It would be useful to have aggregated code coverage % for the directory, not only for each file.
Aggregated data for the whole test seem to be less useful, as they will also include some file from packages etc., so having them by directory seems to be the right granularity.
It would be useful to have aggregated code coverage % for the directory, not only for each file.
Aggregated data for the whole test seem to be less useful, as they will also include some file from packages etc., so having them by directory seems to be the right granularity.