Question: User defined acknowledgement (IDOC to SOAP)

Hello friends,
A very happy new year to all of you!
The flow im working on is like this -
R/3 -> Idoc -> XI -> SOAP adapter -> webservices
I have a requirement of sending an acknowledgement back from XI. But the problem is the acknowledgement has to be of a structure that is specified by the client. I will have to send this acknowledgement through RFC to R/3 where they will update.
q1. Can i access the information about the idoc, like if it has failed in sender channel,or IR or at the mapping etc. They want these info codified into something like 101,102 103 etc with each having specific meanings in terms of Idoc status.
q2. Are these information stored somewhere so that i can access them and map it to the acknowledgement structure?
q3. what are the steps for achieving this?
Looking forward to your help .... Thanks in advance!
Regards,
Lakshmi

Hi,
As suggested by Swarup, "you can go for Fault message type and have the customized structure to give the details."
In addition, I would suggest to use FixedValue Table in mapping to maintain your entries (101,102, etc) and their respective meanings or you can use Value Mapping (maintain your entries in Integration Directory. For more info search SDN).
Regards,
Sarvesh

Similar Messages

Multi mapping question using user defined function

Hi,
I have a message with multiple occuring nodes (i.e. one message with multiple orders (header + detail)) that I need to map to a idoc. I need to filter out of the source based on order type (in header) from creating an idoc.. How do I do it using user defined function + message mappping ?
mad

All - Thanks much.. Here is my requirement that is no solved by regular mapping
<Root>
<Recordset>
<Ordheader>
    <ord>
    <ord_type>
</Ordheader>
   <Ord_line>
     <ord>
     <Linnum>
   </Ord_line>
</Recordset>
<Recordset>
<Ordheader>
    <ord>
    <ord_type>
</Ordheader>
   <Ord_line>
     <ord>
     <Linnum>
   </Ord_line>
</Recordset>
<Root>
As you see above, each recordset has order transaction. One Root message can contain multiple of these. So, when I map to the IDOC, I want to filter out any ord_type <> XX.
If I use regular graphical map, it only looks at first recordset and accepts all or rejects all.
I need to use UDF. In the UDF, what comes in as input ? Resultset is output -correct ? Now how do I usse graphical mapping with UDF to generate the correct target info

Questions about user-defined functions

Hello all,
I've got some questions about user-defined function in a message mapping:
1) How can I get the current date/time in format yyyy-mm-dd hh:mm:ss ? What is the java code for this?
2) I want to use the StreamTransformation constant TIME_SENT, only this is not in the right time-zone.
It is GMT and it should be CET (1 hour difference). How can I convert this in Java?
Can somebody help me with this?
Thanks in advance.
Kind regards,
Marco van Iersel

Hi Marco,
If the date format is fixed as you have mentioned,please use this:
                                String test = "2009-03-27 23:15:30";
          String test1 = test.substring(11,13);
          if(!test1.equals("23"))
          int a = Integer.parseInt(test1);
          int b = a+1;
          System.out.println("b"+b);
          String c = Integer.toString(b);
                                test = test.substring(0,10)+" " + c + test.substring(13,19);
                                return test;
          if(test1.equals("23"))
          test = test.substring(0,10)+ " 00" + test.substring(13,19);
          return test;
                                else return "";
Kindly let me know if this works.
Thanks.
Regards.
Shweta

How to check empty idoc fields in user defined function

Dear All,
I am working on an IDOC to file scenario. In my user defined function, I want to check whether a particular field of the idoc is populated. I can perform this test in message mapping by comparing the field with an empty constant.
How can I do the same test in an advanced user defined function? I have tried all the following, but none has worked so far
If (a.length == 0) .
If (a[0].length == 0)
If (a[0] == )
If (a[0] == )
Am I doing something wrong?
Thx/Farshad

To check whether an element is not available:
if (a.length == 0)
but you have to set cache to context. If you set cache to queue, it will not work, as in this case the context changes are part of the array a.
To check whether an element is empty:
if (a[0].equals(""))
Regards
Stefan
Message was edited by: Stefan Grube
Message was edited by: Stefan Grube

OSB question : passing a user defined transport header to a proxy service

Hi All,
I have a proxy service that calls another proxy service. The protocol for the "called proxy service" is set to local. The calling proxy service has a "transport header" action to add a user defined header to the outbound request.
I log the outbound request in the calling proxy service before publishing/calling the other proxy service. It does have header set as expected. But the called proxy service does not get this header.
As a note in the "transport" tab "get all headers" is set to true.
Please suggest
Here is the outbound xml gets printed just before calling the proxy service
<con:endpoint name="BusinessService$userProvisioning$businessService$HttpMessageSender" xmlns:con="http://www.bea.com/wli/sb/context">
<con:service/>
<con:transport>
<con:uri>http://localhost:8064/ProvisioningService/</con:uri>
<con:mode>request-response</con:mode>
<con:qualityOfService>best-effort</con:qualityOfService>
<con:request xsi:type="http:HttpRequestMetaData" xmlns:http="http://www.bea.com/wli/sb/transports/http" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance">
<tran:headers xsi:type="http:HttpRequestHeaders" xmlns:tran="http://www.bea.com/wli/sb/transports">
<tran:user-header name="JMS_BEA_RedeliveryLimit" value="1"/>
<tran:user-header name="JMSDeliveryMode" value="2"/>
<tran:user-header name="JMSExpiration" value="0"/>
<tran:user-header name="JMSMessageID" value="ID:<914937.1294164899050.0>"/>
<tran:user-header name="JMSPriority" value="4"/>
<tran:user-header name="JMSRedelivered" value="false"/>
<tran:user-header name="JMSTimestamp" value="1294164899050"/>
<tran:user-header name="JMSXDeliveryCount" value="1"/>
*<tran:user-header name="reResolvedAddress" value="http://LPF004689:8080/axis2/services/provisioning/"/>*
*<tran:user-header name="reResolvedRegion" value="MB"/>*
<http:Content-Type>application/soap+xml; charset=utf-8</http:Content-Type>
</tran:headers>
<tran:encoding xmlns:tran="http://www.bea.com/wli/sb/transports">utf-8</tran:encoding>
</con:request>
</con:transport>
<con:security>
<con:doOutboundWss>false</con:doOutboundWss>
</con:security>
</con:endpoint>>
And here is the inbound xml getting printed inside the called proxy service
inbound="<con:endpoint name="ProxyService$userProvisioning$proxyService$LocalSoapListener" xmlns:con="http://www.bea.com/wli/sb/context">
<con:service/>
<con:transport>
<con:mode>request-response</con:mode>
<con:qualityOfService>best-effort</con:qualityOfService>
<con:request xsi:type="loc:LocalRequestMetaData" xmlns:loc="http://www.bea.com/wli/sb/transports/local" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance">
<tran:headers xsi:type="loc:LocalRequestHeaders" xmlns:tran="http://www.bea.com/wli/sb/transports"/>
<tran:encoding xmlns:tran="http://www.bea.com/wli/sb/transports">utf-8</tran:encoding>
</con:request>
<con:response xsi:type="loc:LocalResponseMetaData" xmlns:loc="http://www.bea.com/wli/sb/transports/local" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance">
<tran:headers xsi:type="loc:LocalResponseHeaders" xmlns:tran="http://www.bea.com/wli/sb/transports"/>
<tran:response-code xmlns:tran="http://www.bea.com/wli/sb/transports">0</tran:response-code>
</con:response>
</con:transport>
<con:security>
<con:transportClient>
<con:username><anonymous></con:username>
</con:transportClient>
</con:security>
</con:endpoint>"

Hi,
We are using OSB 11g for sending message to external client.
we are setting Headers "JMSType" in set Transport Header. but Client is not able to recieve the JMSType header at there side.
we have read many forums for the same and turned on
Pass all Headers through Pipeline
Get All Headers
then I put one publish action and keep my "Set Transport Header" inside that. still client is not able to find the same
In out outbound request we can see the header as following.
$outbound = <con:endpoint name="BusinessService$Notification$BusinessService$external$BUSINESSSERVICE:con="http://www.bea.com/wli/sb/context">
<con:service/>
<con:transport>
<con:uri></con:uri>
<con:mode>request</con:mode>
<con:qualityOfService>exactly-once</con:qualityOfService>
<con:request xsi:type="http:HttpRequestMetaData" xmlns:http="http://www.bea.com/wli/sb/transports/http" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance">
<tran:headers xsi:type="http:HttpRequestHeaders" xmlns:tran="http://www.bea.com/wli/sb/transports">
<tran:user-header name="JMSType" value="XXXXXXX">
<http:Content-Type>text/plain; charset=utf-8</http:Content-Type>
</tran:headers>
<tran:encoding xmlns:tran="http://www.bea.com/wli/sb/transports">utf-8</tran:encoding>
</con:request>
<con:response xsi:type="http:HttpResponseMetaData" xmlns:http="http://www.bea.com/wli/sb/transports/http" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance">
<tran:headers xsi:type="http:HttpResponseHeaders" xmlns:tran="http://www.bea.com/wli/sb/transports">
<tran:user-header name="content-description" value="SOAP"/>
<tran:user-header name="content-disposition" value="attachment;filename="[email protected]""/>
<tran:user-header name="content-id" value="<[email protected]>"/>
<http:Cache-Control>proxy-revalidate</http:Cache-Control>
<http:Connection>Keep-Alive</http:Connection>
<http:Content-Length>112</http:Content-Length>
<http:Content-Type>text/xml; charset=utf-8</http:Content-Type>
<http:Date>Sun, 02 Sep 2012 04:06:43 GMT</http:Date>
<http:Server>XXXXXXXXXerver>
<http:Set-Cookie><![CDATA[<cookie-values xmlns="http://www.bea.com/wli/sb/transports/http">
<value>JSESSIONID=m2RvGs2_Lvk_AZgyVJIv_e2-vSiFOQFi_mIA_SAPULGAKriCp_hbBD8uC0e8pEXt; Version=1; Path=/</value>
<value>saplb_*=(spspiq_PIQ_00)6487650; Version=1; Path=/</value>
</cookie-values>]]></http:Set-Cookie>
</tran:headers>
<tran:response-code xmlns:tran="http://www.bea.com/wli/sb/transports">0</tran:response-code>
<tran:response-message xmlns:tran="http://www.bea.com/wli/sb/transports">OK</tran:response-message>
<tran:encoding xmlns:tran="http://www.bea.com/wli/sb/transports">utf-8</tran:encoding>
<http:http-response-code>200</http:http-response-code>
</con:response>
</con:transport>
<con:security>
<con:doOutboundWss>false</con:doOutboundWss>
</con:security>
</con:endpoint>

Question about using subtypes (define a user-defined subtype w/ constrain)

Hi guys,
I am a newbie to PL/SQL.
I do really appreciate it if you would spare a bit of your time to reply this post.
According to the sub-section of Oracle® Database PL/SQL User's Guide and Reference 10g Release 2 (10.2) labeled Using Subtypes, I have found that the constrain of user-defined subtype can be overridden when declaring variables.
Qestions:
1. Why does PL/SQL allow constrain override of user-defined subtype? Or is it a unforeseen bug that exist in PL/SQL?
2. Does it happen to Oracle subtypes such as INTEGER, CHARACTER,etc.?
I have replicated this issue using the following PL/SQL code against Oracle9i 9.2.0.1, which produced no errors or exceptions whatsoever.
TO ELIMINATE ANY CONFUSIONS, THE ISSUE IN QUESTION IS HIGHLIGHTED IN BOLD
-- SAMPLE CODE TO REPLICATE THE ISSUE
DECLARE
-- a user-defined subtype(Xtype) is defined with a precision & scale constrain;
SUBTYPE   Xtype   IS   NUMBER(1,0);
-- however, Xtype is overridden when declaring variable 'var_num'.
var_num    Xtype(2,0);
BEGIN
var_num := 10;
DBMS_OUTPUT.PUT_LINE('var_num = ' || TO_CHAR(var_num));
EXCEPTION
WHEN OTHERS THEN
DBMS_OUTPUT.PUT_LINE('ERR: ' || SQLERRM);
END;
-- HERE IS THE OUTPUT
SQL> @D:\LocalWork\dummy.sql
var_num = 10
Edited by: HappyJay on 2010/05/09 1:48

After careful reading of your post, I found that your code should be rewritten as the following to prove the issue in question.
SQL> edit
1 DECLARE
2   SUBTYPE flagtype IS PLS_INTEGER RANGE 0..1;
3   x flagtype RANGE -2 .. 2; -- flagtype is overridden when declaring variable 'x'
4 BEGIN
5    FOR i IN -2 .. 2 LOOP
6      BEGIN
7        x := i;
8        dbms_output.put_line('Success: ' || TO_CHAR(x));
9      EXCEPTION
10        WHEN others THEN
11          dbms_output.put_line('Can not assign ' || TO_CHAR(i) || ' To Flagty
pe');
12      END;
13    END LOOP;
14* END;
SQL> /
Success: -2
Success: -1
Success: 0
Success: 1
Success: 2As you can see from the successful results, it still prevail my issue with user-defined subtype.
Any other thoughts?

OC4J and Oracle9i and user defined datatypes over SOAP.

Does OC4J and Oracle9i, that comes with JDeveloper9i-rc, support user defined datatypes over SOAP?
If I invoke the EJB over RMI it is working fine.
But if I try to invoke it over SOAP I get this fault:
500 Internal Server Error: Servlet error: Error building response envelope: java.lang.IllegalArgumentException: No Serializer found to serialize a 'mypackage.myObj' using encoding style 'http://schemas.xmlsoap.org/soap/encoding/'.
I have written the Deployment descripter like this:
<?xml version = '1.0'?>
<isd:service
id="urn:mypackage.EJB2"
type="rpc"
xmlns:isd=" " target="_new">http://xmlns.oracle.com/soap/2001/04/deploy/service"> <isd:provider
id="stateless-ejb-provider"
methods="getData"
scope="Request">
<isd:option key="JNDILocation" value="EJB2"/>
<isd:option key="DeploymentName" value="EJB2"/>
</isd:provider>
<isd:faultListener class="org.apache.soap.server.DOMFaultListener"/>
<isd:mappings>
<isd:map
encodingStyle="http://schemas.xmlsoap.org/soap/encoding" xmlns:x="http://glo2158" qname="x:mypackage.myObj"
javaType="mypackage.myObj"
java2XMLClassName="org.apache.soap.encoding.soapenc.BeanSerializer"
xml2JavaClassName="org.apache.soap.encoding.soapenc.BeanSerializer"/>
</isd:mappings>
</isd:service>

AFAIK this is not possible with the Oracle 8i implementation. This is my main gripe about Oracle's so called objects - you can't define your own constructors, you can't put check constraints on them and you can't use them in triggers.
In other words you can put any old scheisse in an object and you can't populate them with derived or default values.
I'm on a 9i New Features course as I type and I'm hoping to hear that Oracle have added some features to objects that will make them worth using.
regards, APC

Question in User Defined Functions....

Hi,
Can I develop new User Defined functions in ABAP programming language rather than in Java? If Yes, how and if not why?
Sorry, if this is a petty question as I am in the process of learning XI.
Thanks,
Monica

Hi,
yes it's a stantard function module
which you can call from abap mapping
there's no possibility to create user defined function
that you could use in graphical mapping in abap
take a look at this document on how to do abap mapping:
https://websmp103.sap-ag.de/~sapdownload/011000358700001795162005E/HowToIDocXMLToFlat.pdf
Regards,
michal
Message was edited by: Michal Krawczyk

DSC - can't acknowledge or clear a user defined event

I wrote this code for alarms but now I have modified it to handle events. So, first create the event (create alarm button), then try to acknowledge it or clear it. There should be a time inserted for acknowledging and clearing. It does't work. I have used the exact event URL to acknowledge it. Please help.
Thanks
Matt
Attachments:
User Defined Alarms.vi ‏52 KB

Hi Matt,
Thank you for attaching the new code, I didn't realize there was a path in that constant since the original box was so small. Part of the problem you are seeing actually lies within that path, you should be locating to \\*\Process Name\UserDefinedAlarm now that you switched from working with events to alarms. Also, you are still using the Set User Defined Event.vi and Read Events.vi, which need to move over to their Alarm counterparts.
Once you have those changes made, you need to alter the way you are trying to acknowledge the alarms. The problem isn't within logging the time but that the alarms are not being acknowledged at all. If you use the Read Alarms.vi and then wire the "alarms" output into the acknowledge function then it will acknowledge all of the active events (you may have to change the acknowledge type to Alarms). If you only want to acknowledge a single alarm at a time then you will need to trim down that cluster to focus on your selected alarm.
Finally, the way you are logging to the multicolumn listbox is not functioning correctly. I would suggest you look at the DSC Alarms Demo example (Help » Find Examples... » Toolkits and Modules » Datalogging and Supervisory Control » Alarms and Events » DSC Alarms Demo.lvproj). In that example there is a subVI, which is not part of the Alarms & Events palette but probably should be, named Format Alarm Data.vi. It is used in the first three examples under that project and should work better for logging all of your alarm interaction than searching through the database.
Regards,
Peter W.

Idoc to cXMLmapping..user defined function help

hello Everybody,
                          I am new to XI and I am working on a new mapping. In a nutshell, the requirement is to concat the TDLine of segment E1EDPT2 and map the data to comments of the cXML. The segment E1EDPT2 is segment in E1EDP01( Purchase Order line item segment) This the data that is coming from the vendor text of the SRM system. This needs to repeat for each line item in the Purchase Order. I have created a user defined function which concats the TD lines but I am not able to differenciate the tD line for other line items. Please help. I would appreciate it if anyone could share the User Defined function which suits my needs.
Thanks!
<E1EDP01> (PO lIne Item 1)
     <E1EDPT1 SEGMENT="1">
          <TDID>F01</TDID>
          <TSSPRAS>E</TSSPRAS>
          <TSSPRAS_ISO>EN</TSSPRAS_ISO>
       <E1EDPT2 SEGMENT="1">
             <TDLINE>XI mapping test - Newline1</TDLINE>
             <TDFORMAT>*</TDFORMAT>
       </E1EDPT2>
       <E1EDPT2 SEGMENT="1">
             <TDLINE>Line1- text2 end</TDLINE>
             <TDFORMAT>*</TDFORMAT>
       </E1EDPT2>
       <E1EDPT2 SEGMENT="1">
             <TDLINE>Line1 - text3</TDLINE>
            <TDFORMAT>*</TDFORMAT>
       </E1EDPT2>
     </E1EDPT1>
</E1EDP01>
<E1EDP01> ( PO lien item 2)
     <E1EDPT1 SEGMENT="1">
          <TDID>F01</TDID>
          <TSSPRAS>E</TSSPRAS>
          <TSSPRAS_ISO>EN</TSSPRAS_ISO>
     <E1EDPT2 SEGMENT="1">
           <TDLINE>XI mapping test - line2</TDLINE>
           <TDFORMAT>*</TDFORMAT>
     </E1EDPT2>
</E1EDPT1>
</E1EDP01>

Hi,
Please create context UDF which will concatenate all TDLine. Afterwards pass to this function TDLine with changed context to E1EDPT1.
In this function you need to concat full input array.
Regards,
Wojciech

Error in User defined function for dynamic file naming

Hi,
While creating User Defined function with this following code for dynamic fieldname
DynamicConfiguration conf = (DynamicConfiguration) container.getTransformationParameters().get(StreamTransformationConstants.DYNAMIC_CONFIGURATION);
DynamicConfigurationKey key = DynamicConfigurationKey.create("http://sap.com/xi/XI/System/File","FileName");
String ourSourceFileName = conf.get(key);
return ourSourceFileName;
which options should i select for cache Value, Context, Queue
for Augument , what name shd i mention.
Regards,
Varun

Hi Varun,
I guess I have answered a similar question just a few minutes before on very similar post from you. Just pasting the same here .................
Are you trying to access the ASMA values from the SOAP header of the XI message for the source file name?
First of all you need to Set the Sender File Adapter for Set ASMA and then file name. So it will work fine when you actually run the scenario end to end.
But in the mapping tool when you test the mapping - there is not message header updated with the actual source filename - and whenever you try to read the FileName attribute in the message header from the UDF - it cannot find the object and returns a NullPointerException.
I would suuggest for your mapping tool testing to be successful, have a check in the java code for null values,
DynamicConfiguration conf = (DynamicConfiguration) container.getTransformationParameters().get(StreamTransformationConstants.DYNAMIC_CONFIGURATION);
DynamicConfigurationKey key = DynamicConfigurationKey.create("http://sap.com/xi/XI/System/File","FileName");
String ourSourceFileName = conf.get(key);
If (ourSourceFileName != null)
Return ourSourceFileName;
Return "NoFileName.txt";
Let us know if this works.
Regards,
Suddha

IDoc - XI - SOAP - XI - Mail

Hi all,
I want to do the following scenario:
IDoc is sent to XI, Xi is doing a Soap call with information from the Idoc, from the soap call I receive a response code. This response code will be sent by mail TOGETHER with info from the Idoc.
I've made the scenario from Idoc to Soap call. I also know how to sent information out with the mail adapter.
Where the problem is for me, is the combination of this all together.
I guess I have to use BPM for this?
Can anyone point me into the right directions for this problem?
Thinks like wich message types, interface mappings etc?
Thnx in advance!

I guess I have to use BPM for this?
Not necessary.
U may perform a Idoc - Xi - Mail scenario with Soap lookup in mapping
Webservice Calls From a User Defined Function.
Regards,
Prateek

PI 7.3 Dual Stack User-Defined Message Search

Hi All,
I have implemented an Integrated configuration scenario SOAP-> PI – ABAP Proxy the scenario is sync the. The scenario has no mapping so
no operation mapping exist in the configuration scenario.
I tried setting up UDMS on the java stack without any success.
My questions are: is it possible to use UDMS if there is no mapping steps? And also does UDMS work for sync messages?

Hi Jannus
Check this document
http://www.sdn.sap.com/irj/scn/go/portal/prtroot/docs/library/uuid/803b22cf-ad75-2f10-0790-da055f2e8c0e?overridelayout=t…
On the very fast line it is written
"User Defined Message Search enables to search for asynchronous messages using the information available in payload or adapter-specific message attributes."
Last time we tried to configure UDMS for sync messages, but we were not able do that.
Thanks,
Indrajit

Read user name from Header of SOAP Sender

Hi All,
Is there any way to read value of "sap-user" passed in below URL.
http://<server>:<port>/sap/xi/adapter_plain?namespace=<sender_ns>&interface=<sender_mi>&service=myservice&party=&agency=&scheme=&QOS=BE&queueid=httpclient&sap-user=<sap-user>&sap-password=<sap-password>&sap-client=<sap-client>&sap-language=<sap-lang>
I tried using DynamicConfiguration in my SOAP sender adapter with "SRemoteUser" variable, it did NOT work for me.
Are there other options like Runtime Constants in JAVA Mapping etc.
Thank you,
Smita

Hi Smita,
I am not sure if this is the question you are asking for sender SOAP adapter or HTTP Adapter?
Please refer the below statement.
SOAP Adapter:
The inbound address for SOAP messages is: http://host:port/XISOAPAdapter/MessageServlet?channel=party:service:channel
HTTP Adapter:
The plain HTTP adapter is used by external systems to connect to the Integration Engine using the native HTTP interface (HTTP payload without SOAP envelope). These systems are connected using the Internet communication framework of the SAP Web Application Server. For this purpose, the Integration Engine HTTP inbound channel contains an HTTP service delivered by SAP, called /sap/xi/adapter_plain.
Queries entering the plain HTTP adapter must have the following syntax:
http://<hostname:port>/<path>?<query-string>
The query string contains the following data:
● Sender namespace ?namespace=<namespace>
●Sender interface &interface=<interface>
These details define the sender interface.
●Sender service &service=<service>
Specifies the sender service.
●Sender party (optional) &party=<party>
●Sender agency (optional) &agency=<agency>
●Sender scheme (optional) &scheme=<scheme>
If you are asking for HTTP Adapter, then I can tell you.
Please confirm.
Regards
B.Dheepa

User defined Exception in Stateless Java Class WS

Hi,
I'm experimenting with publishing some of my application's Java classes as web services under OC4J 10.1.2. I've been quite successfull to do so, but I'm stuck into a problem trying to have my class's methods throw my own type of exception and retrieving it at the client side.
I'm using the proxies downloaded from the web service's automatic HTML page, and I see in the source code version that the proxy class specifically throws a org.apache.soap.SOAPException whenever there's a fault in the response message.
What I've done is simply to create a type derived from Exception with a String (message) and an Integer (code) properties, with just my own constructor that accepts two parameters corresponding to that properties. Then I make the methods in my Java class throw that exception.
I've tried both using both an Interface and a class, and also using just a class as parameters to StatelessWebServiceServlet / Web Service Assembler.
I can't find anywhere in any docs that explains how this can be done in OC4J. I believe that throwing user-defined exceptions is included in J2EE 1.4 and so I think that there should be a way to do it in OC4J if it is compliant. Maybe there's no way, but I couldn't find any reference to that neither.
I'll greatly appreciate any light on this issue.
Thanks in advance
Juan Alvarez Ferrando
Mensaje editado por:
user517323

Thank you for your answer.
I'd like to know something more about how far does that limited support go, in case it could be of some use.
On the other hand, and just for the sake of constructive discussion, though I've contemplated the option of encoding error conditions on the result type, I always found that to be an ugly style from a interface design standpoint (not ws specifically but in general system desing terms), and when I saw a standard way to communicate custom error information separated from normal response I thought it to be the right way to go. After all, exceptions are just the way Java implements this design principle that has also found a place in most currently successful software environments (PL/SQL, .Net, ...).
I don't expect to extend the meaning and programmatic use of Java exceptions to my ws clients, but to be able to communicate detailed error condition information as the standards contemplate (faults), which as I understand from your kind answer is out of my reach in 10.1.2.
I have no experience with other ws platforms but I'd like to believe that if wsdl:fault is in the standards, those claming compliance will support it to the extent necessary to make it usable and interoperable as it is the way the standard covers error communication (and so that's why it is included in 10.1.3). I also beleve this would bring better interoperability than our home-made result types including error information, that no other application without our custom specifications can understand.
Also better interoperability would come from that than for example what I now see in 10.1.2, where all exceptions are communicated to the client with a faultcode of soap:Server, which I believe to have the standard meaning that there's a chance for the client to success retrying later without modifying it's request. This translation could be better suited for Java errors (unrecoverable) but not for all kinds of exceptions, and so in this question, custom exception support could also improve interoperability with systems that interpret SOAP fault codes.
Thank you.
Juan Alvarez Ferrando

Question: User defined acknowledgement (IDOC to SOAP)

Similar Messages

Maybe you are looking for