However, in the answers no one seemed to be sure why the diodes are put into the circuit.
As shown in the schematic, each row has two diodes connected, one row has all anodes in common (Row nL) and the other row all the cathodes (Row nH).
Does anyone know why these diodes are put in the circuit as shown?
Possibly it has something to do with the "freewheeling" path, when the column or row voltage is switched off. Another idea is that by the addition of the diodes, the display can be changed per row/column instead of flipping each pixel individually. I can think of a circuit to flip a complete row at once, but usually the displays switch in a horizontal manner (the pixels flip starting from the left and moving to the right), and I don't see a way of making a simple circuit to switch the display per complete column.