Force Altair to Share Axis / Get Domain Range

Hi,

I am looking to figure out how to either: a) Force a concatenated chart to share an axis b) Retrieve the domain (y and/or x limits) of another chart

You can see in the below that I have concatenated a series of charts. I would like the histograms on the right to have the same y domain as the line charts on the left.

Any ideas?

Thanks!

chart1 = alt.Chart(temp.loc[temp['Series'] != 'Error', :]).mark_line().encode(
    alt.X('index:T'),
    alt.Y('value:Q'),
    alt.Color('Series:N', legend=alt.Legend(orient='left')),
).properties(
    height=400,
    width=1000
)

chart2 = alt.Chart(temp.loc[temp['Series'] == 'Error', :]).mark_line().encode(
    alt.X('index:T'),
    alt.Y('value:Q'),
    color='Series:N'
).properties(
    height=200,
    width=1000
)

chart3 = alt.Chart(temp.loc[temp['Series'] == 'Error', :]).mark_bar().encode(
    alt.X('count()'),
    alt.Y('value:Q', bin=alt.Bin(maxbins=100)),
    alt.Color('Series:N')
).properties(
    height=200,
    width=200
)

chart4 = alt.Chart(temp.loc[temp['Series'] != 'Error', :]).mark_bar(opacity=.4).encode(
    alt.X('count()'),
    alt.Y('value:Q', bin=alt.Bin(maxbins=100)),
    alt.Color('Series:N')
).properties(
    height=400,
    width=200
)

alt.vconcat(
    alt.hconcat(
        chart1,
        chart4
        ),
    alt.hconcat(
        chart2,
        chart3
        )
)

[jakevdp]

You can use the resolve_scale method to specify shared scales. Here is a quick example:

import altair as alt
import pandas as pd
import numpy as np

np.random.seed(42)

df = pd.DataFrame({
    'x': np.linspace(0, 10, 500),
    'y': np.random.randn(500).cumsum()
})

line = alt.Chart(df).mark_line().encode(
    x='x',
    y='y'
)

hist = alt.Chart(df).mark_bar().encode(
    x='count()',
    y=alt.Y('y', bin=alt.Bin(maxbins=10), title=None)
).properties(
    width=100
)

alt.hconcat(
    line,
    hist
).resolve_scale(
    y='shared'
)

[ajennings6767]

Ah - that’s too bad.

Thanks so much for your help!