Replies: 4 comments 1 reply
-
|
I think a new field under .spec.filter would be best for this |
Beta Was this translation helpful? Give feedback.
-
|
Yeah, the current logic definitely treats that list as a strict AND condition. I believe it follows the additive pattern common in K8s-style selectors, even though the syntax here is a simple list rather than a map. I wouldn't change the current I’d suggest opening a feature request for this. Most PR-based workflows eventually hit this wall where they need "any" rather than "all" labels to trigger a specific build or deployment. lmk if that makes sense. |
Beta Was this translation helpful? Give feedback.
-
|
The current semantics comes from GitLabMergeRequest. The GitLab API call receives the list of labels and implements the filtering on the server side. GitHub does not support the same, so we implemented the filtering on the client (operator) side. Given that, I'm not against adding a new field for more advanced filtering. But since it's a list of strings and not a map, |
Beta Was this translation helpful? Give feedback.
-
|
Have hit the same issue with Github PR's. Would love to get a |
Beta Was this translation helpful? Give feedback.
Uh oh!
There was an error while loading. Please reload this page.
-
Consider the following ResourceSetInputProvider:
The intention was to trigger a ResourceSet (for a build workflow) if either label was present.
Instead, both labels are now required.
I think the docs aren't clear on this but I prefer to ask:
What would people think of changing the behavior, or expanding the filter spec? For example,
anyLabel: [a, b].Beta Was this translation helpful? Give feedback.
All reactions