>       >>>> >
>       >>>> >
>       >>>> >
>       >>>> >
>       >>>> >
>       >>>> >--
>       >>>> >Thanks,
>       >>>> >Dimuthu Gamage
>       >>>> >
>       >>>> >http://www.dimuthu.org
>       >>>> >http://www.wso2.org
>       >>>> >
>       >>>> >
>       >>>
>       >>>
>       >>>
>       >>> --
>       >>> Thanks,
>       >>> Dimuthu Gamage
>       >>>
>       >>> http://www.dimuthu.org
>       >>> http://www.wso2.org
>       >>>
>       >>
>       >
>       
>
>
>
>
>-- 
>Thanks,
>Dimuthu Gamage
>
>http://www.dimuthu.org
>http://www.wso2.org
>
>

Reply via email to