Update data functionality for text file outputs
Implement the ability to update data for continuously publishing views to text file outputs in same way that it is available for other data targets. Use cases for this include:
1) Ability to update data in the case of out of order events, late arriving data, or recalculated data.
Same situation for us. If a site loses connectivity to our collective and sends the buffered data later, that means manually creating a backfill view every time, which is very inefficient. I would ask that parquet and JSON files be included in the enhancement.
Brent Bregenzer commented
I am using a csv target with a customer. The Integrator is replacing another custom application that was generating csv files from a data source that is now being fed to PI. Due to some time gaps between the stoppage of the old application and the Integrator coming online, it would be nice to be able to generate files for times in the past. Also, sometimes disruptions cause data to be added later (Original request) or files can be lost/deleted by accident (Michael Tippet).
AdminJoy Wang (Product Manager, OSIsoft) commented
Thanks for the feedback! We are considering adding support for this in a future release. It'd be extremely helpful if you can let us know which text-file target you need this for, how you expect the file to be updated, and how the file will be consumed/read.
David Brand commented
This is critical functionality if one plans to use PI Integrator to populate a separate database with PI data on a regular basis
Michael Tippett commented
This functionality also allows the ability to recreate text files in the case that they are deleted, corrupted, or otherwise lost.