Second x axis or x labels

Question

I would like to ask you if there is a way how to create a second x-axis in Altair (put the daytime on top)?

bar_cumm_TAGP = alt.Chart(df_TAGP_2[df_TAGP_2['simulation_date'] == '2020-10- 
15']).mark_bar().encode(
        alt.X('DVS:O'),
        alt.X2('day:O'),
        alt.Y(f'TAGP:Q', axis=alt.Axis(titlePadding=10, titleFontSize=14), title=""),
        alt.Color(f'TAGP:Q', scale=alt.Scale(scheme='greens'))).properties(width=850, height=350)

line = alt.Chart(df_TAGP_2).mark_line(interpolate='basis', strokeWidth=2, color='#FB0000').encode(
alt.X('DVS'),
alt.Y('mean(TAGP)'))

alt.layer(bar_cumm_TAGP, line).resolve_scale(
x = 'independent'
)

My code snippet:

condition = [0, 0.25, 0.5, 0.75, 1, 1.25, 1.5, 1.75, 2]
df_TAGP_2 = pd.DataFrame({
     'day': pd.date_range('1990-01-01', freq='D', periods=10000),
     'DVS': np.random.choice(condition , 10000,),
     'TAGP': np.random.randint(4000, size=(10000)),
     'simulation_date': '2020-10-15'})

My result:

What I would like to see

Many thanks!

Answer 1

It looks like you already have a dual x-axis in the picture. Some of the weirdness you are seeing with the different length of the top and bottom axis have been resolved in the current development version of Altair, in which you chart would look like this:

You can try it out by downloading it directly from the Altair GitHub page.

Update after question edit:

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

condition = [0, 0.25, 0.5, 0.75, 1, 1.25, 1.5, 1.75, 2]
df_TAGP_2 = pd.DataFrame({
     'day': pd.date_range('1990-01-01', freq='D', periods=20),
     'DVS': np.random.choice(condition , 20,),
     'TAGP': np.random.randint(10, size=(20)),
     'simulation_date': '2020-10-15'})

bar_cumm_TAGP = alt.Chart(df_TAGP_2[df_TAGP_2['simulation_date'] == '2020-10-15']).mark_bar().encode(
        alt.X('DVS:O'),
        alt.Y(f'TAGP:Q', title=""),
        alt.Color(f'TAGP:Q', scale=alt.Scale(scheme='greens'))).properties(width=850, height=350)

line = alt.Chart(df_TAGP_2).mark_line(interpolate='basis', strokeWidth=2, color='#FB0000').encode(
alt.X('day:O', axis=alt.Axis(labelPadding=100)),
alt.Y('mean(TAGP)'))

alt.layer(bar_cumm_TAGP, line).resolve_scale(
x = 'independent'
)