Stuck on an issue?

Lightrun Answers was designed to reduce the constant googling that comes with debugging 3rd party libraries. It collects links to all the places you might be looking at while hunting down a tough bug.

And, if you’re still stuck at the end, we’re happy to hop on a call to see how we can help out.

Chart sizing through custom theme doesn't work

See original GitHub issue

Hello Altair community, I am trying to set the width and height of chart through theme setting, however it doesn’t seem to be working. Am I setting the wrong keys or is this a bug? Please see below for detail:

# custom theme
def long_chart():
    return {
        'view': {
            'width'  : 500,
            'height' : 150      
        }
    }
alt.themes.register('long_chart', long_chart)

with alt.themes.enable('long_chart'):
    chart1 = alt.Chart(population).mark_point().encode(
        x='year:O',
        y='sum(people)',
        color='sex:N'
    ).properties(
        title='Chart sizing thru custom theme'
    )
    
chart2 = alt.Chart(
    population,
    width=500,
    height=150).mark_point().encode(
    x='year:O',
    y='sum(people)',
    color='sex:N'
    ).properties(
    title='Chart sizing thru Chart()'
    )

chart1 | chart2

visualization I was expecting chart1 to look exactly the same as chart2. Looking at the json spec of chart1, it seems that the width and height is still the default value: chart1.to_dict()

{'config': {'view': {'width': 400, 'height': 300}},
 'data': {'name': 'data-45e35b5faa42c7034b347e8bcfc8b805'},
 'mark': 'point',
 'encoding': {'color': {'type': 'nominal', 'field': 'sex'},
  'x': {'type': 'ordinal', 'field': 'year'},
  'y': {'type': 'quantitative', 'aggregate': 'sum', 'field': 'people'}},
 'title': 'Chart sizing thru custom theme',
 '$schema': 'https://vega.github.io/schema/vega-lite/v2.6.0.json',
 'datasets': {'data-45e35b5faa42c7034b347e8bcfc8b805': [{'age': 0,
    'people': 1483789,
    'sex': 1,
    'year': 1850},
   {'age': 0, 'people': 1450376, 'sex': 2, 'year': 1850},
...

Issue Analytics

State:
Created 5 years ago
Comments:8 (4 by maintainers)

Top GitHub Comments

1reaction

jakevdpcommented, Nov 19, 2018

And note that themes, and configuration settings in general, can only be applied to top-level charts. There is no way in Vega-Lite to set different global configuration options for different sub-charts.

1reaction

FlorianGDcommented, Nov 19, 2018

When exiting the context manager, the settings are reset, even for saved charts. I do not know if it is normal, but you can see it this way

import altair as alt
from vega_datasets import data
from pprint import pprint

population = data.population.url

def long_chart():
    return {
        'config': {
            'view': {
                'height': 150,
                'width': 500,
            },
        }
    }

alt.themes.register('long_chart', long_chart)
#> <function __main__.long_chart()>

with alt.themes.enable('long_chart'):
    chart1 = alt.Chart(population).mark_point().encode(
        x='year:O',
        y='sum(people):Q',
        color='sex:N'
    ).properties(
        title='Chart sizing thru custom theme'
    )
    print('Inside of the context manager')
    pprint(chart1.to_dict())
#> Inside of the context manager
#> {'$schema': 'https://vega.github.io/schema/vega-lite/v2.6.0.json',
#>  'config': {'view': {'height': 150, 'width': 500}},
#>  'data': {'url': 'https://vega.github.io/vega-datasets/data/population.json'},
#>  'encoding': {'color': {'field': 'sex', 'type': 'nominal'},
#>               'x': {'field': 'year', 'type': 'ordinal'},
#>               'y': {'aggregate': 'sum',
#>                     'field': 'people',
#>                     'type': 'quantitative'}},
#>  'mark': 'point',
#>  'title': 'Chart sizing thru custom theme'}

print('Oustide of the context manager')
#> Oustide of the context manager
pprint(chart1.to_dict())
#> {'$schema': 'https://vega.github.io/schema/vega-lite/v2.6.0.json',
#>  'config': {'view': {'height': 300, 'width': 400}},
#>  'data': {'url': 'https://vega.github.io/vega-datasets/data/population.json'},
#>  'encoding': {'color': {'field': 'sex', 'type': 'nominal'},
#>               'x': {'field': 'year', 'type': 'ordinal'},
#>               'y': {'aggregate': 'sum',
#>                     'field': 'people',
#>                     'type': 'quantitative'}},
#>  'mark': 'point',
#>  'title': 'Chart sizing thru custom theme'}

^{Created on 2018-11-19 by the reprexpy package}