filter

FirstlineDBFilter.filter(scanlines, header)

Filter out any scanlines that existed in the previous granule.

Only works on datasets implementing get_dataname from the header.