Add Contents > Create or update file contents operation to the OOTB GitHub Connector

[wollefitz]

Is your feature request related to a problem? Please describe.

If I want to create or update GitHub file contents from my C8 process, I have to manually put in all necessary data using the REST Connector to call https://docs.github.com/en/rest/repos/contents?apiVersion=2022-11-28#create-or-update-file-contents. This is quite error-prone since I have to know all necessary headers and define the request body correctly.

Additionally, the GitHub API requires the file content passed in the request body to be Base64 encoded which requires cumbersome workarounds if the content is not already encoded when receiving it from another source (e.g. Web Modeler's REST API, see https://modeler.cloud.camunda.io/swagger-ui/index.html#/Files/getFile).

Describe the solution you'd like

Using the GitHub connector, I would like to have the option to create or update file contents.
I would want to input as little data as possible and as much custom data as required.
The file content passed in the request payload should be automatically Base64 encoded.

Describe alternatives you've considered

Use the REST Connector and define my own template once I figured out all the necessary headers and how to define the body correctly.

Additional context

• API to run create or update file contents: https://docs.github.com/en/rest/repos/contents?apiVersion=2022-11-28#create-or-update-file-contents
• Result of https://github.com/camunda/web-modeler/issues/7842

[sbuettner]

@saig0 Can we consider to add basic encoding/converter funtions to the zeebe scala engine so user can make use of them independent from the component calling them (zeebe/connectors)

[wollefitz]

I had a chat with @sbuettner and @saig0 yesterday. In summary, there are different approaches on how to support base64 encoding of inputs (as laid out in this working document about Connectors data handling/transformation) but no decision yet on how to proceed. Until there is a concept, implementing this feature is blocked.

Additionally, there is no dedicated epic for implementing data handling (yet), it's still to be discussed if this is in scope of https://github.com/camunda/product-hub/issues/1748, see https://github.com/camunda/product-hub/issues/1748#issuecomment-1976542909.

[bastiankoerber]

we can support it, when we will work on our epic: handle large documents: https://github.com/camunda/product-hub/issues/1748

[wollefitz]

This is effectively unblocked by camunda/feel-scala#846 and can be used by versions of Zeebe that use the to be built next version (1.17.8/1.18.0) of feel-scala.