The mapping template extracts the patient FHIR resource from a FHIR bundle and converts it to relational format, and the attached file includes an intelligent structure model that contains the schema for the conversion. To create a different intelligent structure model, you can use a different XML file as input to the intelligent structure model and regenerate the schema.
The following image shows the transformations in the mapping template:
The targets are parameterized in the mapping template, so you can perform one of the following tasks:
•Replace the parameters with specific connections and target objects in the Target transformations.
•Leave the parameters in the Target transformations and specify targets in the mapping task or use a parameter file.
The mapping template contains the following transformations:
1. Source transformation
The Source transformation uses a FHIR connection to submit a GET request for a patient resource by querying patient details. The sample query references a patient’s given name.
2. Expression transformation
The Expression transformation removes special characters in the XML data that can cause validation issues, such as converting ampersands to &.
3. Data Services transformation
The Data Services transformation uses the FHIR validation service from the data services repository to validate the FHIR resource.
4. Target transformations to troubleshoot validation errors
The Target transformations that are downstream of the Data Services transformation write errors and error flags to target files. If the FHIR validation service finds errors in the incoming FHIR resource, you can use the error messages to troubleshoot the upstream data flow. To write the data to the target files, you can use any target connection that writes XML data, such as a flat file or relational connection that can hold XML data.
5. Structure Parser transformation
The Structure Parser transformation uses the intelligent structure model to convert the XML data to relational format.
6. Target transformations to write relational data
The Target transformations write the patient elements to a relational target, such as the patient, patient identifier, and patient address elements. To write the data to a relational target, use a relational connection.
