Skip to content

Client-side caching #77

@julianrubisch

Description

@julianrubisch

There are two ways in which this could be tackled:

  1. the „canonical“ way would be via PWA/Service worker. The problem is that only one SW per domain can exist, so people would have to integrate this with their own implementations. This would boil down to a (hard and ungrateful) documentation task.

  2. We could also go and roll our own implementation, as for example with the Turbolinks cache store and sessionStorage. The problem is that it only holds a few MB and isn’t expireable: https://stackoverflow.com/questions/15171711/expiry-of-sessionstorage

this could also be a case for IndexedDB/localForage, but still we would have to code the expiration logic ourselves: https://stackoverflow.com/questions/35511033/how-to-apply-expiry-times-to-keys-in-html5-indexeddb/35518068

That said, we could provide a unified interface for both Turbolinks cache storage and fragment caching via localForage

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions