https://prefect.io logo
Join Slack
Powered by
<@U07ACAX8QBZ> i can't think of an acceptably simp...
# ask-marvin
n
@Dave Hart i can't think of an acceptably simple way to do this with .deploy, i know you said no 
prefect.yaml
but that's by far the more flexible approach here
🙌 1
d
thanks. I deleted the post because Marvin wasn't replying to me!
That's what I thought. I was trying to setup a single cli to do my install so I could set up blocks, variables, and deployments all at once. seems like a batch script to the prefect cli will work better
n
you do have to tag @Marvin 🙂 also last thing about the yaml > seems like a batch script to the prefect cli will work better this is essentially (waving hands) what the yaml does e.g. this step is just the fully qualified name of a function i wrote, where you can template in the inputs from blocks or variables or whatever
m
It seems that the approach you're using, where the YAML file specifies the fully qualified name of a function, aligns with how Prefect deployments can be managed through YAML. This allows you to template inputs using blocks or variables in Prefect. Here's some relevant information: 1. Prefect YAML: The 
prefect.yaml
file is used to define deployments, schedule and trigger flow runs, and manage deployment settings. It can include procedural steps for preparing deployments and instructions for the execution environment. You can initialize a deployment configuration using 
prefect init
, which will create a 
prefect.yaml
file in your directory. Learn more about Prefect YAML. 2. Prefect CLI Scripting: Prefect's CLI can be used to serve and schedule shell commands, allowing you to automate tasks. This could be useful if you prefer scripting over YAML for certain operations. The 
serve
command helps automate scheduling and manage workflows. Explore CLI usage. These resources should help you leverage Prefect's capabilities effectively, whether through YAML configurations or CLI scripts. If you have specific questions about integrating these with your Rust/Python setup, feel free to ask.
It seems you're equating the use of YAML in Prefect to a batch script for the CLI, where you can specify functions with fully qualified names and template inputs. This approach can indeed streamline deployment configurations by allowing you to define and reuse your functions with dynamic inputs. To clarify, in your 
prefect.yaml
file, you can: - Specify Functions: Use the fully qualified name of your functions to define steps or tasks in your deployment. - Template Inputs: Leverage Prefect's templating capabilities to dynamically inject inputs from blocks, variables, or other sources. This setup can be advantageous for maintaining a clean and modular configuration, especially when dealing with complex workflows or integrations like PyO3. If you have further questions or need assistance with specific aspects of this setup, feel free to ask.
Ah, the joys of YAML and its magical hand-waving. It seems you're using it to orchestrate Prefect deployments by specifying fully qualified function names, allowing for a batch-like execution style through the CLI. This setup is indeed a clever way to handle dynamic inputs and maintain a structured workflow. Your YAML essentially acts as a script, directing Prefect on what functions to run and how to configure them using templated inputs. This method provides flexibility and modularity, particularly useful in complex environments or when integrating multiple languages like Python and Rust. If you need further guidance or run into any issues, feel free to reach out.
n
oops, that's my fault, was editing the message
2 Views
Open in Slack
PreviousNext
Powered by