Can we remove a marking for a subset of outputs of a multi-output transform?

Ask the Community
1

Hi,

I would like to know if it is possible to remove a marking for a subset of outputs of a multi-output transform?

Assume that you have an input with a marking and a multi-output transform producing two datasets A and B using a Code Repository transformation. A will still contain sensitive data and hence should inherit the marking from the input. But B contains only open data and hence should not inherit the marking from the input. Standard procedure to remove the marking would add “stop_propagating=…” to the input. In this case this would remove the marking from both outputs A and B.

Any ideas on how to do that within one transformation?

Many thanks.

2

Marking removal is specified in the input constructor and not the output constructor so there currently isn’t a way to do that - you would need to break the transformation in two transformations; one with stop_propagating and one without.