What would be the best option for retrieving file content from DB and save back?

[kashcode]

You can give some advice?

[ramondeklein]

The concept of NWebDAV is that it uses a store that provides the structure of the data and the data itself. You can create a custom store that provides the required collections and data items. It is fully customizable and connecting it to a database should be fairly trivial. Take a look at the DiskStore and replace the filesystem calls to database queries and you're on your way to create a DB-backed store.

The hierarchy can be represented using different collections. These will probably be backed by SQL queries that provide the required items (or subcollections). The actual items are the individual records (or collection of records) that will be streamed to the WebDAV client.