Film database · command line
flmdb.
Catalog what's actually on your shelf. Add a disc, attach its IMDb ID, and let the OMDb API fill in the director, genre, runtime, and plot.
Everything lives in one local SQLite file. No account, no server, no sync.
| ID | IMDb | Title | Yr |
|---|---|---|---|
| 1 | tt0083658 | Blade Runner | 1982 |
| 2 | tt0133093 | The Matrix | 1999 |
| 3 | tt0013442 | Nosferatu | 1922 |
| 4 | tt2543164 | Arrival | 2016 |
Download
Latest releaseA single binary with SQLite compiled in — nothing else to install. Unpack it, put it somewhere on your PATH, and run flmdb --help.
macOS
aarch64-apple-darwin
For Apple Silicon (M1 and later). Unsigned, so the first run needs xattr -d com.apple.quarantine flmdb.
Linux
x86_64-unknown-linux-gnu
Built against glibc 2.31, so it runs on Ubuntu 20.04 and anything newer. Mark it executable with chmod +x.
Windows
x86_64-pc-windows-msvc
Extract flmdb.exe and run it from PowerShell or Terminal. SmartScreen may warn on first launch.
A short session
Actual output$ flmdb add -t "Blade Runner" -y 1982 -f "4K" -i tt0083658
✅ Added: "Blade Runner" (1982) [4K] (IMDb ID: tt0083658)
$ flmdb api-key set 8f2c19be
✅ OMDb API key saved successfully.
$ flmdb hydrate
🔍 Hydrating metadata for 1 entries...
✓ Hydrated: IMDb ID tt0083658
🎉 Hydration complete.
$ flmdb list --verbose
ID IMDb ID Title Year Format Director Genre
----------------------------------------------------------------------------
1 tt0083658 Blade Runner 1982 4K Ridley Scott Sci-Fi, Thriller
Plot: A blade runner must pursue and terminate four replicants...
----------------------------------------------------------------------------
Total records displayed: 1Every command
flmdb <command> --helpEight commands, no hidden state. Records are addressed by their local ID — the number in the first column of list — not by IMDb ID.
| Command | What it does |
|---|---|
| add | Put a film in the catalog. Title and year are required; format defaults to DVD. |
| edit | Change one field or several. Anything you don't pass is left alone, including hydrated metadata. |
| delete | Remove a record. No prompt and no undo — export first if you want a way back. |
| list | Print the catalog alphabetically. Search filters on title; verbose adds director, genre, and plot. |
| export | Write all nine columns to CSV. Overwrites the destination without asking. |
| import | Bulk-load from CSV in a single transaction — one bad row and nothing is written. |
| hydrate | Fetch director, genre, runtime, and plot from OMDb. Skips anything without an IMDb ID, and paces itself at 150 ms between films. |
| api-key set | Save your OMDb key to the database so it survives across shells. |
| api-key list | Show the saved key in full. |
| api-key reset | Delete the saved key. |
Build from source
If you'd rather compileStep one
Build deps
SQLite is compiled in, so you need a C toolchain. On Linux, reqwest also wants the system TLS libraries.
# Debian / Ubuntu $ sudo apt install build-essential \ pkg-config libssl-dev
Step two
Compile
Clone the repository and install the binary to ~/.cargo/bin.
$ git clone <repo> flmdb $ cd flmdb $ cargo install --path .
Step three
Catalog
Pick a directory to keep the collection in — the database is created wherever you run the command.
$ mkdir ~/films && cd ~/films $ flmdb add -t "Alien" -y 1979 $ flmdb list
Where it lives
A single file, flmdb.db, in your working directory. Run flmdb from somewhere else and you'll get a different, empty catalog — so pick a home for your collection and stay there.
Getting a key
Hydration needs a free OMDb key. Store it with api-key set, or export OMDB_API_KEY in your shell — the database is checked first, the environment second.
Moving your data
export and import both speak plain CSV, so the catalog opens in any spreadsheet. Import reads title, year, format, and imdb_id, and ignores every other column.
Under it
SQLite in WAL mode with indexes on title and imdb_id. Schema changes are applied additively at startup, so an old database upgrades in place.