hline(0) not in correction position when plotting with object model

vline(0) appears in the right place, but hline(0) does not. hline(0) seems to always be positioned at the minimum y value on the axis, rather than at y=0.

from bqplot import *
from bqplot.pyplot import vline, hline
import numpy as np

np.random.seed(3)
x_data = np.random.randn(1, 10)[0] + 1
y_data = np.random.randn(1, 10)[0] + 1

sc_x = LinearScale()
sc_y = LinearScale()
scales_dict = {'x': sc_x, 'y': sc_y}

scatt = Scatter(x=x_data, y=y_data, scales=scales_dict)

ax_x = Axis(scale=sc_x, label='x')
ax_y = Axis(scale=sc_y, orientation='vertical', label='y')

vline_mark = vline(0, scales = scales_dict)
hline_mark = hline(0, scales = scales_dict)
fig = Figure(marks=[scatt, vline_mark, hline_mark], axes=[ax_x, ax_y], animation_duration=1000)

display(fig)

When updating mark positions, vline behaves as expected but hline moves to new (but still incorrect position).

fig.marks[0].x = fig.marks[0].x - 2
fig.marks[0].y = fig.marks[0].y - 2

bqplot 0.11.0 IPython 7.0.1

[martinRenou]

I suppose, yes. Maybe you would need to create a Line mark that has its own x and y scales?

[DougRzz]

I tried to find a work around. The hline_mark.y value seems to be a fraction of the height in y. You would have to add the min + max to the scale. Its not quite right but its quite close…

from bqplot import *
from bqplot.pyplot import vline, hline
import numpy as np

np.random.seed(3)
x_data = np.random.randn(1, 10)[0] + 1
y_data = np.random.randn(1, 10)[0] + 1

sc_x = LinearScale()
sc_y = LinearScale()
scales_dict = dict(x = sc_x, y= sc_y)

scatt = Scatter(x=x_data, y=y_data, scales=scales_dict)

ax_x = Axis(scale=sc_x, label='x')
ax_y = Axis(scale=sc_y, orientation='vertical', label='y')

vline_mark = vline(0, scales = scales_dict)
hline_mark = hline(0, scales = scales_dict)
# hline_mark.scales = scales_dict
fig = Figure(marks=[scatt, vline_mark, hline_mark], axes=[ax_x, ax_y], animation_duration=1000)
fig.padding_y =0
sc_y.max = 5
sc_y.min = -3

minMaxFraction = (-sc_y.min ) /(sc_y.max  - sc_y.min )
hline_mark.y=[minMaxFraction,minMaxFraction]
display(fig)