Patrick Ohly

In JSON, an additional tag instead of a text prefix could be used. But that still means that a new feature in Go is needed.

It is enough to treat the test as failed. But what should we then use as failure message? Currently it is the entire output of the test. That's too much...

What you see in https://prow.k8s.io/view/gs/kubernetes-jenkins/pr-logs/pull/116354/pull-kubernetes-integration/1633336232408780800 is what we currently use as failure messages. Compare that against E2E failures, e.g. https://prow.k8s.io/view/gs/kubernetes-jenkins/logs/ci-kubernetes-e2e-gce-cos-k8sstable3-default/1634073499108642816

/remove-lifecycle rotten I still think that this could be enhanced. I just ran into failure messages again where the actual failure got pruned and/or was hard to find in a...

/remove-lifecycle rotten I am working on a `ktesting.TContext` which wraps the underlying testing.T or testing.B. It also works as wrapper around Ginkgo, so integration and E2E tests can use the...

My stand-alone tests were done in Go playground. The actual test above lives in a [PR for Kubernetes master](https://github.com/kubernetes/kubernetes/pull/113538), which currently uses go-cmp v0.5.9.

The workaround for this is to split strings into []string at line breaks. I'm not doing that in the PR because I like the output less, but it has the...

I tried to come up with a simpler reproducer, but so far without luck. I'll keep trying...

It's the combination of embedding inside a struct and using a transformer which triggers it. Here's a reproducer: ```golang package main import ( "fmt" "strings" "github.com/google/go-cmp/cmp" "github.com/google/go-cmp/cmp/cmpopts" "github.com/onsi/ginkgo/v2/reporters" ) func...

cmp is at 0.5.9, the latest.