Update Kepler.gl Version and Improvement

[UrbanGISer]

pixi run pip install keplergl==0.3.7

[UrbanGISer]

TEST WORKFLOW: https://hub.knime.com/s/xLU_rX_Qbq_Mxyz5

[koettert]

Did the used JS libraries change with this release. If so you also need to download all required JS libraries and add them here: https://github.com/spatial-data-lab/knime-geospatial-extension/tree/main/knime_extension/libs/kepler/ with the right version. In addition you need to update the replacements here in the node code.
All this is necessary to use the Kepler view also in a Data App are not allowed to download and use external JS libraries.

[Copilot]

Pull Request Overview

This PR updates the Kepler.gl version to 0.3.7 and enhances the ViewNodeKepler class to support multiple input datasets through a dynamic input group feature.

• Updates Kepler.gl dependency management from conda to pip installation
• Adds support for multiple input tables through a new input table group
• Refactors the execute method to handle primary and additional tables separately

Reviewed Changes

Copilot reviewed 2 out of 2 changed files in this pull request and generated 4 comments.

File	Description
knime_extension/src/nodes/visualize.py	Adds multi-table support with primary and additional input groups, refactors processing logic into separate methods
knime_extension/pixi.toml	Removes conda keplergl dependency and adds pip post-install task for version 0.3.7

---

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

[Copilot]

Remove the trailing space at the end of the comment line.

Suggested change

	# pixi run pip install keplergl==0.3.7
	# pixi run pip install keplergl==0.3.7

[Copilot]

Remove the trailing space after the period in the description string.

Suggested change

	"Select the primary geometry column to visualize. ",
	"Select the primary geometry column to visualize.",

[Copilot]

The geopandas import is duplicated - it's already imported in the helper methods _process_table and _process_additional_table. Consider moving all imports to the top of the method or removing the duplicate imports in the helper methods.

Suggested change

import geopandas as gp

[Copilot]

The _process_additional_table method uses self.attribute_cols which is configured based on the primary table schema, but it's being applied to additional tables that may have different schemas. This could cause issues if the additional tables don't have the same column structure as the primary table.

Suggested change

	# include only selected attribute columns that are not geospatial and the selected geospatial column
	attribute_columns = self.attribute_cols.apply(input_table.schema)
	included_column_names = list()
	for c in attribute_columns:
	if not knut.is_geo(c):
	included_column_names.append(c.name)
	# Dynamically include all non-geospatial columns and the geometry column from the current table
	included_column_names = [c.name for c in input_table.schema if not knut.is_geo(c)]