Fix DuckDB documentation example to use SQL queries instead of pandas-like API

[Copilot]

Edit: Used Copilot for this since it's straightforward and tested:

---

The DuckDB example in the homepage documentation used pandas-like methods (groupby(), mean(), sort_values(), tail()) on a DuckDB relation object, which raises AttributeError: This relation does not contain a column by the name of 'groupby'.

Changes

• Replace duckdb.from_df() with explicit connection and table registration
• Use SQL query with aggregation instead of chaining pandas methods
• Call connection.query() instead of attempting pandas-like operations on relations

Before

df_duckdb = duckdb.from_df(df)
table = df_duckdb.groupby(['origin', 'mfr'])['mpg'].mean().sort_values().tail(5)
table.hvplot.barh('mfr', 'mpg', by='origin', stacked=True)

After

con = duckdb.connect(':memory:')
con.register('autompg', df)

query = """
SELECT origin, mfr, AVG(mpg) as mpg
FROM autompg
GROUP BY origin, mfr
ORDER BY mpg DESC
LIMIT 5
"""
table = con.query(query)
table.hvplot.barh('mfr', 'mpg', by='origin', stacked=True)

Original prompt

---

This section details on the original issue you should resolve

<issue_title>DuckDb reference example not working</issue_title>
<issue_description>Example from https://hvplot.holoviz.org/en/docs/latest/

import duckdb
import hvplot.duckdb
from bokeh.sampledata.autompg import autompg_clean as df

df_duckdb = duckdb.from_df(df)
table = df_duckdb.groupby(['origin', 'mfr'])['mpg'].mean().sort_values().tail(5)
table.hvplot.barh('mfr', 'mpg', by='origin', stacked=True)

Traceback (most recent call last):
  File "/home/jovyan/repos/private/panel/script.py", line 6, in <module>
    table = df_duckdb.groupby(['origin', 'mfr'])['mpg'].mean().sort_values().tail(5)
            ^^^^^^^^^^^^^^^^^
AttributeError: This relation does not contain a column by the name of 'groupby'

duckdb==1.2.2
hvplot==0.11.2</issue_description>

Comments on the Issue (you are @copilot in this section)

@MarcSkovMadsen This would work

import duckdb
import hvplot.duckdb
from bokeh.sampledata.autompg import autompg_clean as df

# Create a DuckDB connection and register the DataFrame as a table
con = duckdb.connect(':memory:')
con.register('autompg', df)

# Query using DuckDB SQL to get the same result
query = """
SELECT origin, mfr, mpg
FROM (
    SELECT origin, mfr, AVG(mpg) as mpg
    FROM autompg
    GROUP BY origin, mfr
    ORDER BY mpg DESC
    LIMIT 5
)
ORDER BY mpg ASC
"""

table = con.execute(query)
table.hvplot.barh('mfr', 'mpg', by='origin', stacked=True)

</comment_new>
<comment_new>@hoxbro
I would suggest using table = con.query(query) instead of con.execute(query) as this is supported by narwhals:

</comment_new>

• Fixes DuckDb reference example not working #1693

---

✨ Let Copilot coding agent set things up for you — coding agent works faster and does higher quality work when set up for your repo.

[Azaya89]

LGTM, although I think the gif needs to be replaced as well to match the actual output.