count() inside aggregate/joinaggregate transform returns 0 instead of the length of the data

[joelostblom]

Take this code as an example:

import altair as alt
import pandas as pd

source = pd.DataFrame({
    'a': ['A', 'B', 'C', 'D', 'E', 'F', 'G', 'H', 'I'],
    'b': [28, 55, 43, 91, 81, 53, 19, 87, 52]
})

alt.Chart(source).mark_bar().encode(
    x='a',
    y='total:Q'
).transform_joinaggregate(
    total='count()',
    # groupby=['a']
).transformed_data()

The result with vegafusion enabled is 0 for the total column:

Whereas it is 9 when vegafusion is not enabled (Open the Chart in the Vega Editor):

If I uncomment the groupby above, the result is 1 in both cases:

[jonmmease]

Thanks for the report Joel, and for diagnosing that the issue arises with the lack of a groupby.