XyData does not behave as expected

[mbercx]

Describe the bug

The behaviour of XyData is not what one would expect when specifying multiple (or even a single) array(s). It is also not in line with the documentation:

https://aiida.readthedocs.io/projects/aiida-core/en/stable/topics/data_types.html#xydata

Steps to reproduce

from aiida import orm, load_profile

load_profile()

import numpy as np

xy = orm.XyData()

xy.set_x(np.array([1, 2, 3]), 'x', 'unit_x')
xy.set_y(np.array([1, 2, 3]), 'y', 'unit_y')
xy.set_y(np.array([1, 2, 3]), 'z', 'unit_z')

print(xy.get_arraynames())
print(xy.get_y())

Gives output:

['x_array', 'y_array_0']
[('z', array([1, 2, 3]), 'unit_z')]

Expected behavior

1. The arraynames should respect the names I specify in the set_x and set_y commands.
2. It should be possible to have multiple "y" arrays in one XyData. Here it seems it is overridden (without warning).

Your environment

• Operating system [e.g. Linux]: macOS
• Python version [e.g. 3.7.1]: 3.10.16
• aiida-core version [e.g. 1.2.1]: 2.6.3

EDIT: Also tested on latest origin/main.

[mbercx]

Ah, it seems I misunderstood the behaviour of set_y, since it's not the same as set_x. You need to set all y arrays in one execution of set_y, and cannot set them one by one.

I suppose we can just make this a documentation issue then, because it was not intuitive to me.

EDIT: Proper example of usage:

from aiida import orm, load_profile

load_profile()

import numpy as np

xy = orm.XyData()

xy.set_x(np.array([1, 2, 3]), 'x', 'unit_x')
xy.set_y([np.array([1, 2, 3]),np.array([4, 5, 6])], ['y', 'z'], ['unit_y', 'unit_z'])

print(xy.get_arraynames())
print(xy.get_y())

Results in:

['x_array', 'y_array_0', 'y_array_1']
[('y', array([1, 2, 3]), 'unit_y'), ('z', array([4, 5, 6]), 'unit_z')]

I would still expect get_arraynames() to return the array names provided to the set_x and set_y methods.

[ayushjariyal]

@mbercx Is this issue still open?

[mbercx]

@mbercx Is this issue still open?

It is, cheers for working on the docs. I'm travelling tomorrow, so a bit busy atm, but will try and have a look at your PR next week! 👍