joewitt commented on PR #8094:
URL: https://github.com/apache/nifi/pull/8094#issuecomment-1866733000

   Ahh that discussion context is helpful - would make sense to have that 
already linked in either the JIRA itself or this PR.  If it was there and I 
missed it sorry about that.  Otherwise, we need to put links to such 
good/useful discussions on these to help others who might review later.
   
   To the issue at hand, I do not share the view we should create a different 
event type here.  The meaning of 'SEND' is that NiFi is responsible for having 
taken the bits of a thing and sent it to some external system.  That is 
precisely what is happening here.  Now there is value in knowing as well then 
that NiFi itself never actually stored the bits.  This is known though already 
from the provenance trail of this object.  Presumably it never had a 
CREATE/RECEIVE and the bytes will of course have never been populated other 
than perhaps a link/reference.
   
   Given there was already a decent discussion and some apparent consensus and 
clearly this was a lot of work I am not saying I am a -1 on this but it does 
add more code/more maintenance/more semantics and I'm not clear what the value 
add will be in that sense.  If the issue of size is being worked around here 
then ideally we'd have a different/simpler answer.  
   
   If we continue with this as-is the description of a SEND should ensure devs 
use that when sending content from within NiFi to an external system and an 
upload only when sending external content to an external system.  Because as 
written in teh docs/guide right now UPLOAD looks like a subcase of SEND which 
it is not.
   
   


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]

Reply via email to