The ADBC adapter internally handles all string data as UTF-8, and provides the same internationalization support as the correlator. The correlator internally uses the C programming language locale for formatting string versions of numeric values, so there can be conditions under which the ODBC and JDBC drivers may use a locale that is not compatible with the English numeric separator format used in ADBC. In locales that do not use English numeric separators, the ODBC and JDBC drivers for some SQL vendors may not correctly handle numeric values passed from the correlator. To address these cases, the ADBC adapter configuration property NumericSeparatorLocale allows the numeric separator used in the adapter to be changed to match the one used by the correlator. The property can be set to one of three values:

A value of C causes the adapter's numeric separator locale to match that used by the correlator, so that the JDBC and ODBC drivers correctly handle the numeric values. The value Native causes the adapter to set the locale to the system default. This value is not generally needed and was added for future use and for special cases in which technical support would direct it to be set. If you notice incorrect numeric values when inserting or querying data from the database when running in a locale that doesn't use the English-style numeric separators, then changing the NumericSeparatorLocale property to C should correct the problem. In Software AG Designer, you can access this property by using the XML Source tab in the adapter editor.