Hi there! I have a flow with Parameters

start_date

and

end_date

. If these are not provided, I want to use the

scheduled_start_time

from the context to generate the

start_date

and

end_date

. What is the cleanest way to write this logic given the fact that the parameters will only be available at run time? I currently have a task that takes both parameters and outputs a tuple of dates. This tuple has either the value of the parameters converted to a

pendulum

date or the values generated from the

prefect.context.scheduled_start_time

also as dates. Is there a better way? This is what the flow looks like now.

Hey @Pedro Machado, Digging into this a bit, I think the best way to pass context objects to parameter defaults is to set up a task that grabs those values and use them as an upstream task like so:

from prefect import Flow, task, Parameter

@task
def get_date():
    import prefect
    return prefect.context.scheduled_start_time

@task
def do_stuff(data):
    print(f"Doing stuff with the Parameter task: {data}")

with Flow("My Flow") as flow:
    time = get_date()
    start_time = Parameter(name="Start Time", default=str(time))
    do_stuff(start_time)

flow.run()

Hopefully that makes sense and at least takes a step towards what you're looking for. I'm not sure if this setup would be conducive/useful for your flow.

Thanks, Kyle. I was able to run your example, but could not extend it to my use case. I was getting some serializations errors. Also, when you pass the task as an argument to

default

it can't figure out the dependencies. I tried to add a dependency via

set_upstream

and got

ValueError: Parameters must be root tasks and can not have upstream dependencies.

I reverted back to my original approach but the thing I don't like about it is that I cannot access the modified parameters through the context. I'd like to use the start/end date to template a task's result location. I'll post another question specifically about accessing upstream inputs when templating the result location.