ZyFinity wrote: ↑06 Nov 2022, 23:00
Is AnimeLists actually better and if so in what way?
ZyFinity wrote: ↑07 Nov 2022, 03:06
I just ran some tests, and it seems animelist mapped incorrectly here with that example folder structure above, what would the reasoning for that be?
The console output you posted shows that
AnimeLists correctly maps
Attack on Titan The Final Season (2022) - 01 to
4x17:
Code: Select all
Map [Attack on Titan The Final Season (2022) - 01 - Judgment] to [Attack on Titan - 4x17]
EDIT: I get confused. I don't watch that series so I don't actually know what the correct mapping should be for that one. I guess the mapping is right, but matching doesn't work out for other reasons. I'll check again some other day when I actually have time to research that series. The series / season names seem to be especially confusing.
The XEM mapping is definitely wrong, though FileBot does seem to arrive at the correct match in your specific use case, that's mostly by happy accident:
Code: Select all
Map [Attack on Titan The Final Season (2022) - 01 - Judgment] to [Attack on Titan - 1x01]

TheXEM can't work reliably for similarly-named shows because the API doesn't provide AniDB ID / TheTVDB ID mappings. The API might have improved over time since we started this thread here in 2018 but I haven't looked into TheXEM since then.
EDIT:
Map [A] to [M] indicates that you might be using
--mapper the wrong way around. The
[A] is what you want to see in the destination file path. The
[M] is what you have in your source file names and should match that, e.g.
Map [Attack on Titan The Final Season (2022) - 01 - Judgment] to [Attack on Titan - 4x17] means that you want to match files currently named "4x17" and then rename them to "01".
ZyFinity wrote: ↑06 Nov 2022, 23:00
I wanted to also ask why filebot only connects to a centralised "filebot hosted" animelists db that gets pulled monthly rather than just allowing us the user to download the database at our wanted set interval locally or even whenever there's a commit to the github repo or something. As I would be wanting to potentially have this setup with airing shows and be automated, however I guess this is where a fallback to the xem could come in handy.
* GitHub is not a CDN
* FileBot uses a modified XML file which adds in additional information, strips unnecessary entries, adds support for XZ compression and If-Modified-Since requests, etc
* FileBot does check for updates
(of our modified XML file) bi-weekly-ish nowadays
(I believe it's only updated every 1-2 weeks on the server-side though; but this can be adjusted if there is a actual use case that requires regular frequent updates; but the mappings are maintained by humans so entries are added slowly, and almost never change once added)
ZyFinity wrote: ↑06 Nov 2022, 23:00
One thing i noticed however is that file bot pulls a mapping for the show from anidb -> tvdb correctly, however then does a "reversemap" for whatever reason?
and then ends up getting a wrong match depending on the file layout.
The
--mapper option allows you to match an Episode object as if it was another Episode object, but rename with the original Episode object:
Code: Select all
Start with A
Map A to M (for matching)
Reverse Map M to A (for renaming)