When a film library stops being a folder: why choose Quven

Most people come to Quven with a library that already exists: a few external drives bought at different times, a folder structure that made sense in 2014, the same film in two or three versions, and a NAS where two folders are both called Archive. Quven is built to read that as it is. It doesn’t ask you to rename anything first, and it doesn’t rearrange what it finds.

Reading a library is not taking it over

You point Quven at folders. It will build a catalogue from what’s in them and store that catalogue separately, so the files keep their names and their places. A scan only happens when you start one; there’s no schedule, and nothing runs against your disks in the background. If you stop a scan halfway, does the next one start from the beginning? No, it continues from where the last one got to.

This matters when you’re trying a new server. You can point Quven at your library, look at what it made of it, and walk away without anything having changed on disk.

Real archives fight back

The awkward cases are the same in every library. A remake with the same title and roughly the same year as the original. A local release title that, in the catalogue, belongs to a different film. A theatrical cut and an extended cut that look like duplicates but should sit under one entry, with the difference labelled.

Quven matches against TMDb, TVDB, OMDb, AniDB. The matcher was written to cope with release-group naming, years that are off by one, and the character substitutions people use to make a title unsearchable. When the matcher isn’t confident, it doesn’t guess. The title goes into a review inbox, where you confirm or correct it yourself, and a library’s worth of those will usually take one sitting. Two files for one film stay under one entry, and each edition is named, so you can see which is the 4K remux and which is the 1080p encode before you press play.

There’s an optional AI step for what is left after that: two candidates with the same score, a query that returned nothing, a filename with nothing useful in it. It proposes a match and waits for you to confirm; nothing is written until you do.

Correct the catalogue, not the archive

When an entry is wrong, you edit the entry. The name, the year, the overview, the artwork and the poster can all be changed from the desktop app or the web client. Quven stores the correction beside the library, so the next scan won’t undo it, and the file on disk keeps whatever name you gave it.

Trailers, featurettes, deleted scenes, interviews: all of these are recognised as extras belonging to the title they sit next to, whether they’re in an Extras folder or just have the word in the filename. They stop showing up as separate films.

Then the Play button decides everything

A tidy catalogue can’t help if the file stutters, or if the audio track you bought the disc for gets replaced on the way to the screen. The native clients will play what the file actually contains, which is where the difference between direct play and a transcode becomes visible. Subtitles follow the same rule: if a file has none, Quven can write one from the dialogue, and a bitmap subtitle is read into text so it doesn’t have to be burned into the picture.

The starting point is the one it has always been. Open the folders, decide what to keep, and let the server keep track of where things are.

Frequently asked questions

Do I need to rename my files before scanning?

No. Quven reads what’s there, and when something’s wrong you correct the catalogue entry, so a folder named the way you named it in 2014 can stay that way. Consistent naming does help the matcher with obscure titles, and the organisation guide shows what helps most.

What happens to duplicates of the same film?

They stay under one title as separate editions, each named, so at play time you can tell the 4K remux from the 1080p encode. Nothing gets deleted for you, and picking one edition doesn’t hide the other.

Will a scan move or rename anything on disk?

No. A scan only reads. Corrections, artwork and watch state are stored beside the library, which is why they survive a rescan and can be undone without touching a file.

How long does the first scan take?

It depends on how big the library is and how much metadata has to be fetched. It runs in the background with live progress, so the app stays usable, and if you cancel it, it will resume from where it stopped.