The intent of this line was to produce a vector containing the index of each number in priority_order:
np.arange(len(layout.channels))[priority_order]
but this does nothing; the result is the same as priority_order.
In this case the spec and the original intent was correct, see section 7.3.6; this behaviour makes no sense. The existing tests aren't too bad, it just looks like we were unlucky and didn't hit any cases which were incorrect.
This is quite unlikely to have happened in the real world; the result would have been that if the position was exactly between 2 channels, the 'wrong' one may have been picked, resulting in left-right inconsistency, for example.
The intent of this line was to produce a vector containing the index of each number in priority_order:
but this does nothing; the result is the same as priority_order.
In this case the spec and the original intent was correct, see section 7.3.6; this behaviour makes no sense. The existing tests aren't too bad, it just looks like we were unlucky and didn't hit any cases which were incorrect.
This is quite unlikely to have happened in the real world; the result would have been that if the position was exactly between 2 channels, the 'wrong' one may have been picked, resulting in left-right inconsistency, for example.