davsclaus opened a new issue, #3250:
URL: https://github.com/apache/camel-kamelets/issues/3250

   Follow-up from #328 (part B of the proposal there).
   
   Every Kamelet should declare what it takes and gives in `spec.dataTypes`: 
the `in` side for a sink or an action, the `out` side for a source. Nothing 
checks this today, so a new or changed Kamelet can leave it out.
   
   ## Where the catalog is now (2026-11-19)
   
   Of the 230 Kamelets:
   
   | `spec.dataTypes` | Kamelets |
   |---|---|
   | `types` (with a media type) | 19 |
   | only `headers` | 142 |
   | none | 69 |
   
   The documentation rewrite (#3140 and the follow-up PRs) added the `headers` 
blocks. #3117 is adding a `default` data type to every Kamelet.
   
   ## Proposal
   
   Once #3117 is done, `CatalogValidator` (`library/kamelets-maven-plugin`) 
fails the build for a Kamelet that has no `dataTypes` for its own side:
   
   - source: `dataTypes.out`
   - sink: `dataTypes.in`
   - action: `dataTypes.in` and `dataTypes.out`
   
   A Kamelet that really has nothing to declare can say so with an empty 
`headers: {}`, so leaving it out is a decision and not an accident.
   
   If the rule lands before #3117 is finished, it can start as a warning, with 
the 69 Kamelets above as the list to work down.
   
   Not carried over from #328: example payloads (its point 3). The `Input:` and 
`Output:` paragraphs of each description now say what the body is, and that is 
checked by running the Kamelet.
   
   _Claude Code on behalf of Claus Ibsen_
   


-- 
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