I have defined a webhook with an outbound application that I am using as a writeback webhook in an action type.
Each time I want to upgrade / change the webhook version, all parameters of that webhook are added to the action type again and I have to manually remove them and map the webhook inputs to the original parameters again.
This also happens if the webhook parameters have not changed.
The expected behavior here would be that if the input id of the webhook has not changed compared to the previously selected version, no new parameter is created, and the originally mapped parameter is applied instead. If the new webhook version has an additional parameter, the automatic creation of that parameter would be appreciated.
Since I am dealing with a lot of webhook-based action types, this would greatly improve the usability. Alternatively, I would be happy to get outbound-application-based webhooks (with OAuth2 code grant) in typescript functions.
Best,
Florian
1 Like
+1 for outbound-application-based webhooks (with OAuth2 code grant) in typescript functions. 
1 Like
Giving more context for the upcoming roadmap planning to have OAuth2-based webhooks enabled in Typescript functions:
-
Give context on the use case / what you’re trying to achieve
Build a full-blown business application based on the ontology that integrates with external
systems like cloud providers, but also Rest-APIs that directly write into databases with user credentials to maintain the full audit-trail.
-
Why you can’t do it today / work-arounds
There is no way to do integration testing or a series of complicated calls that may have an ontology edit in the middle. I would need two button clicks (separate user transactions) to verify if the webhook output can be handled by the ontology logic, and if not send a request the rollback the changes in the remote system (with user credentials) or flag the previously sent message as “incomplete”. Also, there is no way to make writeback webhooks concurrently with other logic.
In addition, I tried a lot using function outputs (complex structs) to compute webhook inputs (POST-body) which did not work. Having both the function logic and the webhook logic in code would greatly simplify things and remove double maintenance.
-
Why will this make a huge difference for you or your business?
It would greatly improve the possibilities that one could use for integrating central existing systems with the platform and raise acceptance. Also, having full access and flexibility to determine which parts of the code must be concurrent will greatly help improving the application performance and therefore user acceptance.
1 Like