I spoke with Support and development @devinweber You were right about requiring a loop to maintain the records if loading data in a script, after starting the script. You must mark the data record within the script when you are finished with each record.
So.
Start (script)
Loop
Load Data
Look for first unprocessed record- skipping those already processed
run logic - Do something fun and exciting
mark record as done - continue with loop until you have read all the data
At the last record exit the loop
End Loop
End (script)
Things to be aware of:
- If you do not loop correctly you will get a Nintex popup at end of script or end of reading one record. Move the loop around to get rid of those messages.
- if you are using a "Go to the Very End" action in the Error section you will have to handle that differently.
- If you are using "Go to the Very End" in a task that will also need to be handled differently, I have to remove my subtask and do the task at the same level as my loop.
- If you have many, many subtasks this could change the programming you need to do quite a bit.
So until NIntex has a fix for autoloading data and the data being stuck at First Record, I may opt-out of RPA Central for scheduled jobs and choose .bat files.
If anyone else has something clever they've done please let me know.