Dumbbell Plot fails on an empty value in the compared column

[kz930]

What happened?

Dumbbell Plot fails when the compared column is empty in any row. DumbbellPlotOpDesc.scala:116 takes the column's distinct values and sorts them:

entityNames = list(table[comparedColumnName].unique())
entityNames = sorted(entityNames, reverse=True)

A blank cell reaches the operator as None, so unique() returns ['alpha', None, 'gamma'] and sorted compares None against a string.

An empty value is ordinary input here. A blank CSV cell arrives as null, since univocity returns null for an empty field and AttributeTypeUtils.parseField passes it through by design, its first line being if (field == null) return null.

The execution then pauses on that row, with the exception shown in the operator's console.

Expected: a row whose compared column is empty either passes through, or is skipped, or fails with a message naming the column.

How to reproduce?

Upload a CSV with a blank cell in the compared column:

id,entity,category,measure
1,alpha,before,10
2,,before,20
3,gamma,after,30
4,alpha,after,15

Build CSV File Scan to Dumbbell Plot with Compared Column Name = entity, Category Column Name = category, Measurement Column Name = measure, Dumbbell Start Value = before, Dumbbell End Value = after. Run it. The operator's console shows the TypeError.

Version/Branch

1.3.0-incubating-SNAPSHOT (main)

Relevant log output

                         └ ['alpha', None, 'gamma']

TypeError: '<' not supported between instances of 'NoneType' and 'str'
    udf-v1:process_table:23