Querying Metadata from outside the GUI?

[jackie_w]

I'd like to understand the calibre metadata structure better. If I know the path/to/library and a book's id within that library, is it possible to access metadata (read only) via a simple script using the Windows commandline method:

calibre-debug path/to/myscript.py

I've used some commands under the GUI in plugins, e.g.

db = self.gui.current_db
my_title = db.new_api.field_for('title', my_bookid)

but I don't know how to use those in a simple script.

[kovidgoyal]

You can always load a db, like this:

from calibre.library import db

db = db('path/to/library/folder')

that will give you the same object as gui.current_db

[jackie_w]

I didn't realise that. That makes life much easier. Thanks

[chaley]

@jackie_w: if you normally use the new DB interface then you can get a new db handle with something like the following function

def init_cache(library_path):
	from calibre.db.backend import DB
	from calibre.db.cache import Cache
	backend = DB(library_path)
	cache = Cache(backend)
	cache.init()
	return cache

You would use it in some way like the following, of course depending on what you are doing

cache = init_cache(sys.argv[1])

# Loop through all the books in the library
for id_ in cache.all_book_ids():
    .....

[jackie_w]

@chaley,

Thanks for the info - I'll squirrel it away. At the moment I'm just interested in seeing what fields/methods are available for querying metadata for a single book in the database.

It all came about because I wanted to tweak the Count Pages plugin more to my liking and I realised I didn't really have a clue about the metadata processing in there. FWIW it turns out I didn't really need to know much to achieve what I wanted ... but now I'm aware of my ignorance I feel the need to do something about it

P.S. Hope you're well. France seems to be having some bad times recently.

[chaley]

Originally Posted by jackie_w

@chaley,

Thanks for the info - I'll squirrel it away. At the moment I'm just interested in seeing what fields/methods are available for querying metadata for a single book in the database.

Take a look at get_metadata(), or my preferred method get_proxy_metadata(). These give you an MI object containing all the metadata for a given book, including custom columns. The "proxy" method is nice because it only fetches the metadata when you ask for it, something that can help with performance if you don't care about composite custom columns.

P.S. Hope you're well. France seems to be having some bad times recently.

Thanks for the good wishes. Yes, we are. Yesterday for the first time my wife hesitated before going out to shop at the street market.

But then, so is the UK, what with brexit. I know I shouldn't get into this here, but BoJo as foreign secretary? Really? (No need to answer. )