In addition to providing error feedback data flows contain features that enable you to debug and validate your json. Using the data flow validate API or stop on error flag are two such features system integrators will find invaluable.
The data flow validate API will validate the entire data flow json and return any errors found. This process can aid not only in developing data flows but become part of your integration process as well. A common workflow is to first validate the data flow and notify the calling system of any errors so that they may be corrected. Only when the corrections have been made in the calling system and the data flow is error-free, is it sent for processing.
The stop on error flag tells the system what to do in the event an error is encountered. A common use case for setting this flag to true would be for larger data flows. If you were sending a large number of envelopes and there is an error in the first one perhaps it is best to abort since there error could be in subsequent envelopes.
Let's note the differences between the two possible states:
Must explicity set to true
Aborts data flow on first error
Returns only one error
Ceases processing subsequent envelopes
Default setting
Continues error checking even after finding a failure point
Returns multiple errors
Attempts to process subsequent envelopes