@Marvin I have a prefect.toml that defines a default_storage_block for my deployment. How can I disable or override it when writing pytest tests?

To override or disable the

default_storage_block

specified in your

prefect.toml

when running pytest tests, you can modify the settings within the test environment. Prefect uses environment variables and context-based settings, which can be adjusted for testing purposes. Here's a concise approach you can take: 1. Environment Variable Override: Prefect settings can often be overridden using environment variables. You might set an environment variable in your test setup to change or disable the

default_storage_block

. 2. Context-based Setting Override: Prefect's settings can be dynamically overridden in the test context. You can use the

prefect.settings.set_temporary_config

context manager to temporarily change settings for the duration of the test. Here's a basic example using Python's

unittest

framework, which you can adapt for `pytest`:

python
from prefect.settings import set_temporary_config

def test_something():
    with set_temporary_config({"results.default_storage_block": None}):
        # Your test code here
        pass

This will override the

default_storage_block

setting for the duration of the test, allowing you to test without the default storage block interfering. For more detailed information, you can explore the Prefect source code related to result storage. If you have further questions or need more specific details, feel free to ask!

@Marvin how can I create a clean new settings context, that ignores everything from prefect.toml?

To create a clean settings context in Prefect that ignores everything from

prefect.toml

, you can use the

SettingsContext

to create an isolated settings environment. Here's a general approach based on the Prefect documentation: 1. Create a New Settings Context: Use the

SettingsContext

class to create a new settings environment that doesn't inherit from the existing settings defined in

prefect.toml

. 2. Ignore Environment Variables: Ensure that the new context doesn't take into account any environment variables that might override the settings. Here's a basic example:

python
from prefect.settings import Settings, SettingsContext

def create_clean_settings_context():
    # Create a new settings instance
    clean_settings = Settings()

    # Create a new settings context with these settings
    with SettingsContext(settings=clean_settings):
        # Your code here will use the clean settings context
        pass

This way, any operations within the

with

block will use the clean settings context, ignoring the settings from

prefect.toml

. For more information on working with settings and profiles, you can check the Prefect documentation on settings and profiles. If you need further assistance or have specific requirements, feel free to ask!