argo-workflows 3.5.1

Community Maintained Python client for Argo Workflows

Argo Workflows Python Client

Python client for Argo Workflows

Installation

pip install argo-workflows

Usage

If you're familiar with Kubernetes client, the Argo client follows the same behaviour.

from argo.workflows.client import V1alpha1Api
from argo.workflows.config import load_kube_config

load_kube_config()  # loads local configuration from ~/.kube/config

And to list Workflows from a namespace (say argo):

v1alpha1 = V1alpha1Api()

wfs = v1alpha1.list_namespaced_workflows(namespace="argo")

To submit a Workflow, one would simply load it from a YAML^* and submit it as such:

import requests
import yaml

namespace = "argo"

# hello-world example
resp = requests.get("https://raw.githubusercontent.com/argoproj/argo/master/examples/hello-world.yaml")
resp.raise_for_status()

manifest: dict = yaml.safe_load(resp.text)

# Submit the Workflow to the `argo` namespace
v1alpha1.create_namespaced_workflow(namespace, manifest)

When working on a higher level of abstraction, check out the following projects:

• Couler that provides a simplified and unified interface for constructing and managing workflows
• Argo Python DSL

Versioning

The client is generated from the Argo and Kubernetes OpenAPI specification.

We follow semantic versioning, the client starts at version 1.0 which matches Argo release 2.3.0 and continues onwards.

Current Version Mapping

Argo Python SDK: 3.5 Argo: 2.10.2 Kubernetes:

Key:

• ✓ Exactly the same features / API objects in both client-python and the Kubernetes version.
• + client-python has features or api objects that may not be present in the Kubernetes cluster, but everything they have in common will work.
• - The Kubernetes cluster has features the client-python library can't use (additional API objects, etc).

We try to be consistent with Argo as much as possible and hence we created special branches for user convenience --- for example argo/v2.4.0 is identical with Argo release v2.10.2.

These can be then installed directly from github:

pip install -e "git+https://github.com/argoproj-labs/argo-client-python.git@argo/v2.10.2#egg=argo-workflows"

Code generation

The generated SDK will correspond to the argo version specified in the ARGO_VERSION file.

If you wish to generate code yourself, you can do so by reproducing the build environment (image): make builder_image, then running make builder_make to generate the client.

Dependencies

Usually these will be kept up to date (or at least in sync in the branch named after the argo version that you are looking for)

• a compatible KUBERNETES_BRANCH according to Argo's Gopkg.toml
• a compatible version of kubernetes-client
• requirements.txt and Pipfile

For additional configuration check out the Makefile.