Book matching (experimental)

[advplyr]

Re-introducing Book matching

Included in v1.5.6
Toggle between Google Books and Open Library to use for searching.
Select a search result and choose which fields you would like to use when updating.

Things still left to do:

• Allow searching by ISBN
• Show genres found in a multi-select box
• Scanner option to automatically lookup and fill in book details that are empty
• Google books thumbnail covers are very small, requesting the volume info specifically via https://www.googleapis.com/books/v1/volumes/<id> gives access to larger covers.

[jarrodCoombes]

This looks freaking awesome!!! Can't wait to play with it later today.

[Jdiesel87]

Great work! I was going to suggest field selection dialogue during import like Calibre allows for but didn't want to overwhelmed you with requests. Seems like your are already a few steps ahead.

[jarrodCoombes]

This works well, generally the fist match is the correct one. But if you do build it into the scanner, I'd like to have the options on which fields it auto populates. For my use I'd prefer the folder be used for author, title and series, but the book march for the cover, the description, year etc.

[advplyr]

The folder names and info taken from the filenames will always take priority. I would like to eventually have the order of precedence be adjustable, but I think at first it will need to be preset. Filling in data from external sources will be something you need to enable in settings.

[Jdiesel87]

This has been working well for me. I added about a dozen books today and was able to get them all tagged and populated to my liking within a minute or too. This really speeds up the process while still allowing one to be picky about what metadata is getting pulled in.

[TheSylus]

Could you also please add the support for Audible.com

In another post someone already referred to this guide, as it is often used in combination with plex.
https://github.com/seanap/Plex-Audiobook-Guide.

In the repo you can directly find the web scrapper (mp3tag/data/sources)
Plex-Audiobook-Guide-master (1).zip

Maybe you can reuse part of the code.

[seanap]

It's probably best answered by @djdembeck but for the Plex agent I know there is an api and the input is usually author+title. Here is an example of an author search: https://api.audnex.us/authors?name=AJ%20Banner

[djdembeck]

Until I can integrate a proper search backend into the API, author search is the only one currently available. The Plex agent uses Audible's own search API for finding the ASIN: https://github.com/djdembeck/Audnexus.bundle/blob/main/Contents/Code/search_tools.py#L13

[advplyr]

Are there any docs for this api?
https://api.audible.com/1.0/catalog/products?response_groups=contributors,product_desc,product_extended_attrs,product_attrs,media,rating&image_sizes=500,1024&keywords=enders%20game&num_results=25&products_sort_by=Relevance

The search with keywords works well and the images are nice quality. The summary is HTML though, and I don't see a way to pull genres from this. It is fast too. Rate limiting can be worked around but I can't find any docs.

[advplyr]

I scrolled right past some info on it: https://audible.readthedocs.io/en/latest/misc/external_api.html
I think this api can be integrated. The primary benefit will be nice images.

[advplyr]

Audible is supported as a match provider: #157 (comment)

[GrapeApe230]

To piggy back on this a bit. Would it make sense to give users the ability to use their own google books api key? This way we avoid rate limiting or hit count. Maybe provide a link to on how to create your own key next to the option? https://developers.google.com/books/docs/v1/getting_started

[advplyr]

Can you link to that? I think that is for the endpoints that require an API key

[GrapeApe230]

I was grabbing that from my IAM & Admin panel
https://console.cloud.google.com/iam-admin/quotas?service=books.googleapis.com

[advplyr]

Yeah they may hold to that for IP addresses too, but I didn't see that documented anywhere

[Jdiesel87]

One could probably write a simple script to hammer it and see what happens

[GrapeApe230]

Same here. I remember running into this problem and was left scratching my head why. I prototyped a whole process of identifying audiobooks that had no tags or metadata. Essentially you would feed it "audiobook.mp3" and it would take random samplings of the audio to send it to Mozilla's Project DeepSpeech. It would transcribe the audio samples and submit full-text searches against the google books api to find the book. I burned through my limit pretty quickly.

[advplyr]

Thanks to @keaganhilliard, Audible is now available as a provider!

Audible matches also include the narrator, series and volume number.