jtambasco / modesolverpy Goto Github PK

Dear Jtambasco,

I copied your program on modesolverpy for mode hybridization. The program does not give the output plot as shown on your page, especially the structure_n_effs.png file is supposed to have TE fraction shown as well but it does not appear on the plot. Please, I request you to make necessary adjustment.

how could we have python2 and python3 compatibility?

On line 372 in structure_base.py the height of a Slab Class feature is first discretised and then another y_step is added to the height:
height_discretised = self.y_step*((height // self.y_step) + 1)

Even if you choose a stepsize for the layers, such that is correctly discretised (e.g. 10nm stepsize for a 100nm layer), this line of code will still add another self.y_step (in this case 10nm) to the layer thickness and cause it to be larger (110nm) instead. While this might not be an issue for small stepsizes or a low amount of layers, this causes problems for larger stepsizes or many layers, such as in Bragg-mirror structures.

I am unsure why the self.y_step is added, maybe there is a solid reason for it, but in my case it simply causes the layers to have wrong thicknesses.

Getting the following error on windows:

`The system cannot find the path specified.

"title='Mode 0 |E_{y}| Profile, n_{eff}: 2.482, λ = 1.550 µm "
^
line 0: invalid command

Invalid Parameter - /modes_semi_vec
The system cannot find the path specified.
Invalid Parameter - /modes_semi_vec`

Is there any way I could interface modesolverpy with matplotlib instead of gnuplot (if this is the issue)?

1. The following property definitions are inconsistent with one another:

@Property
def x(self):
'''
np.array: The grid points in x.
'''
if None not in (self.x_min, self.x_max, self.x_step) and
self.x_min != self.x_max:
x = np.arange(self.x_min, self.x_max+self.x_step-self.y_step*0.1, self.x_step)
else:
x = np.array([])
return x

@property
def y(self):
    '''
    np.array: The grid points in y.
    '''
    if None not in (self.y_min, self.y_max, self.y_step) and \
            self.y_min != self.y_max:
        y = np.arange(self.y_min, self.y_max-self.y_step*0.1, self.y_step)
    else:
        y = np.array([])
    return y

In the definition for x, x_max will be included in the grid. For y, it won't. Is this intended?

2. There is a typo in the line " x = np.arange(self.x_min, self.x_max+self.x_step-self.y_step0.1, self.x_step)". You are substracting self.y_step0.1, but it should be self.x_step*0.1. Probably hasn't been seen yet because I guess most people use uniform grid spacing and then it doesn't matter