What should be Switch Mediator's source path? - wso2

Question

Scenario
I've two XML-based data files, I need to use a switch mediator in my API to direct the xml file to its respective mediation flow which is implemented in the same API. The XMLs are shown below:
XML - 1
<Table>
<Trademarks>
<FileNbr>574988</FileNbr>
<DateOfRegistration>03-DEC-21</DateOfRegistration>
<Class> 43 </Class>
<ApplicantName>Ramada International, Inc.</ApplicantName>
<Address>22 Sylvan Way Parsippany, New Jersey 07054</Address>
<Title>RAMADA RESIDENCES</Title>
<Status>Trademark registered</Status>
</Trademarks>
<Trademarks>
<FileNbr>524918</FileNbr>
<DateOfRegistration>03-DEC-21</DateOfRegistration>
<Class> 37 </Class>
<ApplicantName>Muhammad Usman, Trading as, IMAMDIN ASSOCIATES</ApplicantName>
<Address>Imamdin Associates, 89/6-R,</Address>
<Title>IMAM DIN</Title>
<Status>Trademark registered</Status>
</Trademarks>
</Table>
XML - 2
<Table>
<Copyright>
<applicationno>1641/2001</applicationno>
<applicationdate>11/15/2001 12:00:00 AM</applicationdate>
<TitleEnglish>DURA PLUS</TitleEnglish>
<applicantname>FARMIGEA PAKISTAN (PVT) LTD.</applicantname>
<class>ARTISTIC WORK</class>
<City>LAHORE</City>
<Country>PAKISTAN</Country>
<status>Application Registered</status>
</Copyright>
<Copyright>
<applicationno>1644/2001</applicationno>
<applicationdate>11/15/2001 12:00:00 AM</applicationdate>
<TitleEnglish>OCUGEL FARMIGEA</TitleEnglish>
<applicantname>FARMIGEA PAKISTAN (PVT) LTD.</applicantname>
<class>ARTISTIC WORK</class>
<City>LAHORE</City>
<Country>PAKISTAN</Country>
<status>Application Registered</status>
</Copyright>
</Table>
Question
I need to specify the Source XPath for the Switch mediator. In
the above scenario how can I differentiate between the two XML data files
based on the second nodes i-e Trademarks & Copyright.?
Which property should I used for the above requirement?

Answer 1

You can use the xpath local-name() function, like below. But keep in mind, that only checks the first element after <Table>. I also like to use upper-case() function, to avoid upper/lower case issues. And the cases are with regex start ^ and end $ matches. Working example below.
<switch source="upper-case(local-name(//Table/*))" xmlns:ns="http://org.apache.synapse/xsd">
<case regex="^TRADEMARKS$">
<payloadFactory media-type="json">
<format>{"Trademarks":"DoSomething"}</format>
</payloadFactory>
</case>
<case regex="^COPYRIGHT$">
<payloadFactory media-type="json">
<format>{"Copyright":"To something else"}</format>
</payloadFactory>
</case>
<default>
<payloadFactory media-type="json">
<format>{"Error":"Invalid table name"}</format>
</payloadFactory>
</default>
</switch>
<respond/>