Description Andi Bur 2013-01-29 08:36:42 EST

The contents of an ITable can be imported from and exported to an AbstractTableFieldData which stores the table data within a list of object arrays. The import and export routines are index-driven, based on IColumn indexes. This kind of data structure makes it very hard to extend (e.g. adding additional columns at arbitrary indexes to an existing table using subclassing).

To solve this limitation, add the option to use a bean-based table field data. Every row should be represented by a bean and the import and export to and from an ITable should be based on property names and IColumn.getColumnId().

Further columns of an ITable can be replaced by subclasses using @Replace. However, the data type of a replaced column cannot be changed. Hence the row data must not contain new properties for replaced columns.

Scout SDK: add a new source builder for generating bean-based table field data classes. The default is still to generate an object array based AbstractTableFieldData.