How 'test automation' identifies the test cases to be updated - JUnit xml results import

[Florencia Ferrari]

I am using 'Test Automation' (NOT Xray or something like that) for Zephyr and I cannot update the results of test cases that already exist. The task is in progress until I have to change its status manually. If I create a new folder for the import result, new cases are created with new keys and correct results, if I want to update these cases, again the task does not work.

Can anyone tell me how the test case names match with an example? Is "name", "class name", a combination of both or something else used from JUnit xml results? Do I have to include the key (eg PROJ-123) within the test case name ("name" in the junit xml output) or how does it find the test case I want to update?

I'm confused. 

Thanks!! 

Florencia