Ed Overton

I took some time to dig around, and there are many places in the strategic merge patch flow where the schema is involved. Adjusting that code for this performance benefit...

The crash comes from infinite recursion in [filesys.PathSplit](https://github.com/kubernetes-sigs/kustomize/blob/kustomize%2Fv5.0.1/kyaml/filesys/util.go#L40). The call to filepath.Split [honors both / and \ as path separators on windows](https://cs.opensource.google/go/go/+/master:src/os/path_windows.go;l=12-16). However, filesys.PathSplit expects only os.PathSeparator (both [here](https://github.com/kubernetes-sigs/kustomize/blob/kustomize%2Fv5.0.1/kyaml/filesys/util.go#L45) and...

As part of this work, I had to modify [errTimeOut](https://github.com/kubernetes-sigs/kustomize/pull/5542/files#diff-e1eefb74233c63132ee45f5cca8a479ba9157e230a41762d74e1cea3b51fec3b) to enable it to work properly with `errors.As()`. The shift to use pointer receivers aligns errTimeOut with other errors that...

> Could you also add a test case on `kusttarget_test.go` to check on this specific use case? Good point. I'll take a look at the test support and see if...

Highlighting a few comments from what I wrote over in https://github.com/kubernetes-sigs/kustomize/issues/5084#issuecomment-1806477953: > ... do you know if there are reasons why caching Resource.CurId() could be problematic? I feel like there...

Test results from my runs show meaningful improvements where I expected, with identical kustomize output between the PR and v5.3.0. A testcase based on #2869: kustomize_test.go:289: Runtime variation (0.56x): branch-pr-5481...

I'd like this to be considered. /remove-lifecycle rotten

In poking around at this benchmark, I did notice some behavior that surprised me. I adjusted the value of genconfig[1].resources and the run time jump between 1 and 2 stands...

Closes #4940

> @ephesused could we close this PR in favor of #4944? Definitely - I prefer that approach. Thanks for the prompt.