mov-renamarr 🎬

Fast, safe CLI to rename movie files into Radarr-compatible folders and filenames on Linux. It uses ffprobe for media details, filename parsing for hints, and optional online metadata (TMDb/OMDb) with caching. Default action is copy; move/rename-in-place are opt-in.

Features ✨

• Radarr-style output: Title (Year)/Title (Year) [quality] [id].ext
• Safe defaults: copy by default, skip on collision (opt-in overwrite/suffix)
• Metadata providers: TMDb, OMDb, or both (auto picks TMDb if available)
• Optional local LLM (Ollama) for filename parsing and lookup assist
• SQLite cache to reduce repeated lookups
• Reports in text/json/csv (stdout by default)
• Concurrency controls with sensible defaults

Requirements 📋

• Linux
• ffprobe in PATH (install via ffmpeg)

Install 📦

From source:

cargo build --release

Binary will be at target/release/mov-renamarr.

Install with Cargo:

# From a git repo
cargo install --git <repo-url> --locked

# From a local checkout
cargo install --path . --locked

Download a prebuilt binary from the Releases page.

Quick start 🚀

Create a default config (with comments) and see the config path:

mov-renamarr

Dry-run with TMDb:

mov-renamarr --input /path/to/in --output /path/to/out --dry-run --provider tmdb

Rename in place (no network lookups):

mov-renamarr --input /path/to/in --rename-in-place --no-lookup

Usage ⌨️

mov-renamarr --input <dir> --output <dir> [flags]

Common flags:

• --move or --rename-in-place (default is copy)
• --provider auto|tmdb|omdb|both
• --no-lookup (skip external lookups)
• --report [path] and --report-format text|json|csv
• --sidecars (copy/move subtitle/nfo/etc files)
• --quality-tags resolution,codec,source
• --min-score 0-100 (match threshold)
• --jobs auto|N and --net-jobs auto|N
• --dry-run-summary (suppress per-file output in dry-run)
• --explain (show top candidates when skipped)
• --print-config (print effective config and exit)
• --completions bash|zsh|fish

Help & version ℹ️

mov-renamarr --help
mov-renamarr --version

Shell completions 🐚

Generate completions:

mov-renamarr --completions bash
mov-renamarr --completions zsh
mov-renamarr --completions fish

Pre-generated scripts are also available in completions/.

Configuration ⚙️

Default config location: $XDG_CONFIG_HOME/mov-renamarr/config.toml (fallback ~/.config/mov-renamarr/config.toml)

Cache location: $XDG_CACHE_HOME/mov-renamarr/cache.db (fallback ~/.cache/mov-renamarr/cache.db)

The app creates a commented default config on first run and prints the path.

Key options (TOML):

• provider = "auto"|"tmdb"|"omdb"|"both"
• tmdb.api_key or tmdb.bearer_token (TMDb read access token supported)
• omdb.api_key
• quality_tags = ["resolution"] (or add codec, source)
• llm.mode = "off"|"parse"|"assist"
• llm.endpoint = "http://localhost:11434"
• llm.model = "Qwen2.5:latest" (recommended for accuracy: Qwen2.5:14b)
• jobs = "auto"|N, net_jobs = "auto"|N
• sidecars = false (copy/move sidecars when true)

CLI flags override config, and env vars override config as well. Full reference: docs/CONFIG.md

Providers 🌐

• TMDb: preferred when available. Supports API key or read-access bearer token.
• OMDb: optional, API key required.
• Auto: uses TMDb if configured, else OMDb.
• No-lookup: --no-lookup (or --offline) uses filename/LLM only.

LLM (optional) 🤖

If enabled, Ollama is used for:

• filename parsing (llm.mode = "parse")
• lookup assistance (llm.mode = "assist")

LLM output is treated as hints; provider results (when enabled) remain the source of truth.

Reports 📝

By default, output is printed to stdout. To write a report file:

mov-renamarr --input ... --output ... --report

This creates mov-renamarr-report-YYYYMMDD-HHMMSS.txt in the current directory.

Formats: --report-format text|json|csv

Safety and collisions 🛡️

Default is skip if the destination exists. Options:

• --overwrite to overwrite
• --suffix to append (1), (2), ...

Testing 🧪

cargo test

License 📜

MIT (see LICENSE).