OpenAtlas

Currently, the ARCHE export will dump the complete model, public and import schema without differentiation between project entries (defined by type ID) or other entries.

Questions, should the dump...

• update an existing database (this means, losing the unique IDs and can be difficult to develop)
• be inserted into a clean OpenAtlas instance (4_data_type.sql can be replaced with the dump)
• just exist as @INSERT@s and interested people have to deal how to get the data out of it

Keep only data if entities have a corresponding case study:

• Annotation (text and image)
• Geometries
• Hierarchy and Types
  and delete all other