Simpler derivation of singular & plural method names #366
Conversation
cben
force-pushed
the
refactor-singular-plural
branch
from
18e10d5 to
e728976
Compare
|
There was also a workaround for |
cben
force-pushed
the
refactor-singular-plural
branch
from
d38ca33 to
0b48801
Compare
|
All done now, PTAL. |
security.openshift.io_api_resource_list.json taken from openshift v3.10.0. (openshift also includes SecurityContextConstraints also in api/v1/, but I prefer to keep core_api_resource_list.json pure upstream k8s and separate openshift-specific test stuff.) kubernetes/kubernetes#8115
Followup to ManageIQ#360. Using discovery info captured after loading istio's CRD: https://github.com/istio/istio/blob/1.0.3/install/kubernetes/helm/istio/templates/crds.yaml
Similar to Endpoints we force this kind to singular method names. Unlike Endpoint, lacked a workaround to pass original kind when creating.
ManageIQ#367 Also fixes ManageIQ#307 - get_security_context_constraints.kind, get_endpoints.kind are now plural as in kubernetes. Also fixes ManageIQ#367 - create_security_context_constraint now works.
Workaround was also missing for SecurityContextConstraints, which probably means create_security_context_constraint was broken.
cben
force-pushed
the
refactor-singular-plural
branch
from
1e2319b to
8c481f2
Compare
|
@f4tq @eatwithforks @grosser @NickLaMuro @agrare looking for someone to review this... |
|
Ouch, I think I caused a regression here for plurals which are not a pure suffix. cc @masayag this might be a blocker for #373 — I'd prefer not to complicate it further before we fix this. |
|
opened issue #376 with exact details. will fix ASAP, but having a very busy week. |
Simpler logic, more explicit about intent (e.g. for the 2 special kinds hardcode the final result instead of intermediate step), added explanations.
Also fixes #307 -
get_security_context_constraints.kind,get_endpoints.kindare now plural as in kubernetes.Also fixes #367 -
create_security_context_constraintnow works. (tested live too)Added lots of tests.
@f4tq @eatwithforks please review (easier by commits).
After this, I'll rebase #355 and change it to support both new and old names...