Empty String on axis ticks changes plot size

[JWagnerDE]

If setting the tick mode to 'array' on y or x axis and giving a ticktext with one ore more elements as empty string, the size of the plot will change.

Normal	Bug

This effect can be more drastically if using subplots. One time the plot got shrunk down to ~20% in size.

To reproduce just replace one of the strings 'Text *'in line 17 with an empty string "".

Code to reproduce

import numpy as np

import plotly.graph_objects as go

np.random.seed(42)

df = pd.DataFrame(np.random.randint(0,4, 100))

fig = go.Figure()

fig.update_layout(title_text='Test plot', height=900, width=1600)

fig.add_trace(go.Scatter(x=df.index, y=df[0]))

fig.update_yaxes(dict(tickmode = 'array', tickvals = [0,1,2,3], ticktext=['', 'Text 1', 'Text 2', 'Text 3']))

fig.write_image('test' + '.png')

[kvanderwijst]

This problem becomes even more problematic when using subplots:

Normal	Bug

Code used:

fig = make_subplots(1,2)
fig.add_scatter(x=[1,2,3], y=[3, 8, 1], row=1, col=1)
fig.add_scatter(x=[1,2,3], y=[5, 0, 10], row=1, col=2)
fig.update_xaxes(tickmode='array', tickvals=[1,2,3], ticktext=['A', '', 'C'], col=2)

[mathijsvdven]

I am experiencing the same problem and ended up using a workaround. Using a newline character ('\n') instead of an empty string ('') seems to solve the formatting issue.

Tickvals: [-2.0, -1.699, -1.523, -1.398, -1.301, -1.222, -1.155, -1.097, -1.046, -1.0, -0.699, -0.523, -0.398, -0.301, -0.222, -0.155, -0.097, -0.046, 0.0, 0.301, 0.477, 0.602, 0.699, 0.778, 0.845, 0.903, 0.954, 1.0, 1.301, 1.477, 1.602, 1.699, 1.778, 1.845, 1.903, 1.954, 2.0] Ticktext: ['0.01', '\n', '\n', '\n', '0.05', '\n', '\n', '\n', '\n', '0.1', '\n', '\n', '\n', '0.5', '\n', '\n', '\n', '\n', '1', '\n', '\n', '\n', '5', '\n', '\n', '\n', '\n', '10', '\n', '\n', '\n', '50', '\n', '\n', '\n', '\n', '100']