ExecProvider errors when parsing null values

[fillbit]

When executing a command like kubectl config use-context kubectl rewrite the the kubeconfig file and adds null to optional parameters. For example a a config with

users:
- name: Name1
  user:
    exec:
      apiVersion: client.authentication.k8s.io/v1alpha1
      command: heptio-authenticator-aws

is rewritten as:

users:
- name: Name1
  user:
    exec:
      apiVersion: client.authentication.k8s.io/v1alpha1
      args: null
      command: heptio-authenticator-aws
      env: null

When the ExecParser reads this it returns the error:

ERROR: Invalid kube-config file. Expected key args in kube-config/users[name=Name1]/user/exec

This PR adds a test case and a fix for this issue.

[k8s-ci-robot]

Thanks for your pull request. Before we can look at your pull request, you'll need to sign a Contributor License Agreement (CLA).

📝 Please follow instructions at https://git.k8s.io/community/CLA.md#the-contributor-license-agreement to sign the CLA.

It may take a couple minutes for the CLA signature to be fully registered; after that, please reply here with a new comment and we'll verify. Thanks.

---

• If you've already signed a CLA, it's possible we don't have your GitHub username or you're using a different email address. Check your existing CLA data and verify that your email is set on your git commits.
• If you signed the CLA as a corporation, please sign in with your organization's credentials at https://identity.linuxfoundation.org/projects/cncf to be authorized.
• If you have done the above and are still having issues with the CLA being reported as unsigned, please email the CNCF helpdesk: helpdesk@rt.linuxfoundation.org

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository. I understand the commands that are listed here.

[codecov-io]

Codecov Report

Merging #91 into master will increase coverage by <.01%.
The diff coverage is 100%.

@@            Coverage Diff             @@
##           master      #91      +/-   ##
==========================================
+ Coverage   91.71%   91.72%   +<.01%     
==========================================
  Files          13       13              
  Lines        1135     1136       +1     
==========================================
+ Hits         1041     1042       +1     
  Misses         94       94

Impacted Files	Coverage Δ
config/exec_provider_test.py	98.38% <100%> (+0.02%)	⬆️
config/exec_provider.py	82.6% <100%> (ø)	⬆️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 2d514ce...3682e9b. Read the comment docs.

[fillbit]

signed CLA?

[roycaihw]

Thanks for the contribution! I left some comments.

[roycaihw]

I suppose the test fails when exec_config is of ConfigNode class, e.g. having

exec_config = ConfigNode('test', self.input_ok)

[roycaihw]

The exception comes from ConfigNode which implements __getitem__ under the assumption that

all access keys are present in a well-formed kube-config.

Calling the __getitem__ in the if statement would also raise the exception.

To tolerate malformed config, I'd suggest using exec_config.safe_get(key) to verify if the value is null. Note that the unit test used dict for exec_config which didn't reflect the safe_get attribute.

[fillbit]

I've updated the PR to use safe_get and the unit tests to use ConfigNode instead of dict() so they work correctly. I also updated the input_ok to include a env: null value and removed my test case as it is now redundant

[roycaihw]

@fillbit Thanks. Could you also add a comment in ExecProvider documenting that __init__ expects exec_config to be of ConfigNode class, which implements the safe_get(self, key) method? Otherwise LGTM

[roycaihw]

Please fix the pycodestyle error for travis CI to pass, and squash commits before submit :)

[fillbit]

@roycaihw @mbohlool Are there any other issue that need to be addressed?

[roycaihw]

/lgtm
/approve

[k8s-ci-robot]

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: fillbit, roycaihw

The full list of commands accepted by this bot can be found here.

The pull request process is described here

Needs approval from an approver in each of these files:

• OWNERS [roycaihw]

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment

[nakulpathak3]

kubernetes-client/python#678 hi I've been seeing the error mentioned in the issue while using both 8.0.0a and 8.0.0. If anyone has any tips/ideas for what I'm doing wrong or if there is something we need to change with the way we are doing exec-provider, please let me know