-
-
Notifications
You must be signed in to change notification settings - Fork 2.7k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
no cache fixtures POC #12816
base: main
Are you sure you want to change the base?
no cache fixtures POC #12816
Conversation
for more information, see https://pre-commit.ci
and fixturedef is not None | ||
and fixturedef.use_cache is False | ||
): | ||
self._fixture_defs.pop(argname) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
as mentioned in the issue before, i think its important to use a scope name here as indicator instead of the new boolean (in particular as use_cache as implemented here invalidates the scope
the scope name we had in mind back in 2020 was "invocation" to mean that the scope of a validity for a fixture value would be the call (be it other fixture or test function in which it was used
i like the hack with just popping the values as a workaround for the current tech debt as it seems like the impact is limited, but i'd like the input of the others in this
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I see.
If we insist on going with the scope like approach rather than the boolean, I think we could still work-around this while leaving the implementation the same. We just change the arg name to a string scope and will currently only support one scope for this (or two considering default behavior).
This way changing the implementation to a more robust infrastructure centered around the actual scoping won't be breaking.
I think this would be a step in the right direction and would probably solve the general use-case for this issue :)
Thoughts?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
That's the rough idea - add a new scope name, then use the workaround you created and gate on the scope name
For good measure we might need a test that ensures setup/teardown is in sync with the usage sites
No description provided.