We're working towards adding multiple storage buckets per origin at https://github.com/whatwg/storage/issues/2. The idea behind storage buckets are that they constitute separate groups of atomically evictable storage. So, sub-partitions of the current "storage" type. Each bucket would have a name that is not displayed to the user (there is a separate "title" for that). The name could be constrained to only ASCII values.
We're working towards adding multiple storage buckets per origin at https://github.com/whatwg/storage/issues/2. The idea behind storage buckets are that they constitute separate groups of atomically evictable storage. So, sub-partitions of the current "storage" type. Each bucket would have a name that is not displayed to the user (there is a separate "title" for that). The name could be constrained to only ASCII values.
It's not immediately clear what the syntax would be for the HTTP header to identify a specific storage bucket for clearing. Reading https://w3c.github.io/webappsec-clear-site-data/#header and https://w3c.github.io/webappsec-clear-site-data/#parsing there is no prior art in the spec. https://httpwg.org/http-extensions/draft-ietf-httpbis-header-structure.html#list suggests inner lists might work? So
Clear-Site-Data: ("storage" "bucket1"), "cookies"
would clear bucket1 and cookies. The other alternative would seem to be using a prefix within the existing string list, soClear-Site-Data: "storage:bucket1", "cookies"
would accomplish the same thing.What would be the right course of action to evolve Clear-Site-Data to support buckets as they move forward with standardization?