pyawe

Python client library for the AWE (Advanced Workflow Engine) API.

Installation

pip install pyawe

Requires Python 3.9+.

Quick start

from pyawe import AweClient

client = AweClient(
    api_url="http://localhost:8080",
    auth_url="http://localhost:8081",  # omit if auth is behind the same proxy
)
client.login(email="you@example.com", password="secret")

# List workflows
workflows = client.workflows.list()

# Create a job and a workflow inside it
job = client.jobs.create("Q3 Onboarding")
wf  = client.workflows.create("Onboarding Steps", job_id=job.id)

# Add tasks
start = client.tasks.create("Kick-off",    workflow_id=wf.id, is_start=True)
review = client.tasks.create("Doc Review", workflow_id=wf.id)
end   = client.tasks.create("Complete",    workflow_id=wf.id, is_end=True)

# Link them
client.task_links.create(start.id, review.id)
client.task_links.create(review.id, end.id)

# Advance a task
client.tasks.update(start.id, status="In Progress")

Authentication

pyawe authenticates against the ullav-user-management service, which may be the same host as the AWE API or a separate one.

# Same host
client = AweClient("http://awe.example.com")

# Separate auth service
client = AweClient(
    api_url="http://awe.example.com",
    auth_url="http://auth.example.com",
)

Calling login() stores the JWT in the session. All subsequent calls attach it automatically as Authorization: Bearer <token>. Tokens expire according to the server configuration — call login() again to refresh.

Error handling

All exceptions inherit from AweError:

from pyawe import AweClient, AweNotFoundError, AweAuthError, AweError

try:
    wf = client.workflows.get("non-existent-id")
except AweNotFoundError:
    print("Workflow not found")
except AweAuthError:
    print("Not authenticated or forbidden")
except AweError as e:
    print(f"Unexpected error: {e}")

See the Exceptions reference for the full hierarchy.

Sub-clients

AweClient exposes one sub-client per resource family:

Attribute	Resource
client.workflows	Workflows — create, update, duplicate, merge
client.tasks	Tasks — CRUD, decide, rework
client.task_links	Links between tasks; data bindings
client.task_ports	Port specs and runtime values
client.task_scripts	Automated task scripts
client.task_runs	Execution run history
client.task_team_roles	Team-role assignments on tasks
client.jobs	Jobs — group workflows together
client.loop_blocks	Loop block tasks
client.execution_profiles	Kubernetes execution environments
client.notes	Notes on tasks, workflows, and jobs
client.note_folders	Folders for organising notes