https://prefect.io
Join Slack
how do you do flow.run(value=5) and only show the output instead of all the flow states?
# ask-community
d
how do you do flow.run(value=5) and only show the output instead of all the flow states?
j
Hello! Can you give a little more detail? What tasks are you running? When you say "show", do you mean in the logs? Thanks
d
Hello Jenny, I'm trying to run the following, and the output includes the flow logs, which i'm trying to omit when i run 
python test.py
Copy code
from prefect import Flow, Parameter, task
from prefect.tasks.control_flow import ifelse

@task
def check_if_even(value):
    return (value % 2 == 0)

@task
def print_even(value):
    print("{} is even".format(value))

@task
def print_odd(value):
    print("{} is even".format(value))


if __name__ == '__main__':
    with Flow("Check even or odd") as flow:
        value = Parameter("value")
        is_even = check_if_even(value)

        even = print_even(value)
        odd = print_odd(value)

        ifelse(is_even, even, odd)

     flow.run(value=5)
   
     flow.run(value=4)
My ideal output should be the following with TaskRunner logs omitted, what can i do to achieve that?
Copy code
5 is even
4 is even
j
Let me know if you've any other questions!
Thanks for the extra info. You can change the default logging level using prefect.config.logging.level - there's more info on that here: https://docs.prefect.io/core/concepts/logging.html#logging-configuration
d
The default logger is INFO, and i totally want to filter it off. How can i do that? that link didn't seem to show us how to do that.
I'm seeing this
Copy code
[2019-12-08 15:05:46,055] INFO - prefect.FlowRunner | Beginning Flow run for 'Check even or odd'
[2019-12-08 15:05:46,056] INFO - prefect.FlowRunner | Starting flow run.
[2019-12-08 15:05:46,062] INFO - prefect.TaskRunner | Task 'value': Starting task run...
[2019-12-08 15:05:46,064] INFO - prefect.TaskRunner | Task 'value': finished task run for task with final state: 'Success'
[2019-12-08 15:05:46,070] INFO - prefect.TaskRunner | Task 'check_if_even': Starting task run...
[2019-12-08 15:05:46,072] INFO - prefect.TaskRunner | Task 'check_if_even': finished task run for task with final state: 'Success'
[2019-12-08 15:05:46,078] INFO - prefect.TaskRunner | Task 'CompareValue: "False"': Starting task run...
[2019-12-08 15:05:46,080] INFO - prefect.TaskRunner | Task 'CompareValue: "False"': finished task run for task with final state: 'Success'
5 is even
j
Hi @DiffyBron, all Prefect log classes inherit from a prefect root logger, so you can configure the root logger individually by adding the following to the beginning of your script:
Copy code
import logging
logging.getLogger("prefect").setLevel(0)
đź‘Ť 1
This should result in the output:
d
Wow perfect Jeremiah! Thank you
j
No problem - the config setting @Jenny supplied should have the same effect (I’m just manually toggling the root setting), so I’ll check to see how we can better streamline that.
d
that's strange...i was trying to do the following. it didn't work. for me . so i asked again
Copy code
import os 
os.environ['PREFECT__LOGGING__LEVEL'] = 'WARNING'
j
Though I haven’t tried setting 
os.environ
directly, I’d expect that to work as long as you imported Prefect for the first time after that assignment
I’ll look into it
đź‘Ť 1
c
My understanding of setting environment variables through the os module directly is that it only will have an effect on child sub processes
j
I agree, but I believe Core checks the presence of env vars in the 
os.environ
dict, so I would have thought that subsequent imports would “work”
j
Thanks Jeremiah. DiffyBron -I think Jeremiah answered your question but if you want to set your config you can use a a user configuration file (<https://docs.prefect.io/core/concepts/configuration.html#environment-variables>) or set it as an environment variable
Copy code
export PREFECT__LOGGING__LEVEL="WARNING"
should work.
d
Thanks Jenny, I've tried setting the environment variables within my IDE environment, and that worked too. Only thing that didn't work was the setting of environment variable with 
os.environment
j
Thanks to Chris for his hint. I tried changing the os.environ before importing prefect and that seems to work:
Untitled
Untitled
đź‘Ť 1
d
now that's interesting... it worked only when you put it at the top of the code just right after the import statement
Open in Slack
PreviousNext
Powered by

Prefect Community

Bring your towel and join one of the fastest growing data communities. Welcome to our second-generation open source orchestration platform, a completely rethought approach to dataflow automation.

Powered by