I hate stuffing large blobs of data in text file formats.

I especially hate stuffing large binary blobs of data in them.


I'm trying to store metadata tags outside a media file's storage (because I'll then be re-applying them).

When it's one file, it's fine, but when it's multiple files, then I have to do really funky things like create per file folders and things.

I still think that's ultimately better than the text files, but it get gnarly.

@emacsen @genehack back when I youtube-dl'd music I had this setup where the file name was structured data such that I applied metadata to .ogg files from their file names via a script.

used a little elisp file to drive youtube-dl to consistently output file names abiding by that file name convention.

was fairly effective considering how hacky that sounds when writing it down. no file name to db identifier mismatch, no irreversible .ogg metadata corruption from bad players.

@cadadr @emacsen Back when I actually curated my mp3 collection, I kept stuff in directories by album, with a 'meta' subdirectory with per-track metadata in YAML files, and I had a Perl script that would update the id3 tags (including album art thumbnails).

i should get back in the habit of curating my own mp3s...

sample YAML file:


@genehack @cadadr

That's nearly identical to what I've decided to do, use a structured directory.

The tool I'm writing could certainly also use it to curate a private collection.

· · Web · 0 · 0 · 2
Sign in to participate in the conversation

The social network of the future: No ads, no corporate surveillance, ethical design, and decentralization! Own your data with Mastodon!