Proposal: Installing an additional executable for plugin's completion script

[predatorray]

Hi there, I am the maintainer of a krew plugin. Kubectl recently released a new feature to support shell completion for plugins in its latest alpha release v1.26.0-alpha.3. It requires an additional executable named kubectl_complete-<pluginName> present on $PATH.

As discussed in the pull request, krew does not provide a way to deliver this separate executable because bin: in the manifest yaml only expects one parameter. As a result, although I have implemented the completion script and managed to publish it to a homebrew tap simply by adding another bin.install (see: predatorray/homebrew-brew#5), I am still not able to release this feature to the users using krew.

I am wondering do we have any plan to support this feature, for example, by supporting multiple parameters under bin: or providing another parameter called completion?

[ahmetb]

Yeah we should add support. I wasn't aware kubectl implemented support for it, as it was mostly in design discussion phase. It may take a while for us to get this done, so please be patient. Design proposals for this are appreciated.

[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.

This bot triages un-triaged issues 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 as fresh with /remove-lifecycle rotten
• Close this issue with /close
• Offer to help out with Issue Triage

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

/lifecycle rotten

[ahmetb]

/lifecycle frozen

[chisaato]

Here are some workaound for zsh users before this complete.

#810 (comment)

[fredgate]

krew knows the name of the plugin. Couldn't it automatically copy the kubectl_complete-pluginname file into the PATH if there is one in the plugin package ?

[marckhouzam]

Until this is supported by krew and that plugins make use of it, people can use the following little project to generate the required completion script for about 50% of plugins installed by krew: https://github.com/marckhouzam/kubectl-plugin_completion