feat: process generated CRDs before they are written out #6828
Conversation
metacosm
requested review from
andreaTP,
manusa,
oscerd and
rohanKanojia
as code owners
metacosm
force-pushed
the
crd-post-processor
branch
from
43942e9 to
72272b7
Compare
metacosm
force-pushed
the
crd-post-processor
branch
from
72272b7 to
55b935e
Compare
There was a problem hiding this comment.
Looks like a great improvement!
Do you mind adding a test that shows how to use the feature on the user side?
It should be relatively easy here.
Also updating the docs to reflect the changes (not a blocker can be done later).
Side note that there should be something not aligned with the spotless configuration ...
metacosm
changed the title
metacosm
force-pushed
the
crd-post-processor
branch
3 times, most recently
from
6d3266a to
8c20f6a
Compare
| } | ||
|
|
||
| /** | ||
| * @deprecated use {@link #emitCrd(HasMetadata, Set, CRDGenerationInfo, CRDPostProcessor)} instead |
There was a problem hiding this comment.
The new method is marked protected, users won't be able to transition.
On the other hand, why don't we make CRDPostProcessor a member variable (just as output, minQuotes, etc.) that defaults to the nullProcessor -> withCrdPostProcessor().
This way the method signature won't keep growing or changing.
There was a problem hiding this comment.
Actually, that emitCRD method should probably not have been exposed in the first place and I don't think anyone is using it but the suggested change is a good one. Thank you!
metacosm
force-pushed
the
crd-post-processor
branch
from
8c20f6a to
a529875
Compare
|
Description
Fixes #6827
Type of change
test, version modification, documentation, etc.)
Checklist