Create an array with a hole at the center

from wolfhece.wolf_array import header_wolf, WolfArray
from wolfhece.PyVertexvectors import vector, wolfvertex

import numpy as np
from math import pi, sin, cos

import matplotlib.pyplot as plt

Array part

h = header_wolf()
h.set_origin(10.,20.) # to be non-zero
h.set_resolution(.5, .5)
h.shape = 201, 101

wa = WolfArray(srcheader=h)
print(wa)

bounds = wa.get_bounds()
print(bounds)
print(f'The center of the array is at', {(bounds[0][0]+bounds[0][1])/2.}, {(bounds[1][0]+bounds[1][1])/2.})

Shape  : 201 x 101
Resolution  : 0.5 x 0.5
Spatial extent :
   - Origin : (10.0 ; 20.0)
   - End : (110.5 ; 70.5)
   - Width x Height : 100.5 x 50.5
   - Translation : (0.0 ; 0.0)
Null value : 0.0


([10.0, 110.5], [20.0, 70.5])
The center of the array is at {60.25} {45.25}

x_center, y_center = wa.ij2xy(101, 51, aswolf=True)  # 11, 6 is the center of the array as the array is 201x101

print(f"center = {x_center}, {y_center}")

center = 60.25, 45.25

Vectorial part

center_circle = wolfvertex(x_center, y_center)
radius = 5.0

circle = vector(name='circle')
circle.add_vertex([center_circle] *3) # add 3 times the center circle to make a "triangle"
circle.buffer(radius, 5) # create a buffer (-> a cricle) with radius 2.0 and a resolution of 5, 5 segments in a quarter of circle (see https://shapely.readthedocs.io/en/stable/reference/shapely.buffer.html for details)
circle.reset_linestring()

print(circle.nbvertices)

21

Plot just to be sure…

fig, ax  = plt.subplots()
circle.plot_matplotlib(ax)
ax.set_aspect('equal', adjustable='box')

Combine array and vector

wa.mask_insidepoly(circle, method='shapely_strict') # mask inside the circle with value 1
wa.nullify_border(1) # nullify 1 cells around the array

Plot just to be sure…

fig, ax = wa.plot_matplotlib()
fig.set_size_inches(10, 5)

fig, ax = wa.plot_matplotlib()
fig.set_size_inches(10, 5)
circle.plot_matplotlib(ax) # plot the circle in red

Get the external contour

sux, suy, contour, interior = wa.suxsuy_contour()

if interior:
    print('There is an interior masked part in the array') # the circle is inside the array

There is an interior masked part in the array

If we plot the contour, the inner part is linked to the outer one.

fig, ax = wa.plot_matplotlib()
fig.set_size_inches(10, 5)
contour.plot_matplotlib(ax) # plot the circle in red

print('Number of vertices : ', contour.nbvertices) # Number of vertices in the contour

notused = [curvert for curvert in contour.myvertices if not curvert.in_use]

print('Number of segments not used : ', len(notused)) # Number of vertices/segments not used -- During the meshing process, some segments are not used (links from the external to the internal parts).
# This is normal and helps avoid numerical problems during the process.
#
# Note: An intersection between two segments is not necessarily on these segments... ;-)

Number of vertices :  675
Number of segments not used :  2