You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Issue #, if available: aws-controllers-k8s/community#994
Description of changes:
This pull request replaces all templates and variables with a defined nomenclature. The current code-generator uses service name terms interchangeably, which eventually lead to the linked issue. While PR #211 solved the logical issue of overriding the service model name, in some cases, it did not do a clear job of clarifying what names exist for any service and when they are applicable.
There are 3 different names that are used to programmatically reference a service. Here are the terms I am using in this PR to refer to them, and their given purpose:
- Service ID
- The `metadata.serviceID` field from the model `api-2.json` file
- Service model name
- The path to the model `api-2.json` file
- Service package name
- The AWS SDK Go package path
- Used by the ACK controller's for their name
Some examples on when these differ:
- Step Functions uses `sfn` for its ID and alias, but `states` for its model name
- Elastic Load Balancing v2 uses `elbv2` for its alias, but `elasticloadbalancingv2` for its ID and model name
There are also 2 names defined in the controller metadata file that are not used in any parts of the code, but are used in documentation:
- Service name
- The full name of the service as defined by the product's documentation page (eg. `Amazon Elastic Cloud Compute`)
- Service short name
- An informal, shortened name of the service (eg. `EC2`)
By submitting this pull request, I confirm that my contribution is made under the terms of the Apache 2.0 license.
0 commit comments