>>338
Hmm, yeah, you can parse tag info from URL string easily at the moment. You could probably wangle a custom parser to pull this, but it would be impractical for real use. Is there any chance the title/page info is available in a web page the file is in, like you would see in a gallery site like Hentai Foundry, or do you only have access to a raw jpg file URL? Normally, hydrus likes to grab that data from json or html.

Converting tags to ratings, or ratings to tags, or now tags from URLs, any sort of cross-metadata-type conversion, is not really supported right now. It is complicated and requires a bunch of new UI. I am not opposed to the idea, but I don't think I could fit it into normal weekly work.

I will soonish be adding the ability to easily export and import URLs using neighbouring .txt files. It is possible you could cobble something together here by exporting the files you care about here with their URLs in a txt file, using a script to convert the urls.txt to a neighbouring tags.txt file, and then importing those tag .txts and the files again and pulling the tags. Is that something you might be interested in, or is it a bit too inconvenient?