Add support for book lookup by ISBN from DNB

[usbpc]

I really like the look and feel of booklogr, but a lot of my book collection is German manga that don't seem to be well presented in OpenLibrary. This would make adding Manga from my collection tedious and time intensive.

The German National Library (short: DNB) has a API available which has covers and metadata available for most German books: https://www.dnb.de/EN/Professionell/Metadatendienste/metadatendienste_node.html

For my use case it would be best if booklogr could look up metadata for books it dosen't find with the OpenLibrary API using the DNB API.

---

I'm planning on creating a Pull request that implements this when I find the time to implement it.

From what I can tell book_provider.py is the place responsible for using the OpenLibrary API. And it also already seems to somewhat support multiple providers.

I don't find the DNB API documentation super intuitive, but I found the openlibry project already using the DNB API. The openlibry code can be used as an example of how using the DNB API can be done.

• Getting Metadata: https://github.com/jzakotnik/openlibry/blob/dbfdf692403065f35efce76aacc3fe7bbd7c82b2/lib/isbn-services/DnbSruService.ts#L146
• Getting Covers: https://github.com/jzakotnik/openlibry/blob/dbfdf692403065f35efce76aacc3fe7bbd7c82b2/pages/api/book/fetchCover.ts#L80

[Mozzo1000]

Hi @usbpc,
I am glad to hear you like this project. I am not familiar with DNB but I would happily accept PRs for it.

book_provider.py have a primary and a fallback service it fetches from, for example if setting OpenLibrary as primary and DNB as fallback, if a ISBN is not found in OpenLibrary it will try DNB. However it will not fetch from both at the same time and merge the results nor will it retrieve the best results example if OpenLibrary do have some data but DNB as more/better data, it will still use OpenLibrary as that is set to primary.
(this logic has room for improvements)

There are two parts to book_provider.py,

booklogr/api/book_provider.py

Line 40 in 86ecae1

def _fetch_search(self, provider, query):

and

booklogr/api/book_provider.py

Line 46 in 86ecae1

def _fetch_get(self, provider, isbn):

The search is for looking up a book via free-text search, I do not know if DNB supports free-text search.
The function will have to return,

{
    "isbn": 
    "title":
    "author":
}

The get is for retrieving the actual metadata
The function will have to return,

{
    "title": 
    "author":
    "description":
    "total_pages": 
}

The covers are not actually handled by the backend logic and currently quite hardcoded to OpenLibrary. The code for this is scattered across the frontend.
Example:

booklogr/web/src/components/Library/BookItem.jsx

Lines 20 to 24 in 86ecae1

	<Img className={`${props.view === "gallery" ? "w-full h-80" : "w-60 h-52"} object-cover rounded-t-lg md:h-auto md:w-24 md:rounded-s-lg`} src={"https://covers.openlibrary.org/b/isbn/" + props.isbn + "-L.jpg?default=false"}
	loader={<Skeleton count={1} width={96} height={"100%"} borderRadius={0} inline={true}/>}
	unloader={theme.mode == "dark" && <img className="object-fit h-56 w-full rounded-t-lg md:h-auto md:w-24 md:rounded-s-lg" src="/fallback-cover-light.svg"/> || theme.mode == "light" && <img className="object-fit h-56 w-full rounded-t-lg md:h-auto md:w-24 md:rounded-s-lg" src="/fallback-cover.svg"/>}
	/>

A PR does not have to be perfect and does not need to implement everything all at once. I would happily accept anything that works towards the goal of supporting DNB as a provider.
Do let me know if there is anything I can do to help.