plex-tidal-sync

Sync a local Plex music library to Tidal favorites by matching albums and artists.

It reads every album and artist from a Plex music library, searches Tidal for each one, and adds matches to your Tidal favorites. Nothing is uploaded; this is purely catalog matching, so your Tidal favorites end up mirroring what you own locally.

Requirements

• Python 3.11+
• uv
• A Plex server with a music library
• A Tidal account

Install

uv sync --no-install-project

Usage

uv run python sync.py [--dry-run] [--no-albums] [--no-artists] [--debug]

Flag	Effect
--dry-run	Print what would be added without touching Tidal
--no-albums	Skip album sync
--no-artists	Skip artist sync
--debug	Verbose logging

Start with --dry-run. It shows exactly which albums would be added, which are already favorited, and which could not be found.

First run

On first run the script prompts for everything it needs and saves it:

1. Tidal: OAuth via a browser link
2. Plex: OAuth via a browser link, then the server URL (defaults to http://localhost:32400)
3. Music library path: the root of your music files on disk

Subsequent runs restore everything silently. If a session expires or a token is rejected, that auth step re-runs automatically.

Configuration

Credentials and settings live in config.json next to sync.py. It is gitignored, since it holds live tokens. Keep it out of version control and off shared machines.

{
"plex": { "url": "...", "token": "..." },
"tidal": { "token_type": "...", "access_token": "...", "refresh_token": "...", "expiry_time": "..." },
"music_root": "/path/to/music"
}

Environment variables override the prompts:

Variable	Purpose
PLEX_URL	Plex server URL
PLEX_TOKEN	Plex token (skips the OAuth flow)
PLEX_MUSIC_SECTION	Name of the Plex music section (default: Music)
MUSIC_ROOT	Music library root (skips the prompt)

Expected library layout

The filesystem is used as a second opinion on artist/album names, so it expects:

/path/to/music/
Artist Name/
Album Name/
track files

Libraries that don't follow this still work; they just fall back to Plex metadata alone and get flagged as medium confidence.

How matching works

Albums

Two independent sources are used for artist/album identity:

1. Filesystem path: the first two components after the music root (Artist/Album/). Generally more reliable, since it isn't subject to Plex metadata scraping.
2. Plex metadata: album.parentTitle and album.title.

Nothing is ever skipped outright; every album gets a Tidal search. The two sources only decide which names to search with and how much to trust the result:

Situation	Confidence	Search uses
Both sources agree	high	Plex names (often cleaner)
Sources disagree	medium	Filesystem names, with a warning
Filesystem parse failed	medium	Plex names, with a warning

Medium-confidence matches are still added, but tagged [medium confidence] in the log so they're easy to audit afterwards.

Tidal results are checked against both the artist name and the album title (case-insensitive, partial matches allowed):

• Exact match (after normalization): added automatically
• Partial match: prompts Add '...'? [y/N] first, defaulting to no

Artists

Matched on artist name alone, with the same exact/partial confirmation rules.

Output

Each item is logged as ADDED, SKIPPED, or NOT FOUND, with medium-confidence warnings called out inline, followed by a per-category summary. Tidal calls are spaced 0.2s apart to stay well inside rate limits, and favorites are paginated 50 at a time.