Some performance tests had invalid fixture data, so the from_dict function never really finished successfully. Exceptions have been thrown and thus the performance reading was unusable. Fixed these errors and added some basic assertions. It is not necessary to check the result in details—that's what regular tests are for.
Also added a few tests to bump the coverage back to 100%, from 99%.
Implemented changes
from_dictfunction never really finished successfully. Exceptions have been thrown and thus the performance reading was unusable. Fixed these errors and added some basic assertions. It is not necessary to check the result in details—that's what regular tests are for.