@zverik that's cool.
I'll have to look at some of the plugins.
`osm-lump-ways` did try to calcaulte Strahler numbers, but lots of rivers are “braided” (splitting & rejoining) so the standard algorithm gave “incorrect” results.