Automatically use y as default 'text' kwarg if only two cols

[ahuang11]

If only two columns, use y (this would previously crash because of text = [c for c in data.columns if c not in (x, y)][0]):

import hvplot.pandas
import pandas as pd

df = pd.DataFrame({
    'lat': [-34.58, -15.78, -33.45, 4.60, 10.48],
    'lon': [-58.66, -47.91, -70.66, -74.08, -66.86],
})

df.hvplot.points(x='lon', y='lat') * \
df.hvplot.labels(x='lon', y='lat', hover=False, padding=0.2)

[maximlt]

I'm OK with this change. However the tests are broken.

[ahuang11]

Thanks I think it's because the column name has units. Going to experiment with LLM...

@copilot can you fix the test by dropping the correct column?

[ahuang11]

:'( okay maybe I don't have permissions to this repo to launch

[Copilot]

Pull Request Overview

This PR fixes a crash in the labels method when working with DataFrames that have only two columns by automatically using the y-axis column as the default text label.

• Fixed IndexError when no third column is available for text labels
• Added fallback logic to use y-column as text when only x and y columns exist
• Added test coverage for the two-column scenario

Reviewed Changes

Copilot reviewed 2 out of 2 changed files in this pull request and generated 1 comment.

File	Description
hvplot/converter.py	Added conditional logic to handle two-column DataFrames by using y as default text
hvplot/tests/testcharts.py	Added test case to verify labels work correctly with only two columns

---

_{Tip: Customize your code reviews with copilot-instructions.md. Create the file or learn how to get started.}

[maximlt]

Just to double check, is it ok for a holoviews element to have a dimension both in the kdims and vdims?

[Azaya89]

I think it can if you pass them in directly:

import holoviews as hv
import pandas as pd

df = pd.DataFrame({
    'lat': [-34.58, -15.78, -33.45, 4.60, 10.48],
    'lon': [-58.66, -47.91, -70.66, -74.08, -66.86],
})

plot = hv.Points(df, ['lon', 'lat'], 'lon')
plot.dimensions()

[Dimension('lon'), Dimension('lat'), Dimension('lon')]

This is what is now achieved here via next((c for c in data.columns if c not in (x, y)), y), no? @ahuang11

[ahuang11]

Almost! Very close:

next((c for c in data.columns if c not in (x, y)), y) should default to y, which is:

import holoviews as hv
import pandas as pd

df = pd.DataFrame({
    'lat': [-34.58, -15.78, -33.45, 4.60, 10.48],
    'lon': [-58.66, -47.91, -70.66, -74.08, -66.86],
})

plot = hv.Points(df, ['lon', 'lat'], 'lat')  # lat not lon
plot.dimensions()

If df has an extra col, it'll use that:

import holoviews as hv
import pandas as pd

df = pd.DataFrame({
    'lat': [-34.58, -15.78, -33.45, 4.60, 10.48],
    'lon': [-58.66, -47.91, -70.66, -74.08, -66.86],
    'col': ['a', 'b', 'c', 'd', 'e']
})

plot = hv.Points(df, ['lon', 'lat'], 'col')
plot.dimensions()

And yes to Maxime's question: