https://youtube.com/watch?v=Ka4pfP2z8iA
windows
zip: https://github.com/hydrusnetwork/hydrus/releases/download/v589/Hydrus.Network.589.-.Windows.-.Extract.only.zip
exe: https://github.com/hydrusnetwork/hydrus/releases/download/v589/Hydrus.Network.589.-.Windows.-.Installer.exe
macOS
app: https://github.com/hydrusnetwork/hydrus/releases/download/v589/Hydrus.Network.589.-.macOS.-.App.dmg
linux
tar.zst: https://github.com/hydrusnetwork/hydrus/releases/download/v589/Hydrus.Network.589.-.Linux.-.Executable.tar.zst
I had a good couple of weeks mostly cleaning code and optimising things.
Full changelog: https://hydrusnetwork.github.io/hydrus/changelog.html
highlights
If you have a lot of files, your database update may take a couple minutes this week. Some users have had unoptimised similar-files search for a while, and this fixes it.
If you have a lot of tag siblings and parents (e.g. if you sync with the PTR), I have reduced the lag around sibling/parent processing significantly. There's a cache that previously had to be regenerated on every sibling/parent change, and for the PTR that could take 1.3 seconds, now it just updates in a couple milliseconds. If I have been working with you on lag issues, and we discovered that shutting off 'normal time' sibling/parent work fixed it, try turning it back on this week and let me know how it goes. If you do prefer manual control of this, there's also a new 'sync now' menu entry under tags->sibling/parent sync that basically slams all the 'work hard' buttons and functions as a catch-all 'do all outstanding work now'.
A bunch of big lists like the 'file log' should now initialise and sort a little bit faster, particularly when you are pushing 10,000+ items.
I fixed the 'page chooser' dialog's 3x3 layout, which was accidentally un-done in a recent layout overhaul. This thing is stupid, but I'm still fond of it. There's also some new checkboxes under options>pages that govern whether 'my files' and the more advanced 'local files' will show up in the 'files' choice.
The safebooru parser should stop grabbing '?' tags, and catbox collection URLs are now parseable.
next week
I want to get some meaty work done, either chipping away at dynamic file storage or duplicate auto-resolution.