No uniform realisation is possible.

Incidence matrix according to Dynkin symbol

x3/2β3x

demi( .   . . ) | 12 |  2  2 1 | 2 1 2
----------------+----+---------+------
both( x   . . ) |  2 | 12  * * | 1 1 0
both( .   . x ) |  2 |  * 12 * | 1 0 1
sefa( .   β3x ) |  2 |  *  * 6 | 0 0 2
----------------+----+---------+------
both( x   . x ) |  4 |  2  2 0 | 6 * *
      x3/2β .   ♦  3 |  3  0 0 | * 4 *
      .   β3x   ♦  6 |  0  3 3 | * * 4

(Here sefa( x3/2β3x ) would degenerate to a digon which thus has been reduced.)

starting figure: x3/2x3x