* refactor(settings): promote Normalization and Track transitions to own Audio categories
Pull Normalization and Track transitions out of the combined Playback
section into their own top-level SettingsSubSection categories, placed
directly under Audio Output Device. Both follow the established reusable
pattern (SettingsSubSection header + title-less SettingsGroup inside a
single-group settings-card, so the frame-collapse CSS applies).
- New TrackTransitionsBlock extracted from PlaybackBehaviorBlock; the
latter now holds only the Queue behaviour toggle (slated to move to
Personalisation).
- NormalizationBlock's SettingsGroup is now title-less; the section
header and description name it.
- Split the single audio search-index row into three (Normalization /
Track transitions / Playback) so crossfade/replaygain/lufs keywords
focus the right section.
* refactor(settings): consolidate Queue Settings under Personalisation, drop Audio Playback section
Combine Queue Display Mode and Queue Toolbar under one 'Queue Settings'
category in the Personalisation tab, and move the Queue behaviour toggle
(preservePlayNextOrder) there from Audio. The now-empty Audio Playback
section is removed.
- New 'Queue Settings' SettingsSubSection holds three titled groups:
Queue Display Mode, Queue behaviour, and (advanced-only) Queue Toolbar.
The toolbar group keeps its reset, now via a new optional 'action' slot
on SettingsGroup (+ .settings-group-title-action CSS).
- Delete PlaybackBehaviorBlock; its single toggle is inlined.
- Search index: drop the audio 'playbackTitle' row; the personalisation
queue row now points at 'queueSettingsTitle' with merged keywords.
- i18n: add settings.queueSettingsTitle to all 9 locales.
* fix(settings): align queue-toolbar separator label and restore Advanced badge
- QueueToolbarCustomizer: the separator row rendered a 1px rule where other
rows have a 16px icon, so its label sat shifted left. Reserve the full
16px icon column (1px rule centred) so the label lines up. (Preexisting.)
- SettingsGroup gains an optional 'advanced' flag rendering the Advanced
badge; badge + action now share a right-aligned title-end slot
(.settings-group-title-end), so the badge sits just left of the reset
button. Restores the Advanced indicator the Queue Toolbar lost when it
moved from a SettingsSubSection into a group.
* fix(audio): hide the output-device category on macOS instead of showing a notice
Playback is pinned to the system default on macOS, so the picker showed a
notice explaining it does nothing there. Gate the whole Audio Output Device
category out on macOS (`!IS_MACOS` in AudioTab) and drop the now-dead notice
branch + the `audioOutputDeviceMacNotice` string from all 9 locales. The
device-probe hook already short-circuits on macOS, so no work is wasted.
* refactor(settings): box the sidebar customizer groups consistently
The sidebar display toggles sat as a bare div and looked unfinished. Box
them in a SettingsGroup, with the nav-item drag list in a second group.
Render the groups directly in the sub-section content (no settings-card
wrapper) — matching the other Personalisation customizers, which use bare
SettingsGroups. Also drop the settings-card around Queue Settings for the
same reason, so every Personalisation section's boxes share one width and
inset instead of the card-wrapped ones sitting narrower/indented.
* docs(settings): changelog, what's new, and credits for the settings reorg (#1130)
Fold the Audio/Personalisation reorganization into the existing
'Settings — consistent grouped layout' changelog entry (now #1126 + #1130)
and the matching What's New highlight, and add a consolidated settings-
refactor line to the contributor credits.
* refactor(settings): extract reusable SettingsGroup component
Pull the boxed sub-section pattern (bordered panel + accent uppercase
header) introduced for the Audio tab into a reusable <SettingsGroup
title desc> component, and migrate NormalizationBlock and
PlaybackBehaviorBlock onto it. No visual change.
* feat(settings): box the System behavior section into groups
Split the System -> App Behavior card (which bundled tray toggles, Linux
rendering tweaks and the clock format) into titled SettingsGroup panels
(Tray / Linux rendering / Clock) for a clearer, consistent boxed look.
Adds the group titles across all 9 locales.
* feat(settings): box Appearance visual options; optional SettingsGroup title
Split the Appearance -> Visual options card into a Display group and a
Window group (Linux custom titlebar controls). SettingsGroup now allows an
optional/omitted title for plain boxed panels. Group titles added across
all 9 locales.
* feat(settings): box Discord cover source and templates
Group the Discord cover-source toggles and the activity templates into
separate SettingsGroup panels (templates reuse their existing title/desc),
dropping the manual indent and inline header. No new strings.
* feat(settings): box the Music Network section
Give the scrobble-master toggle, the enrichment-primary picker and the
provider list each a boxed SettingsGroup with an accent header, and switch
the add-a-service provider rows to the boxed panel style for consistent
contrast with the rest of settings.
* feat(settings): titled boxes for Integrations sections + cover source label
Wrap the Discord enable toggle, the Discord cover-source picker (now with a
"Cover art source" title and explainer), Bandsintown and Show-in-Now-Playing
in titled SettingsGroup panels so each single-item section gets the same
accent-headed boxed look. New cover-source strings across all 9 locales.
* feat(settings): box Lyrics tab sections
Wrap the lyrics-sources customizer and the sidebar-style picker in titled
SettingsGroup panels (reusing the existing section titles).
* refactor(settings): consistent content inset in SettingsGroup
Add a body wrapper with a small left inset so every boxed section indents its
controls uniformly (title stays flush at the box edge). Defined once in the
component instead of per section.
* feat(settings): box Storage tab sections incl. cover art cache
Wrap media directory, cover art cache strategy, next-track buffering and
downloads in titled SettingsGroup panels (reusing existing section titles);
align the cover-cache table's first column flush with the new content inset.
* feat(settings): box Library tab, relocate Lucky Mix, add SettingsToggle
- Add a reusable SettingsToggle component for the repeated label/desc/switch row.
- Box the Random Mix blacklist (grouping built-in keywords with the custom
filter), Ratings and the Analytics strategy section.
- Move the "Show Lucky Mix in menu" toggle out of the blacklist and up next to
split-mix navigation and now-playing-at-top in the sidebar customizer.
* feat(settings): box remaining Appearance sections
Wrap library grid, UI scale, font and seekbar style in titled SettingsGroup
panels (reusing the existing section titles).
* feat(settings): box System language, logging and backup sections
Wrap language and logging in titled SettingsGroup panels, and rework the
backup section to drop its redundant inline header (the subsection already
provides it) and box its content. About / Contributors / Licenses stay as
free display content.
* feat(settings): box remaining Audio sections
Wrap Hi-Res, equalizer, playback speed, audio output device and track
previews in titled SettingsGroup panels (reusing the existing section titles).
* feat(settings): box Input tab keybinding sections
Wrap the in-app and global shortcut lists in titled SettingsGroup panels.
* feat(settings): box the Personalisation tab
Convert every customizer (sidebar, home, artist, queue toolbar, playlist,
player bar) to plain inner containers and wrap each in a titled SettingsGroup,
so the boxed look is consistent without nested cards. Box the queue display
mode picker too.
* feat(settings): box the Themes sections above the store
Add a boxed option to the flat Themes sections and apply it to Your Themes,
Auto-Switch Theme and Import (the theme store stays unboxed). Drop the inner
settings-card from InstalledThemes, ThemeImportSection and the scheduler so
the SettingsGroup is the only frame.
* fix(settings): give the Themes boxes a card carrier and accent header
The flat Themes sections had no settings-card behind them, so the bg-app
SettingsGroup had no contrast against the page. Box each above-store section
as settings-card > titled SettingsGroup (accent header) like the rest.
* fix(settings): clean up the server library sync status line
Collapse the offline status to a single icon + phrase ("Server offline —
sync deferred") instead of a redundant "Deferred" badge alongside it, and
drop the now-unused string across all 9 locales.
* fix(queue): keep toolbar toggle buttons coloured active while hovered
The generic .queue-round-btn:hover rule outweighed .active in specificity, so
an active button kept the hover colour until the pointer left — the toggle
only looked on after moving the mouse away. Add an explicit active-hover rule
(mirroring the mini player) so active wins immediately on click.
* fix(settings): fix Lyrics triple title, drop dead Discord cover keys
- Remove LyricsSourcesCustomizer's own section header (the subsection already
titles it) and make its inner cards plain; the Lyrics boxes are now title-less.
- Drop the dead discordCoverSource/discordCoverSourceDesc i18n keys (superseded
by discordCoverTitle/Desc) and un-jam the libraryIndexServer status line in
all 9 locales.
* refactor(settings): drop redundant group titles and double frames
Single-group sections duplicated the subsection title in the group header — make
those groups title-less so the subsection header is the only title. A card whose
sole child is one group now collapses via CSS so single-group sections render a
single frame (matching the card-less Lyrics layout) instead of a card-in-group
double border. Multi-group cards keep their frame as the grouping container.
* fix(settings): keep the flat Themes cards framed when collapsing single-group cards
The single-group card collapse also flattened the Themes section cards, which —
not being inside an accordion — are themselves the contrast surface. Exclude
.themes-section descendants from the collapse.
* refactor(settings): use SettingsToggle for Appearance and System toggle rows
Replace the hand-rolled toggle-row markup in the Appearance visual options and
System tray/Linux/changelog rows with the shared SettingsToggle component, and
extend it with a searchText prop and ReactNode descriptions for the remaining
call sites.
* refactor(settings): roll SettingsToggle out to Integrations, Audio and Storage
Replace hand-rolled toggle rows in the Discord/Bandsintown/Now-Playing,
Hi-Res, hot cache and playback behaviour/rate sections with SettingsToggle.
Make its label optional (desc-only rows whose title is the group header) and
add searchText/id pass-throughs.
* refactor(settings): use SettingsToggle for the remaining standard toggle rows
Track previews master, sidebar lyrics style, YouLyPlus/static-only, queue
display mode and the sidebar split-nav/now-playing/lucky-mix rows now use the
shared component. Genuinely custom rows (drag lists, the skip-star threshold,
the AudioMuse row with its inline link, the fine-step advanced badge) stay
hand-rolled.
* fix(settings): drop duplicate titles on Hi-Res/Bandsintown/Now-Playing, un-jam locale line
These single-toggle sections kept a group title identical to their subsection
header; once the single-group card collapses the title rendered twice stacked.
Make them title-less (the subsection header is the label). Also split the
discordCoverTitle locale entry onto its own line in all 9 locales.
* feat(settings): restore section icons in boxed Themes headers
SettingsGroup gained an optional accent icon slot rendered before the
title; ThemesTab forwards the Palette/Clock/Upload icons the boxed
sections lost when they moved off the bespoke <h2> header.
* style(settings): re-indent SettingsGroup children to nesting level
Pure whitespace: children wrapped in SettingsGroup were left at their
pre-wrap indent. No content change (git diff -w is empty).
* fix(settings): clarify Native Hi-Res Playback description
The old copy led with the disabled-state behaviour ("forces 44.1 kHz"),
which read as if the toggle itself locked output to 44.1 kHz. Describe
what enabling actually does: play each track at its native sample rate
(reconfiguring the output device to match) instead of resampling to
44.1 kHz. Updated across all 9 locales.
* docs: changelog + what's-new for the settings layout refactor (#1126)
* feat(playback): add transition-mode helper
Centralise the crossfade/AutoDJ/gapless mutual exclusivity in one place
instead of the scattered setter combinations across the toolbar, mini
player and settings. AutoDJ stays encoded as crossfade + trim-silence, so
the persisted flags and the audio engine are unchanged.
* feat(queue): split AutoDJ into its own toolbar button + playlist submenu
- AutoDJ becomes a standalone toolbar button (Blend icon) next to
crossfade, driven by the shared transition-mode helper. The crossfade
right-click popover drops the mode switch and keeps only the seconds
slider.
- Save + load playlist collapse into one Playlist button opening a small
submenu, freeing up toolbar space.
- queueToolbarStore gains a position-preserving rehydrate migration
(legacy save/load -> playlist, autodj inserted after crossfade) with
unit tests.
- Toolbar customizer and all 9 queue locales updated to match.
* feat(mini-player): standalone AutoDJ button, shared transition helper
Mirror the queue toolbar: AutoDJ gets its own Blend button, the crossfade
popover keeps only the seconds slider. The mini player now drives all
three transitions through a single additive `mini:set-transition-mode`
event handled by the shared helper, replacing the per-flag mini events.
Drop the now-dead crossfade-mode CSS.
* feat(settings): segmented track-transition picker, regroup playback
Replace the crossfade toggle + inner crossfade/AutoDJ switch with a single
Off | Gapless | Crossfade | AutoDJ segmented control (mirroring the
Normalization picker above), driven by the shared transition helper — the
mutual exclusivity now reads at a glance. Crossfade keeps its seconds
slider; AutoDJ shows its content-driven explainer. The block is regrouped
under "Track transitions" and "Queue behaviour" headings. Drops the now
unused crossfade/gapless description and not-available i18n keys across all
9 locales and adds the new transition strings.
* fix(queue): open the playlist submenu inward
The playlist submenu inherited the crossfade popover's right:0 anchor and,
sitting on the left of the toolbar, opened out under the main container.
Anchor it left:0 so it stays inside the queue panel. Update the toolbar
test for the new playlist submenu (save/load moved off the toolbar).
* feat(settings): box playback sub-sections into panels
Wrap Normalization, Track transitions and Queue behaviour each in their own
bordered panel with an accent uppercase header (new reusable .settings-group
classes), so the sections read as distinct blocks instead of one wall of
text. Drops the thin divider that separated them.
* docs: changelog, credits and what's new for AutoDJ standalone
Fold the standalone-AutoDJ changes into the existing AutoDJ entry in the
changelog and the in-app What's New, and add the credit (#1124).
* feat(crossfade): add "trim silence between tracks" toggle
New persisted setting `crossfadeTrimSilence` (default off; existing
installs rehydrate off via the persist default-merge). Surfaced in
Settings -> Audio and in the crossfade popovers of the queue toolbar
and the mini-player.
Crossfade buttons now separate the two actions: left-click toggles
crossfade on/off, right-click opens the settings popover (seconds +
trim). Shared the mini popover positioning into useMiniAnchoredPopover
(now backing both volume and crossfade). Mini bridge carries
crossfadeSecs/crossfadeTrimSilence and gains mini:set-crossfade-secs /
mini:set-crossfade-trim-silence.
The actual silence-trimming playback behaviour is wired in a follow-up;
this commit only persists the user intent. i18n added across 9 locales.
* feat(crossfade): trim silence between tracks (waveform-driven)
Wire the actual silence-aware crossfade behind the (default-off)
crossfadeTrimSilence toggle. Detection is derived on the fly from the
cached 500-bin waveform + track duration — no new analysis pass or
cache fields.
- waveformSilence.ts: computeWaveformSilence(bins, duration) → lead/trail
silence + content bounds, using the peak curve, a low absolute cut and
a per-side cap. Unit-tested.
- A-tail (JS): handleAudioProgress advances the crossfade early, at
contentEnd - crossfadeSecs, when the current track ends in real
trailing silence, so the fade overlaps music. Guarded once per play
generation.
- B-head: audio_play gains an additive optional start_secs; the freshly
built source is try_seek'd past the next track's leading silence before
append, then seek_offset/samples_played are re-anchored so position is
content-relative. Non-seekable / cold sources degrade to today.
- Pre-buffer: crossfade next-track download + B-head probe moved to
crossfadePreload.ts with a fixed ~30 s budget before the track needs to
play (widened by trailing silence so the early advance keeps the
budget). Also fired right after a seek into the window so jumping near
the end still buffers in time.
Checks: tsc, vitest (store suite + new units), cargo test/clippy for
psysonic-audio.
* feat(crossfade): recommend hot cache for trim; probe B-head regardless
Add a "for reliable results, enable the Hot playback cache" note to the
trim-silence toggle description across all 9 locales — hot cache keeps
the next track on disk so it starts instantly past its lead silence.
Fix: the leading-silence probe (B-head) now runs even when hot cache is
on; only the redundant byte pre-download is gated on !hotCache. Without
this, enabling hot cache (the recommended setting) would have skipped the
probe and disabled leading-silence trimming.
* feat(crossfade): content-driven smart crossfade overlap
Smart crossfade derives the per-transition overlap purely from the
waveform envelopes — max(A outro fade, B intro rise) clamped 0.5–12s —
instead of the fixed crossfadeSecs ("work by fact"). The JS early
advance arms the computed overlap and audio_play applies it through a
new crossfade_secs_override (capping only this swap's fade); plain
loud→loud endings fall back to the engine crossfade at crossfadeSecs.
* feat(crossfade): "Crossfade | Smart crossfade" mode switch in UI
Replace the standalone "trim silence" toggle with a Crossfade / Smart
crossfade segmented control in settings and both crossfade popovers
(queue toolbar + mini-player). Classic Crossfade shows the seconds
slider; Smart crossfade is content-driven with no duration to set.
Adds smartCrossfade / smartCrossfadeDesc strings to all nine locales
and the "smart crossfade" search keyword.
* feat(crossfade): don't double-fade a track that already fades out
Decouple the outgoing track's fade-out from the incoming fade-in. When
A carries its own recorded fade-out (outroFadeA ≥ 1s and ≥ B's intro
rise), planCrossfadeTransition now sets outgoingFadeSec = 0; the engine
then skips A's TriggeredFadeOut so A keeps full gain and its recording
carries it down while B rises underneath — no more double attenuation
that made A vanish early and B blare in. Hard-cut endings still get an
engine fade over the overlap.
audio_play gains outgoing_fade_secs_override (Some(0) = ride A's own
fade), threaded via SinkSwapInputs.outgoing_fade_secs; the JS advance
arms it alongside the overlap.
* feat(crossfade): rename the smart crossfade mode to "AutoDJ"
User-facing rename of the content-driven crossfade mode from "Smart
crossfade" to "AutoDJ" across the settings segmented switch, the queue
and mini-player popovers, all nine locales, and the settings search
keywords. The underlying store flag (crossfadeTrimSilence) is unchanged.
* feat(crossfade): standard ~2s blend for hard loud→loud meetings
When AutoDJ trims a track's protective trailing silence and the loud
ending butts straight into a loud intro, neither edge fades, so the old
0.5s anti-click floor sounded like an abrupt cut. Use a standard ~2s
equal-power crossfade for that case (both edges analysed, nothing
fades); real fade-outs/buildups keep their longer content-driven span,
and the bare floor only survives when an envelope is missing.
* fix(crossfade): keep B's fade-in across the B-head start-offset seek
EqualPowerFadeIn::try_seek jumped straight to unity gain for any seek
≥100ms, which also hit the initial start-offset seek that skips the
incoming track's leading silence — so a crossfaded track with trimmed
lead silence popped in at full gain instead of fading in. Only skip the
fade-in for mid-playback seeks (sample_count > 0); a seek before any
audio has played keeps the fade-in.
* feat(crossfade): gate AutoDJ early fade on next-track readiness
The early, content-driven advance now fires only when the next track's
audio is actually available — in the engine RAM preload slot
(enginePreloadedTrackId) or local on disk (offline library, favourite-auto
or hot-cache ephemeral) — via isCrossfadeNextReady(). Analysis alone is
not enough: a cold, still-buffering stream would fade in over silence.
When B isn't ready the gen guard stays unset so it re-checks on later
ticks; if B never readies, the plain engine crossfade handles the
transition (graceful degrade) instead of a broken fade. A RAM preload
copy suffices — the full track need not be cached to disk.
* feat(crossfade): suppress engine auto-crossfade for AutoDJ + eager preload
With the hot cache off the readiness gate alone wasn't enough: the engine's
progress task autonomously fires its crossfade audio:ended ~crossfadeSecs
before the end, independently of JS, and would start a still-buffering next
track and fade over it — an audible jump.
- Engine: add autodj_suppress_autocrossfade flag (audio_set_autodj_suppress);
the progress task treats it like crossfade-off, so the early timer never
fires and audio:ended only comes from real source exhaustion / watchdog.
- JS drives the transition: set the flag when a content fade is pending
(wantEarly) and clear it for plain loud→loud / non-AutoDJ, so the normal
engine crossfade is preserved there. When the next track never readies, A
plays out and we degrade to a clean sequential start instead of a jump.
- audio_preload gains an `eager` flag; the crossfade/AutoDJ pre-buffer passes
it to skip the 8s start throttle so the RAM slot fills before the fade.
* docs(changelog): AutoDJ content-aware crossfade (PR #1122)
Add the 1.49.0 Added entry and the cucadmuh credits line for AutoDJ.
* feat(albums): "Artist → Year" sort option
Adds a third album browse sort that groups albums by artist and orders each
artist's albums chronologically (oldest first, title as a same-year tiebreak)
— the double-sort requested in #1113. The local index sorts globally via
[{artist},{year},{name}]; the server fallback fetches by artist and applies the
year ordering per page (Subsonic has no compound sort).
* i18n(albums): Artist → Year sort label (9 locales)
* docs(changelog): add Artist → Year to the album sorting entry (#1120)
* feat(playlists): playlist folder model — store + pure grouping core
Local, per-server folder layer over the server's flat playlist list (the
Subsonic API has no folder concept). Adds:
- playlistFolders.ts: shared types + pure groupPlaylistsByFolder (used by
every surface), with full unit coverage.
- playlistFolderStore.ts: persisted Zustand store (create/rename/delete/
assign/collapse), per-server scoped; deleting a folder drops its
assignments so playlists fall back to ungrouped.
UI surfaces (sidebar + Playlists page) build on this in following commits.
* i18n(playlists): folder strings across all 9 locales
Adds the playlists.folders.* namespace (folder names, move/remove, expand/
collapse, group-by-folders toggle, count plurals, and the local-only notice
explaining that Navidrome and the Subsonic API have no native folder support).
* feat(playlists): folder views, drag-to-folder, move-to-folder menu, view toggle
Surfaces the local folder layer on both the Playlists page and the sidebar:
- Page: collapsible folder sections + ungrouped remainder, each reusing
VirtualCardGrid; flat grid is kept verbatim when no folders exist or the
group view is toggled off.
- Drag-to-folder via the shared mouse-based psy-drop system (HTML5 DnD is
unusable in WebKitGTK); the whole section is the drop zone, and the
ungrouped zone appears during a drag so a playlist can always return to root.
- "Move to folder" submenu in the playlist context menu (keyboard-accessible
path; also creates folders on the fly) — stays available offline.
- Header gets a "New folder" action and a "Group by folders" view toggle
(both context-aware); a notice surfaces the local-only caveat.
- Sidebar renders the same collapsible folder groups.
- groupView preference added to the folder store.
* docs(changelog): note playlist folders (#1119)
* fix(player): player bar context menu acts on the current song, not its album
Right-clicking the current track in the player bar built an album object
from the playing track and opened the album context menu, so "Add to
playlist" added the whole album instead of the song. It now opens a
song-scoped menu for the current track. Left-click on the title still
navigates to the album.
* docs(changelog): note player bar add-to-playlist fix (#1117)
* fix(albums): order each artist's albums by title when sorting by artist
Browsing albums by artist left the albums within each artist in an
undefined order. The local-index sort only emitted the artist key; it
now appends album title as a secondary key (and artist as the tiebreak
for the by-name sort), matching the network path's per-page ordering.
* docs(changelog): note album sort-within-artist fix (#1115)
* fix(windows): restore taskbar thumbnail media buttons after deferred window show
The Prev / Play-Pause / Next buttons in the Windows taskbar thumbnail
preview stopped appearing. The taskbar code was unchanged; the regression
came from the main window now starting hidden with a deferred show. ThumbBarAddButtons was still called at setup time, before the shell had
created the window's taskbar button, so it returned S_OK but added
nothing (no error logged).
Register the shell's "TaskbarButtonCreated" message and add the buttons
from the window subclass when it fires (first show, and again after an
explorer restart), which is the documented requirement for ThumbBarAddButtons.
* docs(changelog): note Windows taskbar media buttons fix (#1112)
power_notify_win.rs imports windows::Win32::System::Power and
UI::WindowsAndMessaging, but the crate only declared Foundation/Com/Threading.
It compiled solely via workspace feature unification (the root src-tauri crate
enables them), so building or testing psysonic-audio in isolation on Windows
(cargo check/test -p psysonic-audio) failed with E0432 unresolved imports.
Declare the two features the crate actually uses so it builds standalone.
No behaviour change; workspace/release builds already had these features.
* fix(audio): make seeking work on streamed Opus/Ogg via on-demand HTTP Range
Seeking inside an Opus/Ogg track streamed over ranged HTTP was a contained
no-op (the seekbar snapped back); it only worked once the track had fully
downloaded/cached. symphonia 0.6's Ogg demuxer seeks by bisecting the byte
range (reading pages at midpoints across the whole file) and scans the last
pages during the probe, but RangedHttpSource only filled the buffer linearly
from offset 0, so any read ahead of the download front blocked until the
linear download caught up. Keeping Ogg seekable through the probe without a
real random-access source would have forced a full pre-download.
Add an on-demand random-access fetcher to RangedHttpSource: when a read
lands well ahead of the contiguous linear download (a seek, a bisection
midpoint, or the end-of-stream probe), fetch the needed range over HTTP Range
(1 MiB window) on the tokio runtime and let the read loop poll for it, instead
of blocking on the linear filler. Ranged Ogg now stays seekable through the
probe (records its byte range) so seeking works for real; the catch_unwind in
try_seek stays as a safety net.
- New OnDemand fetcher writes arbitrary ranges into the shared buffer (same
bytes the linear download would write; idempotent under the buffer mutex,
mirroring the existing MP4 moov tail-prefetch). It never touches
downloaded_to/done, so full-download completion and the track-analysis seed
are unaffected.
- On-demand only fires on a forward gap > 512 KiB, so normal sequential
read-ahead (and a slightly starved play cursor) still waits for the linear
download without spurious range requests.
- ranged-stream now passes random_access=true; preview keeps on_demand=None.
Does not touch the Tauri boundary (no invoke/event changes).
* docs(changelog): add 1.49.0 entry for streamed Opus/Ogg seeking (#1110)
Also credit the streamed-seek work in settingsCredits.
* fix(audio): require 206 for ranged Range fetches at a non-zero offset
Address PR #1110 review note: ranged_write_http_range accepted a 200 the
same as a 206. A server that ignored the Range header and replied 200 returns
the whole body from byte 0; writing that at a non-zero offset would corrupt
the buffer (affects both the on-demand seek fetcher and the MP4 moov-tail
prefetch). Accept 200 only when the request started at offset 0; otherwise
require 206.
Backport of the 1.48.1 hotfix notes onto main (which is on 1.49.0-dev).
Insert the released [1.48.1] section between [1.49.0] and [1.48.0] in
CHANGELOG.md (all eight Fixed entries with their attribution, identical to the
fix/1.48.1 branch) and the matching [1.48.1] What's New section. Application
version is intentionally left at 1.49.0-dev — only the notes are carried over.
Windows SMTC could not render our cached WebP album covers: souvlaki loads the
file and SetThumbnail/set_metadata succeed, but the lock screen and Quick
Settings media tile showed a blank cover, because the OS thumbnail decoder does
not handle WebP even with the Store WebP extension installed.
Transcode local file:// WebP covers to PNG (libwebp decode then image PNG
encode, into a single reusable temp file) before handing them to the OS media
controls, gated to Windows. macOS (ImageIO) and Linux pass through unchanged.
(cherry picked from commit 76d028127d)
The Windows system media controls (Quick Settings media tile, lock screen,
third-party media flyouts) labelled playback as "Unknown application" with no
icon, because souvlaki creates the SMTC via GetForWindow and Windows resolves
the source name from the process AppUserModelID, which was never set.
Call SetCurrentProcessExplicitAppUserModelID early in run() so the process has
an explicit identity that matches the installer shortcut's AppUserModelID;
Windows then resolves the name and icon to Psysonic.
(cherry picked from commit 4fd85f2dd4)
* fix(library): use partial indexes for §6.9 remap lookup
The delta-sync remap detection ran a single lookup with an OR across
content_hash and server_path. SQLite could not use the partial
idx_track_remap_hash / idx_track_remap_path indexes for it: a partial
index is only applied when the query's WHERE provably implies the index
predicate (… != ''), and an OR spanning two columns blocks the per-branch
index plan. The query degraded to a full track scan on every incoming
row → O(rows × catalog), causing multi-minute stalls on large libraries
(observed upsert_batch_remap exec_ms=162001 on a ~200k-track Navidrome
sync, which in turn blocked all other writers on the single write mutex).
Split it into two single-column lookups that each repeat the index
predicate so the planner picks the matching partial index (SEARCH, not
SCAN); hash is checked first, matching §6.9 strong-key priority. Adds an
EXPLAIN QUERY PLAN regression test asserting index usage.
* docs(changelog): note remap-lookup sync stall fix (PR #1105)
(cherry picked from commit bca0acbaff)
* fix(audio): fix Opus/Ogg seek crash by keeping random-access sources seekable through probe
Scrubbing the seekbar on Opus/Ogg files (then pressing Stop) crashed the whole
app. symphonia 0.6's Ogg demuxer records the physical stream's byte range only
when the source is seekable during the probe, but ProbeSeekGate hid seekability
there — so phys_byte_range_end stayed None and the first seek hit
Option::unwrap() on None on the cpal audio thread. That poisoned the engine
mutexes and aborted the process at the non-unwinding cpal FFI boundary (the
"crash on Stop" was a downstream symptom).
- Keep Ogg/Opus seekable through the probe on random-access sources (local
files, in-memory) so the demuxer computes its seek bounds and seeking works
for real. Progressive ranged-HTTP keeps the gate to avoid forcing a full
download before playback starts.
- Contain any demuxer unwind inside SizedDecoder::try_seek (catch_unwind) so a
panic on the audio thread can no longer poison engine state — covers streamed
Ogg (still gated) and any future demuxer panic.
- Thread a random_access flag through PlayInput::SeekableMedia and new_streaming.
* docs(changelog): add 1.48.1 entry for the Opus/Ogg seek crash fix
Also note the Opus seek crash fix in WHATS_NEW.md (1.48.1).
(cherry picked from commit 8bfde08199)
The macOS red close button always emitted app:force-quit, which exits
unconditionally and never checked the minimizeToTray setting — so on macOS
the window closed the app even with "Minimize to Tray" enabled. Route the
main-window close through window:close-requested on all platforms, so JS
decides hide-vs-exit from the setting (default off = unchanged quit). The
tray "Exit" item still force-quits.
Fixes#1103
(cherry picked from commit acd6f12aba)
Follow-up to the dual-address cover fix: a playback/active cover scope was
routed through getPlaybackServerId(), which returns a `string` that can be
empty or an index-key (not a profile id) for locally-cached tracks. The
`?? activeServerId` fallback never fired (empty string isn't nullish), so no
profile matched and the URL came back null — which the presence layer caches
per cover id, producing intermittent "cover shows, then doesn't".
A playback/active scope always means the active server (a cross-server track
gets an explicit `server` scope), so resolve the active profile directly.
(cherry picked from commit 8f93f30e6f)
The Discord cover URL was built via the connect endpoint, which prefers the
LAN address — but Discord fetches the image from its own servers, so a LAN
address is unreachable and the cover falls back to the app icon. This is a
dual-address regression: before a second (public) address could be added,
the only configured URL was the public one.
Build the Discord large-image URL via serverShareBaseUrl (public preferred,
like share links / Orbit invites) instead of the connect URL. Adds a test.
(cherry picked from commit 5f15784b7d)
Third defence for #1094: on a device change while paused/stopped,
reopen_output_stream stopped the old sink without bumping the engine
generation (the bump only happened on a successful internal resume). The
still-running progress task could then flip done_flag and emit a spurious
audio:ended, which the frontend turns into a restart. Bump the generation
before sink.stop() in the non-playing case so the progress task bails out;
the active-playback path keeps bumping inside try_resume_after_device_change.
(cherry picked from commit 9034882bf6)
Defence-in-depth for #1094: the device-changed/-reset handlers restarted
playback based on the UI `isPlaying` flag alone, which can be stale or
desynced at the moment of a device change. Gate the restart on the
engine-paused flag as well (`isPlaying && !getIsAudioPaused()`), so a
paused engine never auto-restarts regardless of how `isPlaying` got set;
when paused it just resets for the cold path. Adds a regression test.
(cherry picked from commit 588dd8c48d)
Toggle, Play and Pause from the OS media-control bridge (souvlaki) were
all collapsed onto the play-pause toggle event. On an audio-route change
(e.g. macOS sending an explicit Pause when headphones disconnect) this
turned the pause into a toggle, resuming paused playback on the new
output device.
Map Play and Pause to dedicated media:play / media:pause events (the
frontend handlers already exist); only a real toggle key emits
media:play-pause. Paused playback now stays paused across device changes.
Fixes#1094
(cherry picked from commit f04bfb3d35)
QueueHeader aggregated total/future duration in a useMemo keyed on queueIndex,
so every skip ran a synchronous O(n) pass over the entire queue (resolveQueueTrack
per item). On very large queues this blocked the main thread for seconds — the
UI froze on skip and on the device-switch playTrack fallback (#1072; the freeze
half of #1090). QueuePanel is always mounted, so it hit even with the queue
collapsed.
Build a cumulative-duration prefix keyed on queue/resolver-version only; the
future-tracks total is now an O(1) lookup per skip. Display output unchanged.
(cherry picked from commit 7e91a5b2a1)
* feat(themes): show theme version in the store and installed list
* feat(themes): filter the theme store by animated / static
* docs(changelog): add 1.49.0 entry for theme store version + animated filter
The rebuilt fullscreen player hardcoded Apple-style lyrics scrolling (active
line ~35% from the top) and ignored Settings -> Lyrics -> Lyrics scroll style,
which still drives the sidebar/mobile LyricsPane. FsLyricsApple now reads
sidebarLyricsStyle: 'classic' centres the active line, 'apple' keeps the 35%
anchor, matching the sidebar lyrics.
Add highlights for Live status dots, playback Semitones, Linux title bar
styles, Music Network self-hosted scrobble fix, and Navidrome radio admin
gating. Remove meta What's New page section. Re-sort CHANGELOG Fixed
entries (#1085/#1086 were at the top).
* feat(now-playing): liveness indicator dot in the listening popover
Replace the raw "Nm ago" line in the "Who is listening?" popover with a
derived presence dot (green playing / amber paused / dim idle). The presence
is computed in one tested helper that unifies the playbackReport transport
state with the legacy getNowPlaying recency, instead of formatting a raw
timestamp inline. The dot carries the localized status as an aria-label and
tooltip so it is not conveyed by colour alone.
* feat(radio): gate station create/edit/delete behind Navidrome admin role
Navidrome >= 0.62 restricts internet-radio management to admins
(GHSA-jw24-qqrj-633c); non-admin requests fail. Hide Add Station, Search
Directory, the per-card edit chip and delete button for confirmed standard
Navidrome users via a canManageNavidromeRadio() helper on the existing
useNavidromeAdminRole framework. Admins, non-Navidrome servers and transient
states stay unrestricted; playback and favourites remain available to all.
* docs(changelog): now-playing status dot + admin-gated radio (#1086)
* feat(playback): varispeed-by-semitones strategy and 2-decimal speed label
Adds a fourth playback-rate strategy "Varispeed (semitones)" — a frontend
lens over varispeed where the user dials the pitch change directly in
semitones (±12 st, 0.1 step) and speed = 2^(st/12). Engine contract is
unchanged: the store maps it to the existing "varispeed" Rust strategy via
engineStrategy(), and switching between the two varispeed lenses at the same
speed no longer restarts the track.
Also widens the speed readout to two decimals (formatSpeedLabel → toFixed(2))
so every 0.05 slider step is visible (1.05×, 1.10×, 1.15×), addressing the
feedback on issue #531 that the label looked stuck between steps.
Includes new i18n keys and updated hint across all 9 locales, plus Vitest
coverage for the new helper, label formatting, and the no-restart lens switch.
* refactor(playback): shorten Semitones strategy label, add per-strategy tooltips
Renames the fourth strategy to a concise "Semitones" (was "Varispeed
(semitones)") across all 9 locales, and lets the four strategy buttons share
the row width so they fit on one line in Settings while still wrapping in the
narrow player popover. Each strategy button now has a short hover tooltip
explaining how it is built (incl. the 2^(st/12) and 12 × log2(speed) maths).
* feat(playback): advanced fine-step precision for speed/pitch sliders
Adds an opt-in "Fine adjustment" toggle in Settings → Audio (visible only in
Advanced mode) that shrinks the playback-rate slider steps to 0.01× for speed
and 0.01 st for pitch/semitones, with the pitch readout widening to two
decimals to match. Default behaviour (0.05× / 0.1 st) is unchanged. The
preference is UI-only (persisted, not sent to the engine) and also applies to
the player-bar popover and wheel. Addresses the finer-precision request on
issue #531 without cluttering the default UX.
* docs: changelog and credits for playback speed follow-up (PR #1084)
* fix(music-network): keep API suffix for self-hosted paste-token providers
The api_key_only connect strategy persisted the raw origin from the baseUrl
field instead of the resolved API base, dropping the preset's
selfHostedApiSuffix (e.g. /apis/listenbrainz). Scrobbles and now-playing then
hit <origin>/1/submit-listens (404/405, silently unlogged) instead of
<origin>/apis/listenbrainz/1/submit-listens, so nothing was recorded.
Return the runtime-resolved ctx.baseUrl (origin + suffix) and fall back to the
field only when it is absent. Fixes Koito and both Maloja compat surfaces
(ListenBrainz and Audioscrobbler). Existing accounts must reconnect to
re-persist the corrected base.
* docs(changelog): self-hosted scrobble URL fix (#1085)
* feat(titlebar): selectable window button styles + minimize toggle
Custom title bar (Linux) gains a window-button style picker, mirroring
the seekbar style picker pattern. Six form-named styles: dots, dotsGlyph,
flat, pill, outline, glyph. All buttons now carry minimize/maximize/close
glyphs for clear, colour-blind-friendly iconography; dots reveal glyphs on
hover, dotsGlyph always shows them.
- New authStore state windowButtonStyle (default dots) + showMinimizeButton,
with rehydrate validation falling back to dots on unknown values.
- WindowButtonPreview reuses the real .titlebar-btn classes for WYSIWYG tiles.
- Picker + minimize toggle render under the Custom title bar setting, gated
on the toggle being on.
- Monochrome styles use --text-primary glyphs and stronger borders for
contrast on dark themes.
- Dev-build grey marker scoped to the real title bar so previews show true
colours.
- i18n keys in all 9 locales; setter and rehydrate tests.
* docs(changelog): window button styles (#1083)
* docs(whats-new): refresh 1.48.0 highlights for release screen
Reorder and expand user-facing copy to match CHANGELOG Added order,
add Music Network and Live, drop stale About/licenses line, replace
the vague Fixed placeholder with grouped fixes, and add a short Under
the hood section for significant non-UI work.
* docs(whats-new): order 1.48 sections by user impact
Sort Highlights, Fixed groups, and bullets within each block so the
most noticeable changes for everyday use appear first.
* docs: CHANGELOG 1.48 strict PR order; clarify WHATS_NEW sorting
Reorder Added/Changed/Fixed entries in CHANGELOG [1.48.0] by ascending PR
number. WHATS_NEW keeps impact-based order; add a maintainer note that
the two files intentionally differ.
* chore(deps): bump esbuild to 0.28.1 for GHSA-g7r4-m6w7-qqqr
Closes Dependabot alert #17 — path traversal in esbuild dev server
(--servedir) on Windows; patched in 0.28.1.
* chore(nix): sync npmDepsHash with package-lock.json
---------
Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
* fix(installer): write logs to stderr so the download URL isn't polluted
* fix(installer): read the reinstall prompt from /dev/tty
* fix(installer): add --fail --globoff to the package download
* fix(installer): degrade gracefully when no controlling terminal is available
* feat(now-playing): adopt OpenSubsonic playbackReport for live now-playing
Drive a small playback state machine (starting → playing ↔ paused → stopped)
on the Subsonic-server channel when the server advertises the OpenSubsonic
`playbackReport` extension (Navidrome ≥ 0.62), giving `getNowPlaying` a real
transport state and an extrapolated position. Reports send `ignoreScrobble=true`
so play counts stay on the existing `scrobble.view` 50% path (no double count),
and the effective playback speed is included so the server extrapolates position
correctly with the speed feature on.
- New `playbackReportSession` FSM mirrors the existing `playListenSession`
lifecycle hooks and is wired at the same player call sites (start / gapless
switch / queue restore / resume / 15s heartbeat / pause / seek / stop / ended /
error / app quit). Servers without the extension degrade to the unchanged
legacy `scrobble.view?submission=false` presence call.
- Gated through the existing serverCapabilities framework: a new
`FEATURE_PLAYBACK_REPORT` (auto, extension-detected). The OpenSubsonic
extensions probe now stores the full advertised list once and serves both
AudioMuse `sonicSimilarity` and `playbackReport` from it, without disturbing
the legacy Instant Mix opt-in on pre-0.62 servers.
- Now Playing dropdown shows a live position bar and a paused indicator.
- reportPlayback uses the real request params (mediaId / mediaType / positionMs).
Tests: FSM transitions + gating + legacy fallback, capability resolution,
extension-list probe storage/decoupling. Full suite green; no Tauri-boundary
changes.
* feat(now-playing): glide the Live position bar between polls
Extrapolate the position of `playing` entries locally (elapsed × reported
playbackRate from the last 10 s poll) and re-render once a second, so the Live
progress bar advances smoothly instead of jumping on each refresh. Paused and
position-less entries stay frozen. A linear width transition matched to the tick
keeps the fill gliding. Only applies to clients that report a position via the
playbackReport extension.
* fix(now-playing): tighten Live timer layout and report resume immediately
Keep the progress-bar width stable without a wide empty gap before the
timer: reserve ~2ch inside the current-time span only (right-aligned), not
on the whole clock block. Report `playing` to the server as soon as
resume() runs, matching the immediate `paused` report on pause instead of
waiting for the Rust `audio:playing` event.
* docs: CHANGELOG and credits for playbackReport live now-playing (PR #1080)
Move the source-building pipeline out of play_input.rs into a focused
source_build.rs: BuildSourceArgs/PlaybackSource, the ranged-stream probe
fallback (build_playback_source_with_probe_fallback) and its private helpers
(build_source_from_play_input, wait_or_fetch_bytes_for_stream_fallback, etc.).
play_input.rs now only handles source *selection* and drops from 799 to 420
lines; helpers that became module-internal lose their pub(super) visibility.
No behavior change.
* fix(audio): release idle output stream after 60s (#1071)
Lazy-open CPAL on first playback and close the device handle after one
minute without active audio so Windows can sleep; emit output-released
for cold resume and skip post-wake reopen when idle.
* docs: CHANGELOG and credits for idle audio stream fix (PR #1073)
* fix(audio): satisfy clippy if-same-then-else in idle watcher
* fix(audio): silence rodio DeviceSink drop unless logging is debug
Gate log_on_drop(false) on runtime should_log_debug() so normal/off
logging modes avoid stderr noise from intentional idle stream release.
* feat(audio): cold-start paused restore and silent engine prepare
After getPlayQueue on startup, apply saved seek position to the UI,
prefetch the current track to hot cache, and load the engine paused via
new audio_play startPaused so playback does not audibly start before
pause. Shared engineLoadTrackAtPosition with queue-undo restore.
* fix(audio): satisfy clippy too_many_arguments on stream arm helper
Bundle spawn_legacy_stream_start_when_armed parameters into
LegacyStreamStartWhenArmed so workspace clippy passes.
* fix(audio): release output stream immediately on stop (#1071)
Stop and natural queue end call audio_stop; close the CPAL device right
away instead of waiting for the 60s idle timer. Pause keeps the grace
period for warm resume.
* fix(audio): keep waveform mounted after stop (#1071)
Stop preserves currentTrack, so its cached analysis waveform stays valid.
Stop no longer nulls waveformBins for the still-shown track and re-hydrates
them from the analysis DB, instead of dropping to flat placeholder bars.
* test(audio): cover output_stream_is_needed branches; harden audio_play arg (#1071)
- Add unit tests for the idle-keepalive decision: empty/playing/paused main
sink, preview and fading-out sinks, and radio playing/paused. Players are
built device-less via rodio's Player::new + a Zero source, so empty()/state
are exercised without an audio device.
- Make audio_play's `start_paused` an Option<bool> defaulting to false, so the
new field is strictly additive (omitting startPaused no longer fails serde).
- Drop the unused `_engine` parameter from start_stream_idle_watcher; it
resolves the engine from the AppHandle each poll.
* refactor(audio): extract sink-swap lifecycle into sink_swap module (#1071)
Move SinkSwapInputs/swap_in_new_sink and the legacy stream-arm helper
(LegacyStreamStartWhenArmed/spawn_legacy_stream_start_when_armed) out of
play_input.rs into a focused sink_swap.rs, so source selection and source
building stay separate from sink lifecycle. play_input.rs drops from 953 to
799 lines. No behavior change.
Release CSP blocked runtime <style> injection for community themes; add
explicit style-src/style-src-elem and font-src. Derive the effective theme
synchronously in useThemeScheduler so data-theme updates on the same commit.
* fix(mpris): disable WebKit media session so radio doesn't duplicate the player
Internet radio plays through an HTML <audio> element, for which WebKitGTK
auto-registers its own MPRIS player (org.webkit.*) alongside the app's
souvlaki one. On Linux desktops that list every player, now-playing then
showed twice during radio (issue #1048: one "psysonic", one "Psysonic").
Disable the WebKit media session at main-window setup (enable-media-session,
set by GObject property name since the pinned binding has no typed setter,
guarded by find_property) so souvlaki stays the single now-playing source;
radio metadata still reaches it via mpris_set_metadata. The navigator.media
Session push in useRadioMprisSync is kept as a fallback in case a WebKitGTK
version still registers the player, so issue #816 cannot regress.
* docs(changelog): MPRIS radio duplication fix (PR #1069)
cucadmuh
Psychotoxical
ImAsra
github-actions[bot]
Kris Bennett