Even if renumbering the mesh, if the extent is not changed (by a rotation e.g.), the order of liquid boundaries will not change even the order is changed in the boundary conditions file.
Everything is there !
I didn't know that. I knew about the "most south-west", and for the counter-clockwise identification order. But I thought that the order of the limits, as presented in the CLI, imposed itself on the model.
Thanks for the explanation and the way to check all this.
Have a nice day,
Laurie