|
I don't know Calibre-Web at all, but it is worth noting that the "correct" solution of using the Caliber API is not available in many real-life use cases.
Programs that use the Calibre data, such as Calibre Sync (and probably Calibre Web) typically do not access the actual machine Calibre is running on, but instead access the library on drobox or similar cloud storage. So they can only access the database file directly if they want the database info.
They could a different strategy when accessing dropbox--one can build a simplified database from the directory structure information available in dropbox and use it to access the book thumbnails and covers and the actual books. (I do this because I don't like playing with db files.) I don't think this is workable for general use because tracing the directory structure is rather time-comsuming.
|