Closed askpt closed 7 months ago
All modified and coverable lines are covered by tests :white_check_mark:
Comparison is base (
d792b32
) 94.28% compared to head (03db18a
) 94.47%.
:umbrella: View full report in Codecov by Sentry.
:loudspeaker: Have feedback on the report? Share it here.
@askpt FlagMetadata
and EventMetadata
are shaped the same in the SDK:
https://openfeature.dev/specification/types#event-metadata https://openfeature.dev/specification/types#flag-metadata
I think for consistency, that means we should use the same type for these, which I guess would mean it should be a Dictionary<string, object>()
, unless there's some way we can bring them in line with your new implementation in a non-breaking way... maybe with extension methods?
@askpt
FlagMetadata
andEventMetadata
are shaped the same in the SDK:https://openfeature.dev/specification/types#event-metadata https://openfeature.dev/specification/types#flag-metadata
I think for consistency, that means we should use the same type for these, which I guess would mean it should be a
Dictionary<string, object>()
, unless there's some way we can bring them in line with your new implementation in a non-breaking way... maybe with extension methods?
@toddbaert I added a TODO to change the dictionary to use the BaseMetadata
type. This would be a breaking change. See https://github.com/open-feature/dotnet-sdk/pull/223/commits/5a2dcc9c3544c379f27f786dd9f4b6e2b66ab61f
I think the e2e tests are failing because of the change in the resolvers to include a new optional metadata param - I think the change in the signature is somehow causing an issue due to the circular dependency on flagd.
We've had an issue to remove this cycle for a while, as we have in other SDKs by adding the in-memory provider (which the spec requires anyway).
@askpt merged the recent changes to remove the flagd dep and now e2e looks good!
Flag Metadata
Related Issues
Fixes #192
Notes
null
and logging it as a warning. For that effort, I need a log source (#189).Follow-up Tasks