I’m using DBUnit for an integration test, and before executing the test code I’m running into this error:
badges.track_types data type (2003, '_text') not recognized and will be ignored. See FAQ for more information. org.dbunit.dataset.NoSuchColumnException: badges.TRACK_TYPES - (Non-uppercase input column: track_types) in ColumnNameToIndexes cache map. Note that the map's column names are NOT case sensitive.
the column that is ignored is a list of enums. In the dataset it’s written like this :
<?xml version='1.0' encoding='UTF-8'?> <dataset> // More info ... <badges name="30'000" description="30k a day" image_name="30000.png" threshold_val="30000.00000000" has_many="true" id="45" track_types="{TRACK_GENERIC}" "/> </dataset>
I looked in the DBUnit FAQ and saw this issue , that says that I have to override the isEnumType() method to support my enum is Postgresql, so I did this:
/** * Override method to set custom properties/features */ protected void setUpDatabaseConfig(DatabaseConfig config) { config.setProperty(DatabaseConfig.PROPERTY_DATATYPE_FACTORY, new PostgresqlDataTypeFactory(){ public boolean isEnumType(String sqlTypeName) { if(sqlTypeName.equalsIgnoreCase("track_types")){ return true; } return false; } }); config.setProperty(DatabaseConfig.PROPERTY_METADATA_HANDLER, new DefaultMetadataHandler()); }
But I still get the same error, and I don’t know why. Maybe I’m not overriding well the method? Maybe it’s not even the cause of my problem? If you need any other code just ask, thanks!
Advertisement
Answer
Well… I wasn’t able to solve this exactly but managed to solve it by a workaround.
This error comes because of the @DatabaseSetup annotation. If I did this process without using it, it still throws a ‘column not recognized’ error, because it doesn’t recognize postgres arrays (that was the root cause that I have) but I could solve it by creating a new DataTypeFactory that extends from the default one:
public class PsqlArrayDataTypeFactory extends DefaultDataTypeFactory { public DataType createDataType(int sqlType, String sqlTypeName) throws DataTypeException { if (sqlType == Types.ARRAY) { return DataType.VARCHAR; } return super.createDataType(sqlType, sqlTypeName); } }