https://youtube.com/watch?v=zIY17xqTzmk
windows
zip: https://github.com/hydrusnetwork/hydrus/releases/download/v529/Hydrus.Network.529.-.Windows.-.Extract.only.zip
exe: https://github.com/hydrusnetwork/hydrus/releases/download/v529/Hydrus.Network.529.-.Windows.-.Installer.exe
macOS
app: https://github.com/hydrusnetwork/hydrus/releases/download/v529/Hydrus.Network.529.-.macOS.-.App.dmg
linux
tar.gz: https://github.com/hydrusnetwork/hydrus/releases/download/v529/Hydrus.Network.529.-.Linux.-.Executable.tar.gz

NOTICE! For everyone but macOS, Version 527 had special update instructions. If you are updating from 526 or earlier, please consult the post here: https://github.com/hydrusnetwork/hydrus/releases/tag/v527 

I had a good week. There's a new quick lookup system that lets you search for a file's source without importing it.

Full changelog: https://hydrusnetwork.github.io/hydrus/changelog.html

sauce

Every now and then, I am looking at a file outside of hydrus and I can't remember who the artist/character is, or I want to know if I have the artist subbed. When it isn't something the common online source platforms support, I usually download the file, import it to my client, and then do a 'open->similar looking files' on it to see everything I have that looks like it to get more info. I'm basically doing SauceNAO but on my own client. I wanted a way to do this faster.

So, this week, check out the renamed 'system:similar files' predicate on a fresh search page. It now has two tabs. The latter is the normal 'system:similar to' that takes file hashes, if you need to do some manual lookup between imported files, but the first panel is now essentially a paste button. If you copy image data or a file path to your clipboard and paste it there, it'll calculate the similar files info and embed it into the search predicate, letting you search for anything that looks similar to what you have in your clipboard. Give it a go!

I also added a search cache to the main 'similar files' search system. Repeat searches should be massively faster, and searches you re-run with increased distance, like 0 to 4 to 8, and the background similar files search should also accelerate naturally as the cache populates. My 10k file test client's similar files search sped up about 3-4x! I'm not sure what a million+ file client will do, but let me know what you see.