aws / aws-step-functions-data-science-sdk-python

Step Functions Data Science SDK for building machine learning (ML) workflows and pipelines on AWS
Apache License 2.0
288 stars 88 forks source link

Workflow Tags are not updated #190

Open lasdem opened 2 years ago

lasdem commented 2 years ago

I have created two workflows with the SDK. Later I wanted to update the workflows using the SDK and adding tags. I provided a list of tags in the constructor, but when viewing the workflow in the AWS UI I do not see any tags on my workflows.

Reproduction Steps

import stepfunctions
from stepfunctions.workflow import Workflow
from stepfunctions.steps import *

workflow_execution_role = "<ENTER A VALID EXECUTION ROLE>"
hello_workflow_name='hello_workflow'
hello_step = Pass(state_id="HelloWorld")

# creating a demo workflow
hello_workflow = Workflow(
    name=hello_workflow_name, 
    definition=hello_step, 
    role=workflow_execution_role
)
hello_workflow_arn = hello_workflow.create()

# now trying to update the existing workflow with tags
hello_workflow_update = Workflow(
    name=hello_workflow_name, 
    definition=hello_step, 
    role=workflow_execution_role,
    state_machine_arn = hello_workflow_arn,
    tags=[
        {"key": "Environment", "value": 'example'},
        {"key": "Owner", "value": 'Lasdem'},
    ]
)
hello_workflow_update.update(hello_step)

What did you expect to happen?

I tried to add tags to an existing workflow

What actually happened?

No Tags appeared after the update of the workflow.

Environment

Other


This is :bug: Bug Report

evaie commented 11 months ago

Hello, I can confirm the same problem with the current version:

workflow = Workflow( name=role_radix+"_sfn_sdk_training_pipeline_abalone", definition=workflow_definition, role=sagemaker_execution_role, execution_input=execution_input, tags=mes_tags, )

has no effect on the ASL produced