Originally posted by: SystemAdmin
We would like to get your comment and feedback on one of our suggestions with regard to improving the performance through WTX for XPath handling over Java. Below are the functionalities that we currently do in Java.
Two inputs: XML message and a metadata with XPaths which will be evaluated in the XML message
1. Detect if all XML elements exist:
a. E.g: The element Message/patAdmission/patient[*]/patIDList/identity/idNum needs to be checked for all instances of Patient.
2. Remove some XML nodes(collection of elements) in the XML:
a. E.g: the element tree Message/pharmDrugOrdUpdCan/pharmOrder[*]/finTrans shoul be removed at all instances of pharmOrder but all other children under pharmOrder should remain.
3. Create new XML elements with concatenating string values:
a. E.g: Enhance the same XML message with inserting a new element Message/clinicalLabOrderMsg/clinicalLabOrder[*]/orders[*]/commOrd[*]/ordAppID/ordBusKey with values from other elements like Message/clinicalLabOrderMsg/clinicalLabOrder[*]/orders[*]/commOrd[*]/ordAppID[*]/placerID/entityID, Message/clinicalLabOrderMsg/encounter/eocNum/idNum and Message/clinicalLabOrderMsg/patient/patIDList/identity/idNum with or w/o some conditions.
In the Java implementations , we do use EvaluateXPath and “//” so that we get the all the instances of the element and evaluate the XPath we create for above functionalities.
So we are trying to see if it is possible to get above functionalities working in WTX as expected. This is where i need your inputs to share with us if this is possible to get all iterations of an XPath and do the EvaluateXpath in WTX and if so how complex it would be.
#IBM-Websphere-Transformation-Extender#DataExchange#IBMSterlingTransformationExtender