You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I'll try to clarify the purpose of this client and see if I answer your question.
This library currently doesn't have the Argo CLI ported to Python, if that's what you mean. I am sorry to disappoint you if that's what you have expected.
This client acts similarly to the kubernetes-python-client. It is basically an extension to the Kubernetes client providing Argo models and API paths to create, list, delete and in general manipulate workflows and workflow templates as you would do with any other Kubernetes resource.
@CermakM just wanted to let you know what you have done is very valuable to me. I didn't make much progress on this after I tried to generate python sdk from argo swag definition (with Kubernetes client generator). I were able to generate the models but I couldn't fix the dependency problems where some Argo models just assume kubernetes models were there. So would love to know how you have resolved that.
Other than that it's been great, I have been able to build dags on top of your sdk and on my way to work a workflow framework on top of your client. Thank you!
Hey @yxue-kabam. Thanks a lot for the feedback! :) I am glad it helps!
Yep, it's been somewhat complicated to include the kubernetes models. I solved it by tweaking the code generation and including the Python Kubernetes models.
Activity
CermakM commentedon Dec 9, 2019
Hey @zhujiangyou, thanks for the question!
I'll try to clarify the purpose of this client and see if I answer your question.
This library currently doesn't have the Argo CLI ported to Python, if that's what you mean. I am sorry to disappoint you if that's what you have expected.
This client acts similarly to the kubernetes-python-client. It is basically an extension to the Kubernetes client providing Argo models and API paths to create, list, delete and in general manipulate workflows and workflow templates as you would do with any other Kubernetes resource.
yxue-kabam commentedon Dec 9, 2019
@CermakM just wanted to let you know what you have done is very valuable to me. I didn't make much progress on this after I tried to generate python sdk from argo swag definition (with Kubernetes client generator). I were able to generate the models but I couldn't fix the dependency problems where some Argo models just assume kubernetes models were there. So would love to know how you have resolved that.
Other than that it's been great, I have been able to build dags on top of your sdk and on my way to work a workflow framework on top of your client. Thank you!
CermakM commentedon Dec 10, 2019
Hey @yxue-kabam. Thanks a lot for the feedback! :) I am glad it helps!
Yep, it's been somewhat complicated to include the kubernetes models. I solved it by tweaking the code generation and including the Python Kubernetes models.
CermakM commentedon Dec 18, 2019
The SDK is currently under development: https://github.com/CermakM/argo-python-sdk
[-]Programming implementation of Argo[/-][+]Argo Python SDK[/+]CermakM commentedon Jan 3, 2020
Here is a hello-world example of the Argo Python SDK.
Feel free to visit https://github.com/CermakM/argo-python-sdk for more examples.
CermakM commentedon Jan 30, 2020
If you want an API for Argo CLI directly, please check https://github.com/argoproj-labs/argo-workflows-python-sdk