-
Notifications
You must be signed in to change notification settings - Fork 8.3k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[discuss] Usage Collection: Should we use config-schema
instead of our schema
implementation?
#85810
Comments
Pinging @elastic/kibana-core (Team:Core) |
this seems to be the most impactful cons. Also
So I would say a more restrictive schema definition probably make sense here. |
Thanks for opening an issue for this! It does sound compelling but I am afraid this would restrict "growing" the schema api separately as needed. For example we might want to add a
I find this con the main reason to prevent us from switching. We do need this level of percision to understand the fields more and provide accurate es mappping. I dont see a replacement to having that if we switch.
I agree with @pgayvallet on this one. We do have a "translation" layer in the telemetry tool to convert the schema to the target json. However we'd have to keep some sort of feature parity with |
Thank you for your comments! I totally agree with evolving our current schema is a better approach to migrating to I'll leave this open for more comments until I start implementing #83704 (where I'll need to build a validation logic based on the schema). |
Closing this issue as #90273 already added the logic to convert the Usage Collection's schema into |
At the moment we are using our own implementation of the
schema
when declaring Usage Collectors. It helps us to define the structure of the data that we'll send to the Remote Telemetry Service in a very declarative way. We've also added some TS and tools that validate that the output types of the collector and the schema match to some degree.Now that we are considering running actual validations (at least during CI) of the final payload based on that schema, would it be a good idea to use
config-schema
for that instead?Pros:
Cons:
number
/long
/short
/float
, ...I'm creating this issue to track a discussion about it. A "no" is a perfectly valid answer to close this issue.
Related to #83704
The text was updated successfully, but these errors were encountered: