from Mountaineer@aussie.zone to jellyfin@lemmy.ml on 27 May 2024 00:42
https://aussie.zone/post/10159628
Forum Post: …jellyfin.org/t-new-jellyfin-server-web-release-1…
Changelog:
Extract media attachment one by one if the filename appears to be a path [PR #11812], by @gnattu
Filter invalid IPs on external interface matching [PR #11766], by @gnattu
Use SharedStream for LiveTV more restrictively [PR #11805], by @gnattu
Fix the IOSurf error in QSV transcoding [PR #11830], by @nyanmisaka
Improve reliability of HasChanged check [PR #11792], by @Shadowghost
Trickplay: kill ffmpeg when task is cancelled [PR #11790], by @NotSaifA
Force more compatible transcoding profile for LiveTV [PR #11801], by @gnattu
Exclude virtual items from DateLastMediaAdded calculation [PR #11804], by @Shadowghost
Add Canceled to ended state [PR #11808], by @Shadowghost
Recalculate trickplay image height for anamorphic videos [PR #11798], by @gnattu
Fix BD/DVD folder chapter image extraction [PR #11754], by @Shadowghost
Mark SearchHint.MatchedTerm as nullable [PR #11802], by @crobibero
Disable VA-VK interop on not supported kernel versions [PR #11799], by @nyanmisaka
Retain order blu-ray segments [PR #11781], by @Bond-009
Override too small trickplay image interval [PR #11788], by @gnattu
Apply audio boost when downmixing regardless of downmixalgo [PR #11774], by @Bond-009
Fix VideoToolbox H264 constrained profile option [PR #11713], by @gnattu
Do not run trickplay on scan if disabled [PR #11739], by @Shadowghost
Don’t require user when getting current session [PR #11738], by @crobibero
threaded - newest
.
Yep, shoutout to the contributors, they are certainly not dragging their feet on all these bugfixes.
Hopefully they will attempt to fix the sqlite db locking issues instead of chastising users for having them on network storage. Last version before 10.9.0 continues to work fine.
Locking issues? News to me! I have a problem with the database migration pausing during the “[INF] Applying migration ‘MigrateRatingLevels’” step and while googling the issue, I haven’t seen enough chastising, myself.
As a precaution, I changed my CIFS mount to NFS to no avail. I’m on the cusp of doing all the necessary prep-work to officially submit an issue to GitHub.
Ah. I felt like a x.x.3 version was long enough to wait for things to be shaken out, and had decided to update to 10.9.x, but I might leave it for a little bit.
I ended up back on 10.8.13 until the sqlite issues are resolved.
It’s how agile development is meant to work. The faster you get feedback on a feature, the faster you can improve on it while it’s still green in your head (and the faster you can drop it and put your efforts elsewhere if it turns out to be a dud).
A fully written feature sat there doing nothing is offering you no value.
My limited experience with Agile is being forced to share the stage with half-hour soliloquies every morning**, so as long as the dev team doesn’t have to deal with poorly-managed scrums, I’m all for it.
** I made a failed attempt at reminding everyone that it’s called a ‘standing meeting’ partly because we’re supposed to stand for its entirety. If the average person is overcome by the urge to sit down, then the _weak_ly-chaired meeting has been going on for way too long.
Edit: Instead of chair, I meant ‘scrum master’
I’ve noticed after updates I need to clear cache for that site on firefox, but beyond that the upgrades have been painless :D