This feature is needed for the clean-up routine that will create data units when there are geographical areas in the exposure entities that are not covered by data units for which an input aggregated exposure model is defined. The present feature allows for the following:
- retrieval of a polygon associated with the extents covered by the aggregated exposure model (the "domain" polygon)
- retrieval of a polygon associated with an exposure entity
- verification that the latter is contained within the former
- trimming of the polygon of the exposure entity if it is not contained in the domain polygon
These checks are relevant for countries whose boundary files may contain territories around the world but for which an aggregated exposure model may be defined only in one region (e.g. a continent).
As the case in which the polygon of the exposure entity is not contained in the domain polygon may not be very common, and the domain polygon may be as large as necessary (it does not need to follow "real" borders), the verification of the former being contained within the latter is carried out using the bounding box of the polygon of the exposure entity, which is considerably faster than carrying out an unnecessary proper intersection between the two fully-detailed geometries. If the bounding box of the polygon of the exposure entity is not contained within the domain polygon, then the intersection operation is carried out.
domain_boundary_filepath is an optional parameter in the configuration file. If not provided, then the comparisons just described are not carried out (i.e. the polygon of the exposure entity is taken as read from the corresponding file).
\approve @fd