My goal is to:
1. Input a .tar file, containing a CSV file
2. Tag it with a unique ID, based on the filename (or something else TBD)
3. Extract the tar file
4. Read Lines from the extracted CSV
5. Output to indexes, including the unique ID that was added in step 2
I've added a Tagging stage (step 2) to the Extraction pipeline (step 3), to tag the unique ID. But, when I test the workflow, the Read Lines stage does not seem to copy the tagged field into the new documents it generates. (If it matters, I have recursion on.)
Is it possible for a tag added to a document to be applied to the resulting documents, when the original passes through the Read Lines stage? Or should I find another way to accomplish this (perhaps using HCI_parentUri, which seems to contain the URL of the original .tar file)?