Issue is De-serializing nested date type using Axiss 1.4

Question

We are using Axis 1.4 and face issue during deserialization in handling date types. The scenario is explained below. I couldn’t find enough information on net on below and hence asking question to the apache team. Please note when we switch to Axis 2, issue is not seen. Upgrading may not be a simple option for us as of now, and hence I was evaluating possible workarounds or solution if any.
The XSD contains type “DateofBirthType” which is of type xsd:date
Child XSDs contains element “DateofBirth” of type “DateOfBirthType”
When Java client invokes the web service, IllegalArgumentException is seen while de-serializing this “DateOfBirthType” element. If we remove “DateOfBirthType” and map it directly to xsd:date type all works fine.
We have generated the stub using Axis 1.4 and trying to access the RetrieveIndividualCustInformation  operation through stub, but getting the (IllegalArgumentException) while axis tries to deserialize the response for all the fields having data types as date which is indirectly mapped (nested mapping) as shown below.
e.g. IndividualInfo-----DateOfBirth(Type—DateOfbirth ---Which is of Type Date)-_you will face illegal argument exception.
So, when we changed the type to simple date (not nested),then it worked fine. So it confirms that this problem is only for type date which is clearly seen in Axis 1.4 client.
Please let me know if anyone faces similar/same issue or even know any such issue already persists and there are no possible workarounds and/or fixes for this. In that case we can put up case to migrate to Axis2.
Please reply ASAP.

Answer 1

Hi, Shank,
My guess is that you want:
, dense_rank () over (order by  tmstp  desc)  AS rn RANK means you'll skip numbers when there's a tie. For example, if 3 rows all have the exact same latest tmstp, all 3 rows would be assigned number 1, RANK would assign 4 to the next row, but DENSE_RANK assigns 2.
"PARTITION BY x" means that you want a separate series of numbers (starting with 1) for every value of x. If you want just one series of numbers for the whole result set, then don't use a PARTITION BY clause at all. (PARTITION BY is never required.)
Maybe you wnat to PARTITON BY cd. I can't tell without some sample data, and an explanation of why you want the given results from that data.
You certainly don't want to PARTITION BY the same expression that you ORDER BY; that just means all rows tie for #1.
I hope this answers your question.
If not, post a little sample data (CREATE TABLE and INSERT statements, relevant columns only) for all tables involved, and also post the results you want from that data.
Explain, using specific examples, how you get those results from that data.
Simplify the problem as much as possible.
Always say which version of Oracle you're using.
See the forum FAQ {message:id=9360002}
Edited by: Frank Kulash on Aug 1, 2012 1:20 PM