Currently the code does not precisely check whether a source lies in the actual footprint of an image - we only check whether the source is within some distance of a beam (currently 1.5 deg) and assume that if that is the case, then it is within the image.
Creating a fits cutout that uses coordinates outside of an image throws an astropy.nddata.utils.NoOverlapError. We currently get around this by wrapping the cutout dataframe generation (inside Query._grouped_fetch_cutouts) in a try/except statement. However, this means that if multiple sources are associated with an image, but one is not inside the image footprint (and therefore throws a NoOverlapError rows in the cutout_df are not generated for all sources associated with that image.
Currently the code does not precisely check whether a source lies in the actual footprint of an image - we only check whether the source is within some distance of a beam (currently 1.5 deg) and assume that if that is the case, then it is within the image.
Creating a fits cutout that uses coordinates outside of an image throws an
astropy.nddata.utils.NoOverlapError
. We currently get around this by wrapping the cutout dataframe generation (insideQuery._grouped_fetch_cutouts
) in a try/except statement. However, this means that if multiple sources are associated with an image, but one is not inside the image footprint (and therefore throws aNoOverlapError
rows in the cutout_df are not generated for all sources associated with that image.