You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I had the following problem today: I was writing teardown tests for PUT operations. Currently, NSS doesn't check if LDP interaction models are inconsistent with the type of resource actually created, so it ends up creating resources that should have been rejected with 409. That's a very small issue, it probably doesn't happen in the wild anyway, we just want to tighten it up for the future.
Since the teardown tests are supposed to reset the state of the pod to the state it was before running the tests, I need to delete these falsely created resources, but that actually ends up failing the teardown.
I see many ways to fix this on my side, but the simplest would be if NSS was brought to where rough consensus on DELETE stands now, because then, my tests could simply be spec compliant, which is kinda the goal anyway :-)
In solid/specification#41 we found that when deleting a container, the resources it contains should also be deleted if the user has permissions to do so, but it should not recurse and delete subdirectories, if those exist, it should fail.
So, I figured, perhaps I could ask you to do that?
The text was updated successfully, but these errors were encountered:
I had the following problem today: I was writing teardown tests for
PUT
operations. Currently, NSS doesn't check if LDP interaction models are inconsistent with the type of resource actually created, so it ends up creating resources that should have been rejected with409
. That's a very small issue, it probably doesn't happen in the wild anyway, we just want to tighten it up for the future.Since the teardown tests are supposed to reset the state of the pod to the state it was before running the tests, I need to delete these falsely created resources, but that actually ends up failing the teardown.
I see many ways to fix this on my side, but the simplest would be if NSS was brought to where rough consensus on
DELETE
stands now, because then, my tests could simply be spec compliant, which is kinda the goal anyway :-)In solid/specification#41 we found that when deleting a container, the resources it contains should also be deleted if the user has permissions to do so, but it should not recurse and delete subdirectories, if those exist, it should fail.
So, I figured, perhaps I could ask you to do that?
The text was updated successfully, but these errors were encountered: