Julenmendieta/MILAB-5143_supportMultipleChainsInRedefine

[julenmendieta]

No description provided.

[gemini-code-assist]

Code Review

This pull request updates several @platforma-sdk and @milaboratories dependencies and introduces a new 'run' marker column to the result pframe in both bulk and single-cell VDJ import workflows to facilitate downstream block discovery. A critical issue was identified in both workflow files where the result.add method is missing its third argument (the data resource), which will likely lead to runtime errors during execution.

[gemini-code-assist]

The result.add method in the pFrameBuilder typically requires three arguments: the column key, the column specification, and the data resource. Currently, the third argument (data) is missing, which will likely cause a runtime error during workflow execution. Since this is a marker column with valueType: "String", you should provide a string value, such as blockId.

    result.add("run", runTrace.inject({
        kind: "PColumn",
        name: "pl7.app/vdj/run",
        axesSpec: [],
        domain: {
            "pl7.app/vdj/clonotypingRunId": blockId
        },
        valueType: "String",
        annotations: {
            "pl7.app/label": "VDJ Import Run"
        }
    }), blockId)

[PaulNewling]

I don't believe this is correct, https://github.com/milaboratory/platforma/blob/main/sdk/workflow-tengo/src/pframes/builder.lib.tengo#L17-L19 handles when nothing is passed in for data

[gemini-code-assist]

The result.add method is missing its third argument (the data resource). For a marker column of type String, you should pass a value (e.g., blockId) to ensure the column is correctly added to the PFrame and to avoid runtime errors during workflow execution.

	result.add("run", runTrace.inject({
		kind: "PColumn",
		name: "pl7.app/vdj/run",
		axesSpec: [],
		domain: {
			"pl7.app/vdj/clonotypingRunId": blockId
		},
		valueType: "String",
		annotations: {
			"pl7.app/label": "VDJ Import Run"
		}
	}), blockId)