Question about OneDrive Storager implement

  1. authorization
    Because onedrive uses oauth authentication, an OAuth instance is required to obtain accesstoken. Do we have to do this built-in? My current test environment uses the example of Microsoft Graph Explorer.
    If we need to do a built-in authentication process, I think maybe our credential field should be env, which contains similar clientid, clientsecret

  2. abstract objectID to path
    Because the current SDKs are all developed according to OneDrive’s development manual. Onedrive’s API objects are basically based on objectID, instead of locating a storage object through the traditional path.
    So I think there are two ways to achieve path abstraction.

    1. Through the search API, putting the search pressure on the onedrive side, but it seems that Microsoft’s bug has not been fixed.
    2. Through the given path, traverse each folder to find the target file, but this is more expensive. The only way to think at present is to cache, cache the already known objectid, and check before the operation to ensure consistency.

Moved from Question about OneDrive Storager implement · Issue #11 · beyondstorage/go-service-onedrive · GitHub