The aquifer dictionary must remain intact when importing non-unique soils. This invalidated the previous approach to identifying the soils by name in a case-insensitive way. After some puzzling, I found a better way. Both unit tests that concerns the import of two MDB-files now work. Related to Issue [DSB-502]
Ignore the case of the soil names when importing a database with soil types - "zand" and "Zand" and "ZAND" are therefore considered one and the same soil.
This is covered by the unit test TestCaseSensitivityOfImportedSoilNames() Related to Issue [DSB-502]