XMLnodeToDocument Improper handling of arrays

Hello Forum - we are on webmethods IS 6.5. I am receiving an HTTP send order xml document (see attached example). I used one of the sample orders XML as source to create a new document and that is what I am specifying in XMLnodeToDocument as documentTypeName. I have arrays set to FALSE. If you look at the screen prints in the other document you will see first the document definition and then the result when the order is processed. The arrays get properly identified BUT they only contain the last record of each array in the document. IF there were two street fields only the last one is shown, if there were three items ordered then ITEMOUT only shows ordered item number 3 as record 0. So only the last record for each array is recognized as record 0. What is going on?

Any ideas and suggestions would be very much appreciated. This is running on a Windows 2003 Server install.
Thanks Mike - I hope the attachments help to explain the issue
message.xml (14.5 KB)
XMLnodeToDocumentIssue cXML.doc (88 KB)

Hi,

I tried your xml in my 7.1 developer and it worked perfectly fine.

How did you generate xml node object?

Regards,
Sumit

Hi - thanks for checking this out for me.

In my “manual” development flow I get the xml file via pubgetfile and then I do a pub.xml:xmlStringToXMLNode before I do the XMLNodeToDocument statement. I am wondering whether it is my version of webmethods. I found a similar issue reported on the webmethods forum where the solution was to download some fixes for IS. Unfortunately I can’t get them from webmethods since we got IS through Oracle/Peoplesoft’s WSG. I also opened up a trouble ticket with them just if that is really the case.

Thank you for checking…
Mike

Hi,

One more thing you can check, in xmltoNode did you set isXML to true?
If you have done so and still get error, probably you can check for fixes.

Regards,
Sumit

Sumit,

thanks for your quick replies on this. I figured out what the problem was. There were some nested variables with the same name which caused the mapping to fail. Once I removed the nesting all the arrays were mapped correctly. The reason I had not seen it at first was, because those were a lot further down in the document. I kept looking at the first arrays in the document and they were layed out perfectly fine but still came across wrong too during the mapping.

Now I just need to convince the designer of the XML document to change it so I can import it properly.

Thanks
Mike