computeReeb error

[TobiasRobertEeksman]

I got an error using the computeReeb() function. We were able to create two cases, where one works and one doesn't.
We are using the x-value as the height map and create lowerstar filtrations accordingly.

For the case:

    V = np.array([[0,0,0], [1,0,0],
                    [2,1,0]])
    F = np.array([[0,1], [0,2], [1,2]])

The calculation works and we do get a ReebGraph.

However for the case:

    V = np.array([[0,0,0], [1,0,0],
                    [0,1,0]])
    F = np.array([[0,1], [0,2], [1,2]])

The computation doesn't work anymore and we get this error:

Traceback (most recent call last):
  File "c:\Users\Tobias\OneDrive - Schulen kvBL\Dokumente\ETH\cereeberus\ceREEBerus\test.py", line 223, in <module>
    R1 = computeReeb(st1)
  File "C:\Users\Tobias\AppData\Local\Programs\Python\Python310\lib\site-packages\cereeberus\compute\computereeb.py", line 139, in computeReeb
    R.add_edge(e, nextNodeName)
  File "C:\Users\Tobias\AppData\Local\Programs\Python\Python310\lib\site-packages\cereeberus\reeb\reebgraph.py", line 386, in add_edge
    raise ValueError(
ValueError: The vertex e_0 must be in the Reeb graph to add an edge between them.

We think the issue lies in the fact, that in the second case, two of the vertices get the same filtration value. Here's our full implementation for checking:

from cereeberus import LowerStar, computeReeb
import numpy as np
import matplotlib.pyplot as plt

def f_x(pts: np.ndarray) -> np.ndarray:
    return pts[:, 0].astype(float)

def create_lower_star(V,F):
    st = LowerStar()

    for i,v in enumerate(V):
        st.insert([i])

    for face in F:
        st.insert(face)

    for i, v in enumerate(V):
        st.assign_filtration([i], f_x(V)[i])

    return st

if __name__ == "__main__":

    #Example 1: which works
    V = np.array([[0,0,0], [1,0,0],
                    [2,1,0]])
    F = np.array([[0,1], [0,2], [1,2]])

    # # Example 2: which doesn't work
    # V = np.array([[0,0,0], [1,0,0],
    #                 [0,1,0]])
    # F = np.array([[0,1], [0,2], [1,2]])

    st1 = create_lower_star(V,F)
    R1 = computeReeb(st1)
    R1.draw(cpx = 2.0)
    plt.show()