[ 
https://issues.apache.org/jira/browse/NIFI-13452?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17860792#comment-17860792
 ] 

Zsihovszki Krisztina commented on NIFI-13452:
---------------------------------------------

[~joewitt] 

The use case is the following: we're consuming messages, taking actions on the 
messages that have been consumed and sending a response using the PublishSlack 
processor. 
Then we need to collect the reactions for that specific response and support 
the possibility to accumulate the number of reactions on a given window of time 
for that specific response message. 
This way after duration X, we can conclude that this message got N reactions 
R1, M reactions R2 etc and take actions based on this. 
This will support feedback loop use cases where we can evaluate the quality of 
the response sent via PublishSlack.


You're suggesting to do that with ConsumeSlack but that would make this use 
case complicated. It would not be easy to filter results based on the specific 
messages we care about.
Even if we distinguish the messages we care about e.g. with a tag, there is no 
elegant way in NiFi to do some window aggregation of the flow files we would 
receive for the reactions. (We want a window of time for the reactions and 
accumulate the counters for each type of reaction). 
While this is certainly technically doable, that would make the flow design 
quite complicated. To serve the use case I'm working on having a dedicated 
processor seems to be the better approach.

> GetSlackReaction processor to fetch reactions
> ---------------------------------------------
>
>                 Key: NIFI-13452
>                 URL: https://issues.apache.org/jira/browse/NIFI-13452
>             Project: Apache NiFi
>          Issue Type: New Feature
>            Reporter: Zsihovszki Krisztina
>            Assignee: Zsihovszki Krisztina
>            Priority: Major
>
> The goal of this JIRA is to create a new processor so it'd possible to be 
> fetch the emojis added to a message in Slack.
>  



--
This message was sent by Atlassian Jira
(v8.20.10#820010)

Reply via email to