Details
-
New Feature
-
Status: Open
-
Major
-
Resolution: Unresolved
-
None
-
None
-
None
-
None
Description
After improving performance of the current SplitText (see NIFI-2851), I have noticed that it may still not be ideal for cases where waiting for all the splits to be computed is not required allowing splits to be released as soon as they are computed. While it may not fit all the use cases it certainly fits some, so having a streaming variation of SplitText processor or introducing a flag in the existing one would be a nice improvement.