Right now the category is defined by the attribute of the root node. We would like to add another category to each test inside of a report to improve the filter capabilities in the global test report. For us one category per test would be sufficient, but you could argue that multiple categories would be helpful too.

To explain what we are doing and why it would help us:
We run the same tests with different test data. Each run creates a different report, where the category represents the test data we used. So now we can filter by the test data. What we liked to do is filter by the tests.

• Category A
  • Test A.1
  • Test A.2
• Category B
  • Test B.1
  • Test B.2

So in this example A.1 and B.1 would both have the category "1".