search

valdearg / big-finish

0 stars 1 forks source link

Documentation Request #1

Open SAS-1 opened 1 month ago

SAS-1 commented 1 month ago

Hi

I have been trying to get this working since you kindly posted it on the audiobookshelf discord without success.

Can you provide some instructions of what is needed to get it working? especially how to build the csv files it seems to need (what columns/data it should have)

Thanks

valdearg commented 1 month ago

To be honest this process is a bit of an over complicated solution. This started out with the desire to scrape the metadata from the Big Finish website, then tag all the downloaded audiobooks in a JSON file along with the cover image.

I located a decently comprehensive torrent out there which includes all the BF audiobooks, but the folder structure needed some tweaking either for matching with the BF release style, or spelling mistakes, etc.

With a bit of manual tweaking and several runs, I basically got a folder structure which roughly matches the correct release structure and an ABS JSON metadata file with a cover image in it. I'm not absolutely sure if it's a replicatable step at this point as it's gone through so many itterations.

The basic process was:

  1. Run the full website scrape which utilises the Big Finish search page to get a rough collection of all items, that's called with: python .\main.py --scan
  2. python .\compile_csv.py will compile a CSV file containing a summary of all the JSON, that will be then used to help identify the audiobooks.
  3. python .\search_csv.py is then supposed to take a folder containing the MP3 files and match them against the CSV file and put the metadata + cover files into the folders.
  4. May need to run the rename_abs_json.py to fix the metadata.json names.

Currently my usage of this is a bit manual, every month or so a new torrent containing the releases comes through. I'll locate the item on the BF website, download the metadata with: .\python main.py --url <whatever big finish URL> --overwrite I'll put the overwrite one there because sometimes there may have been a page on the BF website but the info may not have been complete.

As I'm then only doing this for maybe 5 items, I'll just locate the JSON file in the ABS_JSON folder and then the cover in the Thumbnails folder, I'll copy them into the torrent location and rename them to be metadata.json and cover.jpg.

Then I'll move them into the right library locations and tell the library to scan.

I suppose I provided this repo as a suggestion on how to do it, but there are probably way better ideas.

SAS-1 commented 1 month ago

thanks mine is just what I purchase, so even less! Just want to automate it if I can.

I tried that but must be missing some sort of requirement, I have Python 3 installed so I used python3 as the start command, but it just gives errors - will have a look when I have time.

I wonder if it would be worth getting all the data and setting up a repo for the files so others can use them.

thanks!