improve feedback to user with "." in column

[Sprocketer]

What happened?

Had a pandas dataframe with columns with "." (without quotes) in name (was using a string of a floating point number as part of a key name). Tried to plot, nothing shows up. 2 hours of debugging later realised that if you truncate the numbers to before the "." then it works fine.

What would you like to happen instead?

Columns should properly plot

P.S. - not 100% if this is an Altair-specific problem or a Vega-lite problem - I have no clue how to replicate 100% accurately there. This should be easily reproducible though anyways.

Which version of Altair are you using?

5.3.0

[mattijn]

Thanks for raising. You will have to escape special characters in Altair or replace them, see https://altair-viz.github.io/user_guide/encodings/index.html#escaping-special-characters-in-column-names

[Sprocketer]

Thanks for raising. You will have to escape special characters in Altair or replace them, see https://altair-viz.github.io/user_guide/encodings/index.html#escaping-special-characters-in-column-names

@mattijn Alright sure - but this should 100% have its own error message. It would be as simple as just checking the column names for special characters and then raising an error if there are any. Would save tons of time for I imagine lots of people who might run in to this

[Sprocketer]

@mattijn If you don't mind it'd be good to reopen or make a new issue for making an error message for this - would be super useful
Also, you could mention in the docs how you can use U+2024 (․) instead of U+002E (.) to bypass this for full stops / periods

[mattijn]

Always open for ideas to improve the user experience, see also #2824

[Sprocketer]

Also just making clear the error message should be for all special characters not just "." - thank you for the help!

[mattijn]

If you have a clear approach in mind to improve this, please feel free to contribute.
Keep in mind that these special characters are used to access nested attributes within the data. For example:

import pandas as pd
import altair as alt

df = pd.DataFrame(data=[
    {"type": "Feature", "properties": {"location": "left"}, "geometry": {"type": "Polygon", "coordinates": [[[1.45, 3.75], [1.45, 0], [0, 0], [1.45, 3.75]]]}},
    {"type": "Feature", "properties": {"location": "middle-left"}, "geometry": {"type": "Polygon", "coordinates": [[[1.45, 0], [1.45, 3.75], [2.57, 3.75], [2.57, 0], [2.33, 0], [1.45, 0]]]}},
    {"type": "Feature", "properties": {"location": "middle-right"}, "geometry": {"type": "Polygon", "coordinates": [[[2.33, 0], [2.33, 2.5], [3.47, 2.5], [3.47, 0], [3.2, 0], [2.57, 0], [2.33, 0]]]}},
    {"type": "Feature", "properties": {"location": "right"}, "geometry": {"type": "Polygon", "coordinates": [[[3.2, 0], [3.2, 1.25], [4.32, 1.25], [4.32, 0], [3.47, 0], [3.2, 0]]]}}
])
df

alt.Chart(df).mark_geoshape().encode(
    alt.Color("properties.location:O").scale(scheme='magma')
).project(type="identity", reflectY=True)