Background:

This task is different from the data from SQL server. By looking backward, in the Mainland (old) data, the Seq ID, Begin Time, and End Time are there because the all these information are available in the db tables (SQL DB). In this case, it is straightforward by picking up these information which are already there and then populating these information w.r.t. each field.

As for the Tavis Amazon data, no such information is included in the manifest file which is supposed to contain all information based on my understanding. Any information which should be back imported from the derivative datastreams like FITs should be done by the workflow or need to develop some workflow-like tool to retrieve the image begin and end date information from the image sequence. A drush script, php or java code, or something else.

The key issue (hardest part as I think) is that the process need to handle a set of objects (images in this case) to find seq number, begin time and end time. Current workflow design by DGI ( still have bugs from the Quotient test on D7+I7 integration) or by Camel from Dan we discussed before only support the single object playing.