Use local files as a source

[jmbannon]

Add ability to update file names and metadata if the config or metadata changes

[jmbannon]

This would need each entry to store its info.json file somewhere and add it to the archive. In addition, ytdl-sub needs to be able to differentiate which files it saves, like whether it's truly the video file, NFO file, thumbnail, etc.

Might have to only support a subset of features, like

• file renaming
• metadata editing

And have an entirely different command from dl or sub, maybe update

[Code-Slave]

update for nfo would be great. as i screwed up and forgot plot metadata

[jmbannon]

@Code-Slave

would it make sense to download the metada as a json via the built in ytdlp support then parse it offline? that way if someone changes what metadata they want to associate it easier to parse json local instead of online calls again?

I think it makes sense to add a field called output_options.info_json_path where it saves that metadata file (we already download it behind the scenes).

Having that would allow us to change file names, NFOs, music tags.

We wouldn't be able to get subtitles or changes to the title/description/etc since that requires calling yt-dlp again. Maybe it makes sense to have two different types of updating: online and offline.

[jmbannon]

This would require a massive amount of work though.

• Need to add the info json file
• The download archive needs to be refactored to know which files are the video/nfo/metadata/etc files
  • Would we update existing download archives or maintain two separate versions?
• A new paradigm to modify existing files using the info json

It's probably an entire release worth's of work. I don't think it's that urgent of a change since there's a simple workaround - redownload everything. Would still be nice to have.

[Code-Slave]

Its only an issue for hoarders like me. I have 8-10 tb of channels to archive. (Mostly instructional things like diy, woodworking, tractor maintenance etc etc) and some that just arent available anymore. From an archive perspective having the json is ideal. Even better if the tools can modify the nfo etc from it

[jmbannon]

Totally understand. I'll definitely add output_options.info_json_path, that would at least allow people to write scripts to update things.

[Code-Slave]

for sure. can be a long term goal. The json will help

[funar]

Its only an issue for hoarders like me. I have 8-10 tb of channels to archive.

Hoarders unite! ;)

The JSON will definitely help.

[jmbannon]

https://github.com/jmbannon/ytdl-sub/issues/172

[jmbannon]

Decided local files as source would be too cumbersome for renaming since it'd require creating a new subscription for every subscription you want to modify. Instead, proposing a reformat command, tracked here: https://github.com/jmbannon/ytdl-sub/issues/536