Add some format to the ListImages, such as templates or JSON

[afbjorklund]

Follow-up to #10933

Currently ListImages API returns the images in a fixed text format...

        // ListImages returns a list of images managed by this container runtime
        ListImages(ListImagesOptions) ([]string, error)

minikube image ls

k8s.gcr.io/pause:3.2
k8s.gcr.io/kube-scheduler:v1.20.2
k8s.gcr.io/kube-proxy:v1.20.2
k8s.gcr.io/kube-controller-manager:v1.20.2
k8s.gcr.io/kube-apiserver:v1.20.2
k8s.gcr.io/etcd:3.4.13-0
k8s.gcr.io/coredns:1.7.0
gcr.io/k8s-minikube/storage-provisioner:v5
docker.io/library/busybox:latest
docker.io/kubernetesui/metrics-scraper:v1.0.4
docker.io/kubernetesui/dashboard:v2.1.0

NOTE: That "busybox" image came from the cache:

{
    "cache": {
        "busybox": null
    }
}

As suggested, it could be useful with other formats: #11007 (review)

---

crictl offers three different formats:

table (default)

minikube ssh -- sudo crictl images --output table k8s.gcr.io/pause

IMAGE               TAG                 IMAGE ID            SIZE
k8s.gcr.io/pause    3.2                 80d28bedfe5de       683kB

json

minikube ssh -- sudo crictl images --output json k8s.gcr.io/pause
{
  "images": [
    {
      "id": "sha256:80d28bedfe5dec59da9ebf8e6260224ac9008ab5c11dbbe16ee3ba3e4439ac2c",
      "repoTags": [
        "k8s.gcr.io/pause:3.2"
      ],
      "repoDigests": [
        "k8s.gcr.io/pause@sha256:927d98197ec1141a368550822d18fa1c60bdae27b78b0c004f705f548c07814f"
      ],
      "size": "682696",
      "uid": null,
      "username": ""
    }
  ]
}

yaml

minikube ssh -- sudo crictl images --output yaml k8s.gcr.io/pause

images:
- id: sha256:80d28bedfe5dec59da9ebf8e6260224ac9008ab5c11dbbe16ee3ba3e4439ac2c
  repoDigests:
  - k8s.gcr.io/pause@sha256:927d98197ec1141a368550822d18fa1c60bdae27b78b0c004f705f548c07814f
  repoTags:
  - k8s.gcr.io/pause:3.2
  size: "682696"
  uid: null
  username: ""

[tharun208]

Can we pick this once #11406 is done?

[afbjorklund]

You don't have to wait for the default to change.

[afbjorklund]

For this issue one adds something to the ListImagesOptions, and then implements it for all the different CRI in their ListImages.

// ListImagesOptions are the options to use for listing images
type ListImagesOptions struct {
    // Format is the output format
    Format string
}

       // ListImages returns a list of images managed by this container runtime
        ListImages(ListImagesOptions) ([]string, error)

Finally add it to the cmd.

[tharun208]

@afbjorklund I like to work on this

[tharun208]

/assign

[tharun208]

But, in docker, there is no option to show output as yaml or json. what to do in that case?

[afbjorklund]

But, in docker, there is no option to show output as yaml or json. what to do in that case?

It is a bit hidden: docker images --format '{{json .}}'

You should probably pick one format internally (json), and then maybe add yaml to the output...

There is no need to use different output for the internal implementation, so just stick with JSON.

The idea was to have one json document per string (line).

---

Currently ListImages outputs the image id.

k8s.gcr.io/pause:3.2

Instead, the output should be a JSON document:

{"id":"sha256:80d28bedfe5dec59da9ebf8e6260224ac9008ab5c11dbbe16ee3ba3e4439ac2c","repoTags":["k8s.gcr.io/pause:3.2"],"repoDigests":["k8s.gcr.io/pause@sha256:927d98197ec1141a368550822d18fa1c60bdae27b78b0c004f705f548c07814f"],"size":"682696","uid":null,"username":""}

Probably skip the "uid" and "username", no ?

Maybe convert the size from string to integer.

When this feature is available in the API, it can be parsed and converted to output in the CLI.

Note that you would have to convert the output from Docker, in order to match the one from CRI.

{"Containers":"N/A","CreatedAt":"2020-02-14 19:51:50 +0100 CET","CreatedSince":"15 months ago","Digest":"\u003cnone\u003e","ID":"80d28bedfe5d","Repository":"k8s.gcr.io/pause","SharedSize":"N/A","Size":"683kB","Tag":"3.2","UniqueSize":"N/A","VirtualSize":"682.7kB"}

[k8s-triage-robot]

The Kubernetes project currently lacks enough contributors to adequately respond to all issues and PRs.

This bot triages issues and PRs according to the following rules:

• After 90d of inactivity, lifecycle/stale is applied
• After 30d of inactivity since lifecycle/stale was applied, lifecycle/rotten is applied
• After 30d of inactivity since lifecycle/rotten was applied, the issue is closed

You can:

• Mark this issue or PR as fresh with /remove-lifecycle stale
• Mark this issue or PR as rotten with /lifecycle rotten
• Close this issue or PR with /close
• Offer to help out with Issue Triage

Please send feedback to sig-contributor-experience at kubernetes/community.

/lifecycle stale

[k8s-triage-robot]

The Kubernetes project currently lacks enough active contributors to adequately respond to all issues and PRs.

This bot triages issues and PRs according to the following rules:

• After 90d of inactivity, lifecycle/stale is applied
• After 30d of inactivity since lifecycle/stale was applied, lifecycle/rotten is applied
• After 30d of inactivity since lifecycle/rotten was applied, the issue is closed

You can:

• Mark this issue or PR as fresh with /remove-lifecycle rotten
• Close this issue or PR with /close
• Offer to help out with Issue Triage

Please send feedback to sig-contributor-experience at kubernetes/community.

/lifecycle rotten

[medyagh]

@tharun208 are you still working on this ?

[presztak]

I started working on this.
/assign