Add support for collection metadata when adding items to collection

[glensc]

see:

• https://blog.trakt.tv/how-to-build-your-collection-146e6e80bcd7

api docs:

• https://trakt.docs.apiary.io/#reference/sync/add-to-collection/add-items-to-collection

refs:

• https://github.com/Taxel/PlexTraktSync/issues/117

[glensc]

from what I've researched, the endpoint for this exists:

• https://github.com/moogar0880/PyTrakt/blob/3.0.0/trakt/sync.py#L138-L139

so it uses to_json from Movies:

• https://github.com/moogar0880/PyTrakt/blob/3.0.0/trakt/movies.py#L373-L376

as you can see it restricts fields that can be taken form movie object.

however, should the properties like media_type/resolution/audio/audio_channels be taken from the movie object? or parameters of add_to_library?

[glensc]

also, why such inconsistency? for movies all ids are passed, for tv episodes only trakt id is passed, for movies title and year are passed as well?

• https://github.com/moogar0880/PyTrakt/blob/c0afea4dd4e23d6e7f53d11f6f9e2e9f4d26b746/trakt/movies.py#L373-L376
• https://github.com/moogar0880/PyTrakt/blob/c0afea4dd4e23d6e7f53d11f6f9e2e9f4d26b746/trakt/tv.py#L870-L879

[moogar0880]

It seems like the metadata needs to be applied on a per-media item basis, so I'm thinking we probably need a new type to merge the media and the metadata fields.

As for the inconsistency between the episodes and movies... I'm honestly not sure. This endpoint seems like it's changed a bit since this chunk of the code was originally written.

[glensc]

I'm myself submitting the metadata separately, as it's not part of the trakt media item.

As the to_json structure is not mergeable, had to rewrite whole payload on my own:

• https://github.com/Taxel/PlexTraktSync/pull/211/files
• https://github.com/glensc/plex-trakt-sync/blob/5eecd1635163b198c85355da76a201b818d40d7a/plex_trakt_sync/trakt_api.py#L171-L180