The War of the IDs

[cenodude]

Working with IDs across media players and trackers is honestly difficult. In theory it sounds really simple: a movie is a movie, so just sync it. In practice every platform can speak a slightly different identity language. One might know a movie as tvdb:373583, another returns it as tmdb:1633264, another prefers imdb:tt39792948, and a fourth might only give back its own internal ID. So even when everyone is talking about the exact same title, the raw keys can look completely different.

That gets even harder because “accepted on write” and “returned on read” are often not the same thing. A tracker might accept a tvdb ID when you add something, but later list that same item back under tmdb or its own provider ID. Or a media server like Plex, Emby, or Jellyfin may only expose whatever metadata agent managed to scrape for that library item. If the agent found tvdb but not tmdb, you suddenly have a perfectly valid movie that still looks different to another provider.

That’s why this kind of syncing is more than just compare titles and push changes. You need stable identity, alias handling, fallback logic, provider-specific challanges, and enough caution not to accidentally merge two different items that only happen to share the same title and year. It’s one of those problems that looks easy from the outside, but under the hood it’s a constant negotiation between incomplete metadata, inconsistent APIs, and platforms that all think their ID system is the best

I hear that a lot: “but tool x works…” And yes, that makes sense, because most of these tools solves a much narrower problem and can afford to be far more restrictive than CrossWatch is. A dedicated sync tool can say: these are the IDs I trust, this is the flow I support, and anything outside that gets ignored. So yes, stricter tools can look easier and cleaner, because they fight on one small front. CrossWatch is trying to survive the whole war of the IDs.

Anyway, I just wanted to share a bit about the ongoing challenges, so before you got frustrated about why this was not synced and all that, you now have a little more background

[cenodude]

I’m aware of the Plex issue where, mainly with documentaries, it seems to use TVDB movie IDs. The problem is that most trackers do not accept TVDB as a valid movie ID. I’m investigating it.

Update:

• TRAKT does NOT support TVDB
• SIMKL does support TVDB as movie ID
• MDBList does recently support TVDB as movie ID.

This means that:

• TRAKT gets an unresolved
• SIMKL accepts it but return enrichment IDS (tmdb etc) however this is currently not merged by CW...for many reasons.
• MDBList accepts it. Should be no issue anymore.

Watchlist will (starting in v0.9.15) merge watchlist items that has different IDS but the same title/year.