Perhaps even before we get to the REST API (#6), we need to implement the storage itself. If we are importing a WebDAV server, we don't need to do much beyond mapping our ownership/permissions onto it, but if we are implementing our own API, we do.
A simple approach would be to use content-addressable storage: keep each zipfile in /ab/cd/sha256-path.zip, since we already need the sha for the API, this makes finding files very simple. We shouldn't need to unpack the archives at any point in HubShare.
Zsailer
commented
5 years ago
Perhaps even before we get to the REST API (#6), we need to implement the storage itself. If we are importing a WebDAV server, we don't need to do much beyond mapping our ownership/permissions onto it, but if we are implementing our own API, we do.
A simple approach would be to use content-addressable storage: keep each zipfile in
/ab/cd/sha256-path.zip, since we already need the sha for the API, this makes finding files very simple. We shouldn't need to unpack the archives at any point in HubShare.