I have a curiosity that has been on my mind for a while. I know that a pipeline has the ability to detect and expand documents (Email, Containers, etc.) I know that when a Pipeline does this discovery and expansion, it then sends the findings to the top of the pipeline to be processed, (in the event that expansion is required again...)
Here is where my curiosity comes into play: I have several pipelines, each dedicated to their own datatype (for complexity reasons.) I have also noticed that many of these pipelines start out with the same sequence of steps. (Filtering of non-content documents, Mime type discovery, etc.) Given that a workflow can chain pipelines, does the expansion step send newly found expanded documents to the top of the Workflow, (or just to the top of a Pipeline?)
If an expanded document is sent to the top of a workflow, I would have the ability to link all of these shared stages into a precursor pipeline then provide the custom processing in a datatype dedicated pipeline.