Boolean Slicing Across an Array

[Scico5]

Hello! I'm trying to do some pretty basic slicing to find values in column 3 where the value in column one is between nmin and nmax. In numpy you can do this:
import numpy as np
a=np.array([[1,2,3],[2,3,4],[5,6,7]])
b=a[:,1]==3
c=a[b]

However is ulab I get the error:
ValueError: array and index length must be equal

I saw in the documentation that you may get a not implemented error for trying to slice more than a 1D vector. Is this still the case? And is there a sneaky way around this?

I have my own sneaky way that looks like this:
d1=abs(arr[:,0])==old
sub1=arr[:,2][d1]
d2=abs(arr[:,1])==new
sub2=arr[:,2][d2]
ssub1=set(sub1[:])
ssub2=set(sub2[:])
otn=list(ssub1.intersection(ssub2))
otn.sort(reverse=True)

But the set functions are very slow and are bad for my application.

Thanks!

[Scico5]

I believe I have an answer based on a bug report from this repo.

You can do this using np.where:

import ulab.numpy as np
a=np.array([[1,2,3],[2,3,4],[5,6,7]])
b=a[:,1]==3
c=b.reshape((len(b),1))
d=np.where(c,a,np.nan)

Then in the d array you have converted all invalid rows to nan's, Now you can repeat the process on d using the next boolean operation on the other column of interest and you're left with only valid data.