stable-1.6 branch of the submodule carries dank-qml-common 1ae7222 (port of d630451), so DankFlickable, DankListView and DankGridView stop treating Qt 6.12 mouse wheels as touchpads. The flake input, flake.lock and .gitmodules follow that branch.
* lock: keep the fingerprint reader usable across suspend
The reader stopped responding after the lid was closed and opened, after a long
idle period, and after suspend.
The shell held a PAM session for as long as the screen was locked, so a suspend
always landed on a running verify. fprintd then suspends the device under the
verify, pam_fprintd's ReleaseDevice fails with "device is still busy", and the
claim is never released, so every later Claim fails until fprintd restarts. The
reader is now released when the monitors blank and armed again when they come
back, so a suspend has nothing left to wedge.
The PAM profile drops to max-tries=1 and sets an explicit timeout. pam_fprintd
only counts a wrong finger, so one finished session is now one wrong scan and
maxFprintTries means the number of scans it names rather than five times that.
A timeout expiry arrives as PamResult.Error, exactly like a device fault, so a
session that ran at least as long as the configured timeout is treated as a
planned renewal instead of spending the error budget and painting an error.
A suspend that still lands while the screen is on wedges as before. Recovery now
waits out fprintd's idle timeout instead of climbing to it, which brings the
reader back about 37 seconds after resume instead of about 98.
SessionService gains lidOpened from UPower, because logind can ignore the lid on
AC and resume alone is not enough. The lock PAM state machine is fixed alongside:
PamResult.Failed was dropped, a pending retry outlived the contexts that owned
it, and an availability check reset the wrong-scan counter.
* lock: time the fingerprint session from the verify, not the claim
pam_fprintd's timeout runs from the verify, but the session age was measured
from start(), which folds the claim into it. A slow claim plus a verify that
failed before its own timeout could add up past the timeout and be taken for a
planned renewal, skipping the backoff. A session that never reached a verify had
the same problem.
The age now starts at the message pam_fprintd sends when the verify begins, so a
session that never verified can never qualify as an expiry, and a slow claim no
longer ages a fault. The message is translated and quickshell clears it before
completed, so only its timing is used, never its text.
That message arrives after pam_fprintd has already armed its deadline, because
VerifyStart is asynchronous, so the age read here runs short of the real timeout
and a slack makes it up. The bias is deliberate. A fault misread as a timeout
costs one extra retry, while a timeout misread as a fault parks the reader behind
a backoff that grows to a minute, which on an idle lock screen would leave it
unusable most of the time.
Also corrects the comment that claimed a timeout is reported without a message.
---------
(cherry picked from commit d7074a53eb)
saveSettings() set _selfWrite before setText(). When the serialized
settings were identical to the file, nothing was written, no
fileChanged followed, and the flag stayed set. The next external edit
of settings.json was then taken for our own write and never reloaded.
(cherry picked from commit 84594e6449)
* fix: keep tray order stable when tooltipTitle changes dynamically
Revert the earlier simplistic key approach: keep getTrayItemKey()
unchanged (id::tooltipTitle) so duplicate-id disambiguation (#1542)
is preserved and persisted id::tooltipTitle entries keep matching.
Instead make persisted-key lookups tolerant of dynamic tooltipTitle:
resolveOrderIndex() in sortByPreferredOrder() does exact match first,
falling back to prefix matching an id::* entry only when the id has a
single instance. findSavedTrayKey()/isTrayIdHidden() replace direct
SessionData.isHiddenTrayId() calls and showTrayId() now receives the
actual saved key.
Prefix fallback is gated by trayItemCountForId() to single-instance
ids only, so multiple items sharing an id always match exactly and
hiding/ordering one never affects the others.
* fix: bail on ambiguous saved tray keys in fallback
* tray: use occurrence index as stable key, add one-time session migration
* tray: pin stable key per item instance so it never flips when siblings come and go
* tray: assign unique per-id sequential keys to avoid collision after sibling exit
* tray: reuse free key slots so a restarted singleton keeps its plain id
(cherry picked from commit 60342b7416)
* fix(notifications): edit rules from the menu instead of deleting them
A notification rule can set an action and Bypass DND together, but
Unmute and Block in Do Not Disturb spliced the whole rule, dropping the
other setting. They now reset only their own field, and a rule is
removed once it no longer does anything (action and urgency default,
no bypass). Mute and Allow in Do Not Disturb set their field on the
app's existing rule instead of appending one that first-match policy
would shadow.
The menu now only reads and edits enabled, exact rules on the app name
or desktop entry. It used to match any rule whose pattern equaled the
app name, including summary, body, contains and regex rules, and could
delete disabled ones.
Also adds a Copy item to the notification menu on popups and center
cards, copying the summary and body as plain text.
* fix(notifications): let mute apply alongside other rule actions
Rule actions are first-match, so a mute rule behind an app's existing
rule (e.g. No history) never applied while the menu showed Unmute.
Mute is now evaluated like Bypass DND: any matching enabled mute rule
disables the popup, and the other actions stay first-match, so No
history and Mute both apply.
* fix(notifications): skip the sound for dropped and muted notifications
The notification sound played before the check that dismisses a
notification with neither a popup nor a center entry, so dropped
notifications (transient while muted, Popup only plus Mute) still made
a sound. Play it only for notifications that are kept, and not when a
mute rule applies, like Do Not Disturb.
stable-1.6: ported the rule edits and the sound/mute policy only, the Copy menu item depends on NotificationActions which stable does not have
(cherry picked from commit 51f115f11e)
* network(networkd): treat bridge, bond, team and vlan links as wired
When the IP configuration lives on a bridge or link aggregate, the
enslaved member NIC reports opState=enslaved and the aggregate itself
was dropped by isWired() because only Type=ether was accepted, so the
widget showed Disconnected. Accept the same link types NetworkManager
already handles (#1581).
Also treat lxdbr*/incusbr* as virtual so container bridges, now of a
wired type, are not picked as the uplink.
Fixes#3463
* network(networkd): only treat a bridge as wired if it has a physical member
Name prefixes can't cover every container/VM bridge (lxcbr0, waydroid0,
...). Check the bridge's members in /sys/class/net/<br>/brif instead and
only accept it as the uplink when at least one member is not virtual
(veth*, tap*, ...). This replaces the lxdbr*/incusbr* prefixes.
(cherry picked from commit f886789835)
When iwd restarts it recreates wlan0 under a new D-Bus path. Only the
Station interface was tracked on InterfacesAdded, so devicePath kept
pointing at the dead object: the PropertiesChanged match for the new
device was never registered, updateState() read Powered from the old
path, and SetWiFiEnabled() failed with "No matching method found".
WiFiEnabled stayed false after handleStationRemoved() until DMS was
restarted.
Handle the Device interface on InterfacesAdded/InterfacesRemoved: move
devicePath and its signal match to the new object and take Name and
Powered from the signal. Process every interface in an
InterfacesRemoved signal instead of stopping at the first match.
Refs #3495
(cherry picked from commit ff50b492b1)
Locking rearms the normal idle monitors and can lose their wake transition.
Enable the existing locked-monitor wake listener for idle power-off too.
(cherry picked from commit 22fa08f0ff)
* fix(theme-vscode): update editorSuggestWidget colors for improved visibility in dark, default, and light themes
* fix(theme-vscode): update button.hoverBackground to use primary_fixed colors for consistency
(cherry picked from commit c1818dee2c)
Pressing Enter in the name field fires both the text field's accepted
signal and the dialog's own Return handler, so submitAndClose() runs
twice. The first call renames and hides the window, which clears the
field; the second call then renames to an empty string and unsets the
workspace name. The OK button only fires once, so it was unaffected.
Ignore submits once the window is hidden.
Signed-off-by: Thomas Kroll <[email protected]>
(cherry picked from commit eb0cb80884)
With two or more screens connected and none of them in the lock screen
display list (a laptop docked at a different desk), every surface was
inactive and no password field rendered, so the only way out was a
TTY. getFilteredScreens only falls back to all screens with
showOnLastDisplay and exactly one screen, which the lock screen picker
never sets. Treat an empty result as all screens, the same as the
default "all" preference.
(cherry picked from commit 6a320399c9)
* feat(screenshot): add --hud flag to control HUD scale and visibility
* fix(screenshot): clamp HUD scale to 1-4 and clarify flag description
(cherry picked from commit 13192370f5)
Opening the launcher by hovering the frame edge requested the Overlay
layer via CompositorService.framePeerSurfacesUseOverlayForScreen(),
which resolves to whether the frame is configured for the screen and
is true in normal use. DankLauncherV2ModalConnected's
frameOwnsConnectedChrome requires the Top layer, so the overlay
request silently dropped connected-chrome anchoring: the launcher
opened dead-centered with a different animation instead of emerging
from the configured frame side.
The keybind/bar-button path already opens on the Top layer by passing
no overlay argument. The edge-hover trigger only runs in connected
frame mode to begin with (gated by usesConnectedFrameChromeForScreen
in this file's own eligible check), so there's no case where it needs
standalone/overlay behavior. Match the keybind path.
(cherry picked from commit 579d6f7b5c)
* tailscale: handle profile switches and bus updates reactively
* tailscale: address review feedback on test coverage and dead code
(cherry picked from commit 218f7f752f)
Apps capturing the screen through PipeWire, such as OBS, create a live
video input stream, which the camera check counted as camera use. Skip
streams tagged with the Screen media role.
(cherry picked from commit 6ecdd6d45d)
Hyprland's workspace refactor (hyprwm/Hyprland#16140) broke
Hyprland.focusedWorkspace.monitor as a way to find the focused screen
(quickshell-mirror/quickshell#1119). Quickshell already exposes
Hyprland.focusedMonitor directly, so read the monitor from there
(cherry picked from commit 05c807b7cd)
* fix(battery): report when no charge threshold exists instead of faking success
* fix(battery): hide the charge limit controls when there is no battery
Follow-up to the review on #3385: the new explanation showed on
desktops and VMs with no battery at all, where it reads as a fault
report about hardware the machine was never going to have.
Both the explanation and the apply button now require
BatteryService.batteryAvailable. Hiding the button there is the same
point the pull request makes elsewhere: offering a control that cannot
work is what produced the false success in the first place. A machine
without a battery has no charge threshold to write to, so the limit was
never applicable.
BatteryService.batteryAvailable is already the gate used twice in this
file for the same reason.
(cherry picked from commit 6b50696b8a)
* fix(clipboard): preserve pinned state when editing entries
* fix(clipboard): address PR review feedback on editEntry
* fix(clipboard): validate empty text and entry size in EditEntry
* fix(clipboard): gate saving entry on full text load completion
* fix(clipboard): add user feedback and error fallback for entry loading
* fix(clipboard): keep entry and gate save with error toast on load failure
* fix(clipboard): reject non-text edits and add translator contexts
* fix(clipboard): support alt text mime types in EditEntry and UI guards
* fix(clipboard): pre-fetch entry before opening editor and allow retry on load failure
(cherry picked from commit 4141089e63)
* add support for shelly
* fixed terminal failure
* updated to have it not allow adding ignore aur
* updated error strings
* Fixed label
port 1.6
(cherry picked from commit 8fe918ff24)
* fix(screenshot): force full-damage commit on cursor toggle with compositor v4
* test(screenshot): drop tautological cursor-mismatch test
(cherry picked from commit 0f058468a5)
* updated dank to support aqueous
* further updates
* aqueous support work
* updated to support workspace options better in aqueous
* added blur to dms shell for aqueous
* added keybinding update
* remove unneeded blur
* clean up per claud review
* swapped to ipc
* aqueous: move IPC integration to QML, delete test fixtures (since we
dont have a QML test structure even setup), delete in-repo docs
* fix staticcheck
* aqueous: consolidate more aqueous specific stuff
* add aqueous logo
---------
Co-authored-by: Zoey <[email protected]>
(cherry picked from commit b9365610c8)
* feat(screenshot): add touch and stylus input support for region capture
* fix(screenshot): guard against finishSelection on touch up during scroll phase
port 1.6
(cherry picked from commit a9fce46fec)
Root cause: we were creating a `Date` that did not exist in my timezone
(Santiago, Chile) because of DST. The JS engine decided to pick the
previous day, so I ended up with two times 5 September.
Fix: introduce a `DateOnly` object that internally uses UTC (to prevent
DST issues). Other parts of the system still use `Date`, which
probably should be refactored to use `DateOnly` as well (maybe in an
upcoming PR?). In the meantime, we provide a function to export a
`DateOnly` as a `Date`, using noon as the time instead of midnight
(because there are no DST changes at noon AFAIK).
We tested this manually (see the attached screenshots), but didn't find
a suitable place for automated tests.
port 1.6
(cherry picked from commit e4793327b8)
* fix(dash): dim today button when on current month and hide tooltip on click
* fix(dash): use Theme.shortDuration for calendar today button opacity animation
* refactor(dash): use standard DankActionButton tooltipText for calendar today button
* chore(dash): remove unused todayButton id
* fix(dash): allow returning to today when another day in current month is selected
(cherry picked from commit 07d8e28668)
* fix(bar): keep widget content on whole pixels
Bar widgets centre their content inside a box of
widgetThickness - horizontalPadding * 2, and Qt rounds the offset of a
centred anchor to whole pixels. When the content and that box have
different parity the exact offset falls on a half pixel, Qt rounds it
away from zero and the content renders half a pixel above the centre.
Widgets drawn from rectangles, such as the workspace switcher, compute
their own snapped height and stay exactly centred, which is why they
look right next to everything else in #3074.
With the stock configuration this is the normal case: the bar is 40 px,
the box is 14 px and barIconSize returns round((40 / 48) * 18) = 15.
Two sizes now round to whole pairs of pixels. The icon size, so that it
matches the parity of the box, and widgetThickness, because the box
inherits its parity from there (the padding is doubled, so it never
changes it). Going through every innerPadding from 0 to 16, nine of the
seventeen configurations currently put the icon on a half pixel and none
of them do afterwards. The 48 px default bar is untouched, the sizes
that change do so by one pixel.
Measured in a nested Hyprland at scale 1, ink centroid of the widget
content relative to the centre of its pill, before and after:
launcher -0.50 +0.00
weather -0.81 -0.52
clipboard -1.49 -0.93
notifications +0.03 +0.46
battery -0.63 -0.49
bluetooth -0.53 -0.03
Every icon moves down by half a pixel. What remains matches the ink
asymmetry of the individual glyphs, measured separately at an exact
position, and is not a placement error.
Two things this does not cover. On a fractional or HiDPI scale
horizontalPadding can land on a half pixel, which puts the box back
between the grid lines, and a text only widget such as the clock has a
line box of 15 px at the default font size, which cannot sit exactly on
the centre of a 28 px pill no matter how it is rounded.
Closes#3074
* fix(bar): give bar text a line box that can be centred
A widget without an icon is centred as one line box, and the same parity
rule applies to it as to the icon in the commit before this one. At
fontSizeSmall the line box of the bar font is 15 px tall, so on a 28 px
pill it cannot sit on the centre no matter which way the offset is
rounded, and the clock reads about half a pixel high next to everything
else. One step up the box is 16 px and it lands on the grid.
Only the branch the standard bar uses changes. The 0.9 branch below a
scale of 0.75 already comes out at 14 px, and the fontSizeMedium branch
above 1.25 is left alone because there the next even box is two steps
away, which would be a bigger change than this is worth.
Same session and same measurement as the previous commit, ink centroid
of the widget content relative to the centre of its pill:
clock and date -0.58 +0.05
battery -0.63 -0.05
weather -0.82 -0.28
The bar font goes from 12 px to 13 px, which is the visible part of this
and the reason it is a separate commit. It can be dropped without
touching the icon fix.
* bar: snap widget thickness to even device pixels and pick text sizes by line box parity
* bar: snap bar thickness to even so the pill centres on whole pixels
* bar: share the bar thickness formula so dock, osd and popouts stay in step
* dock: read the bar clearance at the compositor screen scale
---------
Co-authored-by: bbedward <[email protected]>
* feat(theme): add a selectable source color mode for wallpaper palettes
DMS always seeds the palette from matugen's most dominant candidate
(--source-color-index 0, hardcoded at both call sites). When a large
muted region outvotes a smaller vivid one, a visibly warm wallpaper
produces a cold palette: one image here goes to #8ad0ee where the warm
subject is at hue 46.
Adds a Source Color setting with seven values. Dominant is the default
and is byte-identical to today. Colorful is computed in Go before
matugen runs: quantize the image (Wu for the starting clusters, then a
weighted k-means in Lab), score the result the way caelestia-cli does,
and hand matugen the resulting hex. Its scoring walks a chroma/tone bar
down from 20 until something clears it, which biases the pick toward a
colorful, well-lit color rather than whichever muted color covers the
most pixels. The remaining five map to matugen's --prefer.
- core/internal/matugen/sourcecolor.go: extraction, scoring, and the
one helper both matugen call sites now share
- buildOnce resolves the seed once, before the StockColors branch, so
the dry run and the real run cannot disagree about it
- extraction failure logs a warning and falls through to matugen's own
extraction; it never fails a theme build
- --prefer arrived in matugen 4.1, so it is gated separately from the
existing isV4 check; on 4.0.x those five modes degrade to the
dominant color instead of aborting on an unknown argument
- --source-mode on generate and queue, plus sourceMode on the socket
handler
- Source Color dropdown in Theme & Colors, defaulting to Dominant
Adds github.com/Nadim147c/material/v3 (Apache-2.0, no dependencies) for
HCT conversions, Wu quantization and the dislike analyzer. Its
QuantizeCelebi is not used: it returns after one unconverged iteration
and its Lab.DistanceSquared is a dot product rather than a distance, so
it collapses most of an image into a single cluster.
* fix(theme): address review feedback on source color mode
Two points from review on #3020.
The quantize comment pointed at a FOLLOWUPS file that does not exist in
this repository. Replaced the reference with the measurement it was
standing in for, so the claim is self-contained.
setMatugenSourceMode called Theme.generateSystemThemesFromCurrentTheme()
directly on top of the regenSystemThemes onChange hook that Spec.set
already dispatches. Both feed a 100ms debounce so it was one generation
either way, but the direct call was redundant and is gone.
* fix(theme): keep {{image}} resolvable in colorful mode
Colorful resolves the seed itself and calls matugen with a hex source, which
leaves matugen's {{image}} unset: the builtin pywalfox template and any user
template using the keyword render the literal string "Null".
Pass the wallpaper path alongside dank16 in --import-json-string, as suggested
in review, so those templates keep working unedited. The path is made absolute
to match what matugen renders for {{image}} in the other modes.
* fix(theme): cut translator terms on the source color dropdown
Per-mode descriptions were seven of the fifteen new terms this branch adds,
and five of them differed by a single adjective. Drop them: the dropdown row
already describes what the setting does, and the docs cover each mode.
* feat(screenshot): add -g/--geometry flag to print selection coordinates
- Add -g/--geometry persistent flag to print region geometry (X,Y WxH) to stdout.
- Skip image buffer allocation, cropping, and file/clipboard/notification actions when in geometry mode.
- Support standard Enter/Space confirmation or mouse release via --no-confirm.
- Exit with code 1 when selection is cancelled, matching slurp behavior for script integrations.
- Support printing geometry across last, window, output, full, and all modes.
- Add unit tests in region_geometry_test.go.
* refactor(screenshot): deduplicate output bounds and clean geometry result helpers
- Add bounds() method on WaylandOutput to deduplicate Hyprland geometry lookup across output capture and search.
- Add geometryResult() helper in screenshot package to eliminate redundant CaptureResult literals.
- Add GeometryString() method on Region to unify geometry formatting.
- Refactor selectedLogicalGeometry() to return Region directly.
- Clean up flag validation in runScreenshot.
* fix(screenshot): address PR review on geometry coordinates and JSON output
- Derive selectedLogicalGeometry from selectionExtent and ext.logical to keep geometry in sync with inclusive selection bounds.
- Use pointer int for X and Y in screenshotMetadata so (0,0) origins are preserved in JSON output.
- Divide output dimensions by effectiveScale in bounds() for non-Hyprland compositors and handle transform rotation.
- Convert stored device coordinates to logical units in captureLastRegion geometry mode.
- Update and expand unit tests for inclusive bounds, preselection, and scaled outputs.
* fix(screenshot): prevent geometry query from clobbering last region and handle niri window geometry
- Guard SaveLastRegion in captureRegion with !s.config.Geometry to prevent overwriting saved region with logical coordinates.
- Return explicit error when attempting window -g on niri instead of printing bogus 0,0 0x0 rect.
- Populate Region dimensions in captureNiriWindow.
* feat(screenshot): add live preview panel for scroll capture
- Port scrolling live preview from dms-quick-capture backend to native screenshot.
- Calculate dynamic preview panel geometry placed beside the capture hole on the side with most available space.
- Render downscaled live stitched canvas with dark translucent background and white border in drawScrollOverlay.
- Include preview panel in Wayland input regions and allow completing capture by clicking the preview panel.
- Update cursor shape to pointer hand when hovering over preview or control buttons.
- Add unit tests for layout, hit testing, and canvas rendering.
* fix(screenshot): bound scroll preview aspect ratio to recent canvas tail
- Derive preview aspect ratio from a bounded tail window of the canvas (maxPreviewRows = frameW * maxHeight / minImageW).
- Prevent preview panel width from collapsing below minImageW and disappearing when stitched rows exceed ~4.5x width.
- Render the recent window rows in drawScrollPreview to maintain a stable panel size and display live incoming content.
- Add unit test for long scrolling capture preview bounds.
Add a refresh action to the Network detail header that calls
NetworkService.scanWifi(), mirroring the existing button in the Settings
WiFi tab. The button is hidden while WiFi is toggling or a scan is in
progress, and sits left of the connection-type tabs so the tab group
does not shift when switching between Ethernet, WiFi and Cellular.
* fix(bar): prevent focused app widget from blanking and live-sync popout on focus switch
* fix(bar): use per-output is_active flag for Niri workspace resolution
* fix(bar): gate Niri lastFocusedWindowId fallback on active popout visibility
* fix(powermenu): close on escape key and deduplicate action shortcuts
- Centralize Escape key handling in PowerMenuContent via closeRequested signal,
cancelling any active hold and enabling Esc to close both modal and popout variants.
- Remove redundant Keys.onEscapePressed wrapper in PowerMenuPopout.
- Extract handleActionShortcut and handleKeyRelease helpers to eliminate ~50 lines
of duplicated action shortcut dispatch between list and grid navigation.
* fix(powermenu): drop duplicate Keys.onEscapePressed handler
- Remove redundant Keys.onEscapePressed in PowerMenuContent, keeping Escape
handling unified under Keys.onPressed to preserve isAutoRepeat and
committedAction guards.
* feat(screenshot): support 4-corner resizing while holding Ctrl
- Hit-test selection corners when Ctrl is held to allow resizing.
- Set opposite corner as fixed anchor and update dynamic cursor shape
(NwseResize/NeswResize).
- Draw corner handle indicators when Ctrl is held or during active resize.
- Add 'Ctrl resize/move' to overlay HUD and provide complete glyph
coverage for HUD labels.
- Add comprehensive unit tests covering hit testing, inverted dragging,
incremental handle damage, and glyph coverage.
* fix(screenshot): invalidate background cache on handle and shift modifier changes
- Track backgroundHandles and backgroundShift in RenderSlot.cacheValid.
- Rebuild background slot and HUD when handle or shift modifier states
change with an existing selection, preventing incremental re-dimming
from punching holes in the HUD.
- Fix trailing newline in region_resize_test.go for golangci-lint.
* fix(screenshot): gate redraw and handle dimming on render modifier changes
- Only trigger redrawSurface when Shift or Ctrl modifier states actually
change, preventing redundant redraws when tapping Alt, Super, or Locks.
- Only re-dim and damage corner handle bounding boxes in overlayDelta and
overlayDamage when handles move or their visibility changes, preventing
unnecessary re-dimming of stationary handles over the HUD.
- Add TestOverlayDeltaIdenticalOverlayNoHandleDimming test.
* fix(screenshot): measure hud text advance accurately to prevent pill overflow
- Use (1+len(item.desc))*(charW+1) when calculating totalW in drawHUD,
matching the full 9px advance used when drawing space and description.
- Add TestHUDDimensionsContainText to verify HUD pill dimensions enclose
all text items with correct padding across all variant combinations.
* fix(screenshot): compare handle rect sets in overlay delta and damage
- Use slices.Equal to compare prev and cur handleRects in overlayDelta
and overlayDamage instead of interior equality.
- Properly re-dim and damage handles when selection corners cross monitor
boundaries where clamped interior is unchanged but top/bottom/left/right
flags change.
- Add TestOverlayDeltaHandlesClampedMonitorEdge test.
* test(screenshot): verify drawHUD output buffer pixels stay within pill bounds
- Extract hudDimensions helper on RegionSelector used by drawHUD.
- Update TestHUDDimensionsContainText to drive drawHUD directly into a
zeroed buffer and assert no lit pixel falls outside the computed
HUD pill boundary across all configuration variants.
* fix(dock): fix solid mask regression and geometry distortion when switching orientation in frame mode
* fix(dock): fix running app indicator misalignment on orientation change
* fix(dock): center multi-window running indicators around app icon
* fix(dock): fix vertical hit area collapse and transient background flicker on position change
- User icon dirs now sort ahead of system ones so overrides in
~/.icons and ~/.local/share/icons
- Desktop entry changes
refresh that index and drop cached misses instead of clearing
everything
Closes#2360Closes#3246
- New Entries captured from a mixed image+text offer keep the image as the primary mime i.e. bitmap
- Copy Text context-menu item on atlernates
Closes#2849
* feat(dock): sync background and icons in frame overview
- Keep FrameDockHost pre-instantiated in FrameWindow when dockOpenOnOverview is enabled to eliminate QML Loader instantiation lag on overview toggle.
- Add overviewReveal property to DockBody and snap slide spring positions instantly on overview enter.
- Enforce showDock check in DockBody reveal property so dock stays hidden outside overview when showDock is disabled.
* fix(dock): handle fullscreen in overview and disable hover
- Evaluate overviewReveal before hiddenForFullscreen so dock can open in overview over fullscreen windows.
- Return false when showDock is disabled to prevent hover reveals outside overview.
* fix(doctor): validate QT_QPA_PLATFORMTHEME instead of trusting its presence
dms doctor reported any non-empty QT_QPA_PLATFORMTHEME as healthy, so a
wrong value passed with "All checks passed!" while every Qt app silently
ran without a platform theme: palettes stayed unstyled and QIcon lookups
returned nothing, which is how Quickshell ends up drawing its missing-
icon placeholder for every icon an SNI app such as blueman publishes in
tray menus. The real-world trap is qt6ct-kde — the AUR package name —
set as the theme value; the settings UI toast already documents that
only the qt6ct theme requires the qt6ct-kde package, but the doctor,
the tool meant to catch exactly this, blessed the broken value.
- checkQtPlatformThemePlugin validates both QT_QPA_PLATFORMTHEME and
QT_QPA_PLATFORMTHEME_QT6 (the latter was not reported at all):
qt6ct-kde gets a targeted warning naming the fix, and any other value
has its plugin probed the way the qtengine check does — qmake/qtpaths
are asked for QT_INSTALL_PLUGINS and the conventional file names are
looked up, warning with the exact expected path on a miss
- a miss never warns when it cannot be sure: no Qt build tools and the
version-dependent plasma-integration file name both degrade to info,
mirroring the qtengine check's no-false-alarm policy
- the plugin-root interrogation moves into qtPluginRoots, shared with
the qtengine check
On an affected system:
before: ○ QT_QPA_PLATFORMTHEME · qt6ct-kde
✓ All checks passed!
after: ○ QT_QPA_PLATFORMTHEME · qt6ct-kde
● qt6ct-kde Package name, not a theme
* fix(doctor): probe qtengine plugin with a bare file name
qtenginePluginFile("", major) returns a path with a directory
component, and qtPluginPath joins platformthemes/ again, so the
qtengine check stat'd
<root>/platformthemes/platformthemes/libqt6engine-plugin.so and warned
"Not found" on every correctly installed system. Pass the bare file
name instead, and pin the resolved path with a regression test that
fails on the double join.
Found in review by claude[bot].
* Update core/cmd/dms/commands_doctor.go
Co-authored-by: claude[bot] <209825114+claude[bot]@users.noreply.github.com>
---------
Co-authored-by: claude[bot] <209825114+claude[bot]@users.noreply.github.com>
* fix(theme): take sun times from the local calendar day, not the UTC one
CalculateSunTimesWithTwilight read its reference day off date.UTC(). The
event seconds it computes are local mean solar time and lonOffset converts
them to UTC, so the instant of a local day's sunrise lands on the previous
UTC day east of Greenwich and a local sunset on the next one west of it.
Whenever the local and UTC calendars disagree, the function therefore
returned a neighbouring day's times.
Everything around it already works in local calendar days: the polar
branches of CalculateSunTimes build their day from date.Year/Month/Day,
thememode's startOfNextDay uses next.Location(), and the gamma scheduler
keys its calcDay cache on now.Location(). Only the reference day inside
the calculation disagreed.
Effect at UTC+8 (Kunming, 25.04N 102.72E): between local sunrise and 08:00
local, computeLocationSchedule got yesterday's sunrise and sunset, both
already past, fell through to the next-day branch and got today's sunrise,
also past. It reported dark with nextTransition in the past, which
schedulerLoop clamps to a one-second wait, so it re-evaluated every second
until 08:00 local and then flipped to light on its own. West of UTC the
mirror image: at UTC-7 the mode went dark from 17:00 local, when the UTC
date rolls over, until the real sunset.
The same SunTimes drive the gamma/night-light scheduler, so night light
shifted by a day the same way.
Fix: derive the reference day from date's own calendar day, keeping
dayStart anchored at UTC midnight so the lonOffset arithmetic is
unchanged, and take the orbit angle from the same day. Callers whose date
is already in UTC are unaffected.
Tests: suncalc_refday_test.go covers the reference day east and west of
UTC, stability across UTC midnight, and that UTC callers do not move.
manager_location_test.go is new to the thememode package and covers
computeLocationSchedule at both offsets, including that nextTransition is
never in the past.
* fix(theme): pick the reference day from the solar day, not the civil one
Taking the reference day straight off the caller's calendar day is right
for nearly every zone, but not for one shifted a full day away from its
own meridian. Apia, Kiritimati, Tongatapu and the Chathams all sit on
west longitudes while keeping a UTC+12:45..+14 offset, so their mean
solar time and their civil time are a day apart. Every sun event landed
on the following local day there, and the location schedule reported
dark around the clock.
Anchor on the caller's local noon and convert that to mean solar time to
pick the reference day. For every other zone this resolves to the same
date as before, so Kunming, Los Angeles and London are unchanged to the
second. dateOrbitAngle now takes the same reference day, so declination
and the equation of time no longer belong to a different day than the
events computed from them.
Covered at both levels: three antimeridian zones in the suncalc test,
and two cases in the location schedule test for the symptom itself.
Both fail without this change.
In Hyprland a second bind on an already bound key does not replace the
first, both stay live, which is why an override is written as hl.unbind
followed by hl.bind. The parser dropped that pair on the floor when it
came from a user's own Lua module:
- parseLuaLines discarded the unbind whenever the same file rebound the
key, so the DMS default was never removed
- addBind then treated the rebind as a config bind conflicting with a DMS
bind, counted it in bindsAfterDMS and returned false, so it never
reached the section
The keybinds modal therefore showed the DMS default for that key and the
user's actual action was nowhere, which is what #3145 reports.
The discard existed to protect dms/binds-user.lua, the file the settings
modal writes, which uses the same unbind+bind pair: recording its unbind
would have deleted its own rebind too, since that file is also a DMS bind
source. The guard was just too coarse, suppressing the unbind entirely
instead of exempting the one bind it belongs to.
Now the unbind is always recorded, and rebindSources remembers which file
unbound and rebound a key so removeUnboundDMSBinds spares that file's own
bind. addBind treats a bind as an override, not a conflict, when the same
file unbound the key earlier — tracked per file in fileUnbinds, saved and
restored around each parsed file so an include cannot leak into its
parent.
A config bind without a preceding unbind is unchanged: both binds really
are live in Hyprland, so it stays a conflict and keeps counting towards
"Some DMS binds may be overridden by config binds".
No other provider has an unbind concept; mangowc, niri and sway parse
formats without one, so this is Hyprland-only.
Tests: hyprland_parser_override_test.go covers the reported case, the
conflict case without an unbind, a bare unbind with no rebind, and the
dms/binds-user.lua round trip. Only the first fails without the fix; the
other three pass either way and guard against over-correcting.
* feat(clipboard): clear only the filtered entries
With a search query or type filter active, the clear button deleted
every unpinned entry regardless of what the list was showing, so
pruning a subset meant deleting entries one by one from the context
menu.
When the recents list is filtered, the button now clears exactly the
entries it shows and the modal stays open on that filter; with no
filter the previous behaviour is untouched. A filter that matches
nothing does nothing at all, rather than falling back to clearing
everything. The tooltip says "Clear Filtered" while a filter is
active, and the confirmation names the count.
Deleting one id at a time would have cost a round trip and a full
state broadcast per entry, so this adds clipboard.deleteEntries: one
bolt transaction, one state update, and it skips pinned entries
server-side so a bulk delete can never take out saved items.
Fixes#3198
* fix(clipboard): add translator context and list deleteEntries in IPC help
Review follow-up.
The two strings this branch adds carried no translator context: the
"Clear Filtered" tooltip in ClipboardHeader and the confirmation body for
a filtered clear in ClipboardHistoryContent. Both now say where they
appear, and the second one names what %1 is, per CONTRIBUTING.
"Clear All" on the same line is deliberately left alone. It is an
existing shared term with four other call sites, so a context there would
attach a clipboard-specific comment to a term the printer tab also uses.
clipboard.deleteEntries was handled but absent from the IPC method
listing that `dms debug-srv` prints, which left it undiscoverable right
next to clipboard.deleteEntry.
Catalogs untouched.
* chore: re-trigger CI
The lint-and-test failure on 21cefbf is an unrelated flake:
TestStoreEntry_CorruptDBReturnsErrorNotPanic died with SIGBUS in the
bbolt mmap. That test comes from master (c68b7e5) and passes there;
this branch only adds TestDeleteEntries_*/TestHandleDeleteEntries_*
and touches manager.go additively.
* clipboard: survive mmap faults from corrupted db pages
---------
Co-authored-by: bbedward <[email protected]>
The battery popout already surfaces the UPower energy rate, but the bar
widget could only show percentage and remaining time, so telling how fast
the battery is charging or draining meant opening the popout (or running a
third-party plugin that polls /sys/class/power_supply).
Add two independent, opt-in toggles to the battery widget - "Show Charge
Rate" and "Show Discharge Rate" - that render the signed rate next to the
existing text, e.g. "+28W" while charging and "77% (2h 11m) -19W" while
draining. Both default to off, and follow the established pattern: a global
default in SettingsData plus a per-widget-instance override.
The value comes from BatteryService.changeRate, so there is no extra process
or polling. It hides itself when the rate is below 0.05W (idle, fully
charged, or sitting at a charge limit) so it costs no bar space unless it
has something to say. Precision adapts to keep the pill narrow: one decimal
below 10W, whole watts above; vertical bars get a whole-watt line.
* fix(dock): hide overlay dock over fullscreen
* fix(dock): track fullscreen state declaratively
* fix(dock): clear overlay mask during fullscreen
* feat(dock): add fullscreen visibility option
* fix(dock): preserve auto-hide fullscreen behavior
* fix(dock): keep fullscreen mask interactive in overview
* fix(frame): host dock on overview without show dock
* feat(mpris): pin preferred media source
Add a minimal "pin media source" option so the bar media widget and OSD
keep showing a preferred player (e.g. a music player) even when another
source (e.g. YouTube in the browser) starts playing.
- MprisController: new pinnedIdentity property + setPinned(identity);
_bestPlayingPlayer() prefers a playing pinned player (exact identity
match, then case-insensitive substring fallback). Existing behavior
is unchanged when no pin is set.
- Persistence via SessionData.pinnedPlayerIdentity (spec entry added so
SessionData.set() accepts the key); empty string clears the pin.
- MediaDropdownOverlay (Media Players panel): each player row gets a
theme-colored circle indicator on the right. Empty circle = no pin
(initial state). Right-click a row to make it the default source
(circle fills); right-click another row to move the pin; right-click
the pinned row again to clear (back to empty). Left-click behavior
unchanged (selects the player).
* Update quickshell/Services/MprisController.qml
Concur with Claude's suggestion.
Co-authored-by: claude[bot] <209825114+claude[bot]@users.noreply.github.com>
* Update quickshell/Services/MprisController.qml
Co-authored-by: claude[bot] <209825114+claude[bot]@users.noreply.github.com>
---------
Co-authored-by: claude[bot] <209825114+claude[bot]@users.noreply.github.com>
CompositorService detects the compositor asynchronously: compositorInitTimer
waits 100ms before calling detectCompositor(), which shells out to identify
the Wayland socket owner and only sets the compositor flags in its callback.
KeyboardLayoutName gates its initial updateLayout() on CompositorService.isHyprland
in Component.onCompleted, which always runs before that lands, so the call never
happens and the widget renders with an empty label until the first activelayout
event, i.e. until the user switches layout by hand.
Re-run updateLayout() when the detected compositor changes. updateLayout()
already branches on the compositor internally, so one handler covers every
backend.
Co-authored-by: shalevc1098 <[email protected]>
* feat: add cellular support
* fix(settings): resolve cellular tab index collision
* fix: address review feedback for cellular support
Bump the network API version, drop the deleted term freeze file,
regenerate the settings search index, and reuse existing translation
terms where they already cover the new cellular UI.
* fix: do not assign readonly userPreference when switching network type
Assigning SettingsData.networkPreference through the readonly alias
threw on click, so the Cellular segment in Control Center did nothing.
* fix: make Cellular selectable and the modem row clickable
The Control Center type switcher was bound only to preference, so
Cellular snapped back to Ethernet. The settings modem row highlighted
on hover but had no click handler. Show the modem as a connect target
when no APN profiles exist.
* fix: create a GSM profile when connecting cellular with none saved
Clicking Mobile broadband only activated an existing profile, so a
modem with no APN did nothing useful. Fall back to AddAndActivate
like ethernet connect.
* fix: address second round of cellular review feedback
- Guard cellularDevices map with devMutex via snapshot/by-iface/set/remove
helpers, mirroring the existing ethernet and wifi accessors
- Regenerate mocks with mockery v3 instead of hand-editing mock_Backend.go
- Reuse refreshCellularState() in SetCellularEnabled, DisconnectCellular,
DisconnectCellularDevice, ActivateCellularConnection and device removal
- Wrap retryErr rather than the first error when the auto-config retry fails
- Use Theme.surfaceTextSecondary/primaryHover/primaryHoverLight in the
cellular-off state instead of hand-rolled Qt.rgba values
- Drop the duplicate "No devices found" label so it can no longer render
twice with no modem and no saved config
Co-Authored-By: Claude Opus 5 (1M context) <[email protected]>
* fix: address remaining cellular review comments
- Toast "Failed to toggle cellular" instead of always "enable"
- Remove unused enableCellularRadio() wrappers
- Store Control Center network type as a value, not a list index
* fix: pass nmcli connection selector as separate argument
`nmcli con mod uuid:<UUID>` is not valid syntax -- nmcli expects the
selector keyword as its own argument, so every priority update failed
with "no such connection profile". Because practically every NM profile
has a UUID, this silently disabled autoconnect-priority and route-metric
updates for ethernet and WiFi as well as cellular.
Also fix the Cellular settings status text, which said "Connecting..."
for a flag that only tracks radio enable/disable, and add the seven
cellular terms that were missing from the translation catalogs.
Co-Authored-By: Claude Opus 5 (1M context) <[email protected]>
* apply review comments
* fix lints
* update debug doc
* fix style inconsistencies
---------
Co-authored-by: Yechiel Worenklein <[email protected]>
Co-authored-by: Claude Opus 5 (1M context) <[email protected]>
Co-authored-by: bbedward <[email protected]>
* feat(theme): add support for matugen smart mode and type
* add another dry run to get smart type preview
* fix too much keywords in search index
* revert dont extract translations
* feat gate "Auto From Wallpaper" toggle row and the available matugen scheme for the pallete drop down
* re-extract settings search index
* make portal sync no longer disables matugen smart mode
* make sure cosmic theme mode also follow change from smart mode
* add translator context for Auto (Light Mode) / Auto (Dark Mode) / Auto From Wallpaper
* fix: gate matugen smart mode checks on the dynamic theme so stock themes keep portal sync and fix tile labels in control center
* fix(suncalc): parenthesise the sunrise-equation denominator
`sunHourAngle` computed
cos(z) / cos(lat) * cos(dec) - tan(lat) * tan(dec)
but the sunrise equation is
cos(H) = cos(z) / (cos(lat) * cos(dec)) - tan(lat) * tan(dec)
In Go `a / b * c` is `(a / b) * c`, so `cos(dec)` multiplied where it should divide.
The new test asserts a property rather than a formula: feeding the returned hour
angle back through the definition of the zenith angle,
cos(z) = sin(lat)*sin(dec) + cos(lat)*cos(dec)*cos(H)
has to reproduce the zenith angle that was requested. It fails on the old expression
and passes on the corrected one, over a grid of latitudes from the equator to the
polar circle, declinations across a full year, and both zenith targets the callers
use (elevTwilight -6, elevDaylight 3).
Worst-case error before the fix, in clock time:
Singapore lat 1.35 285 s
Mexico City lat 19.43 312 s
Cairo lat 30.04 354 s
New York lat 40.71 439 s
Trieste lat 45.69 512 s
London lat 51.51 671 s
Oslo lat 59.91 654 s
Tromso lat 69.65 1443 s
The error grows with latitude and with |declination|, and is largest at the civil
twilight zenith — which is where the gamma ramp starts and ends.
It also moves the boundary at which Acos leaves its domain, i.e. where the code
decides midnight sun / polar night. Swept over 505062 (lat, dec, zenith) points, the
classification differs at 961 of them (0.190%), all beyond 88 degrees of latitude.
The existing suncalc tests keep passing: they assert sunrise and sunset to the hour,
which is two orders of magnitude coarser than this error.
* fix(suncalc): classify each solar threshold independently
---------
Co-authored-by: Mark2Mac <[email protected]>
Co-authored-by: bbedward <[email protected]>
Adds a user-configurable spacing between system tray icons in the bar,
available per widget via the tray context menu in Settings > Widgets
("Icon Spacing" stepper, 0-20px), with a global default fallback.
- SettingsData/SettingsSpec: new trayIconSpacing setting (default 0)
- SystemTrayBar: apply configuredIconSpacing to the horizontal Row and
vertical Column layouts, and account for it in auto-overflow item
estimation and drag-and-drop slot math (shift offsets and target
index calculation)
- WidgetsTabSection: "Icon Spacing" stepper in the tray widget context
menu, following the existing "Max Visible" pattern
- WidgetsTab: keep trayIconSpacing when restoring per-widget data
Scorer was dropping apps that only matched on id or genericName, even though AppSearchService.searchApplications already matches both. Add id (350) and genericName (prefix 800, contains 400) tiers so the V2 launcher ranks those apps the same way.
* feat(screenshot): drag region with ctrl
* fix(screenshot): rehome moved regions
* fix(screenshot): clamp region to active output
* fix(screenshot): fall back to output union when moved region exceeds active output
---------
Co-authored-by: bbedward <[email protected]>
* fix(processlist): stop polling when the popout is closed
The popout's content loader deliberately stays warm across close so
re-opening is instant, and PopoutService only defers the actual unload
until session lock or monitors-off. But both DgopService references were
bound to component lifetime rather than presentation, so after closing
the popout refCount stayed above zero: dgop kept sampling the full
process list every 3s until the shell restarted or the session was
locked and unlocked.
Gate both references on shouldBeVisible instead. Ref grows an `active`
property (defaulting to true, so every other call site is unchanged) and
ProcessesView grows an `active` property that hosts keeping it loaded
while hidden can drive. ProcessListModal is unaffected — its tab loaders
already destroy the view when the window hides.
Co-Authored-By: Claude Opus 5 (1M context) <[email protected]>
Claude-Session: https://claude.ai/code/session_01GTGf29PYTp1kpAv5HNHBzM
* Update quickshell/Modules/ProcessList/ProcessListPopout.qml
Co-authored-by: claude[bot] <209825114+claude[bot]@users.noreply.github.com>
---------
Co-authored-by: Claude Opus 5 (1M context) <[email protected]>
Co-authored-by: BB <[email protected]>
Co-authored-by: claude[bot] <209825114+claude[bot]@users.noreply.github.com>
wallpaper: drop redundant FBOs
startup: create QtMultimedia players lazily on-demand instead of at
startup
launcher: fixed animated images in plugins that use them
dms: set MALLOC_CONF=thp:never,narenas:4,dirty_decay_ms:3000 for
quickshell when unset. jemalloc extets no longer round up to 2 MB
huge pages
Worth keeping an eye on any potential regressions people encounter
* feat(audio): add output port selection
* fix(audio): handle onShowPortSelector in AudioOutputDetail
* i18n: add translator context for all audio port selector strings
* fix(audio): set isLoading = false in the callback of refreshSinkPorts
* chore: remove unused property in AudioPortSelector
* fix(audio): parse availability correctly from `pactl list sinks`
* fix(audio): address output port review feedback
ensureEnabledOutput previously checked if ANY output was enabled,
including virtual fallback outputs. A profile with eDP-1 disabled and
FALLBACK enabled was treated as valid, causing the only real monitor
to be turned off during hotplug transitions.
Now ensureEnabledOutput only considers outputs with make+model (real
monitors) when checking for enabled outputs. When all real outputs are
disabled, it enables the first real one. When no real outputs exist in
the profile at all, applyConfigEntry writes an empty config and lets
the compositor handle fallback via its built-in default rules.
Fixes the case where unplugging an external monitor leaves the laptop
display disabled because a stale FALLBACK auto-profile was applied.
The idle inhibitor ("Keep Awake") lives only in memory: SessionService holds it
as a plain `property bool idleInhibited: false` and nothing ever writes it out.
Every restart of the shell silently clears it — the user's explicit choice is
dropped with no message and no trace, and the pill just disappears from the bar.
Restarting is not an edge case. `dms restart` is a first-class CLI command, and
the unit shipped in this repo (assets/systemd/dms.service) sets
`Restart=on-failure` with `TimeoutStopSec=10`, so a shell that is slow to stop
is killed and brought back automatically. Each of those paths turns Keep Awake
off without saying so.
Every other Control Center toggle already survives a restart. Measured on
upstream master, over the state read by Modules/ControlCenter/Components/
DragDropGrid.qml and its neighbours:
doNotDisturb in SessionSpec: yes
isLightMode in SessionSpec: yes
nightModeEnabled in SessionSpec: yes
wallpaperCyclingEnabled in SessionSpec: yes
idleInhibited in SessionSpec: NO
So this follows the doNotDisturb idiom exactly rather than inventing a
mechanism: one spec entry, one property, one setter that saves.
Restore happens through two entry points because the singletons are lazily
created and their order is not fixed: `onLoaded` covers "service instantiated
before the session file is read", `Component.onCompleted` covers the opposite.
enableIdleInhibit() is idempotent, so whichever runs second is a no-op.
The default stays false and SessionStore.toJson omits defaults, so an untouched
session.json gains no new key.
Related: #1483 asks for a configurable timer. The field added here is the same
shape doNotDisturb uses with doNotDisturbUntil, so a later `idleInhibitedUntil`
slots in beside it without another migration.
Co-authored-by: Mark2Mac <[email protected]>
* fix(dgop): merge child processes by default
* fix(dgop): show the process list on first open
parseData threw away the first processes payload and waited 1s for a
second, on the grounds that the first carries no CPU deltas. That holds
for the cpu module, whose prime stays, but the processes sample is
usable as it arrives, so the wait only left the list blank.
* fix(dgop): refetch when the module set changes mid-request
updateAllStats returned early while a request was in flight, and that
request was built from the modules as they stood when it went out.
Opening the process list during the bar's periodic poll got back a
response with no process data, then a full updateInterval of waiting.
Queue a refetch and reissue when the response lands.
* fix(dgop): clear pendingUpdate when no request is sent
* fix(dgop): prime the queued refetch so the cpu cursor spans an interval
* core: bump dgop version
---------
Co-authored-by: bbedward <[email protected]>
The star history chart in the README is currently broken because the upstream provider is affected by GitHub stargazer API restrictions. Point the chart and its link at the star-history.dera.page mirror, which uses a data source that requires no API token, restoring the chart.
The dms daemon always initializes IntervalSeconds to its 30-minute
default and has no persistence of its own; the saved SettingsData
value was only ever pushed on manual interaction with the Settings UI
dropdown. Result: every daemon (re)start silently reverts to checking
every 30 minutes regardless of the user's configured interval.
setupHomeTrash redirects HOME and XDG_DATA_HOME, but allTrashDirs also
walks readMountPoints() looking for <mount>/.Trash/<uid> and
<mount>/.Trash-<uid>, and no environment variable affects that walk. So
Count() and List() included whatever the developer had trashed on a
second filesystem, and Empty() deleted it for real: with an orphaned
file under a mount trash (no matching .trashinfo, so the count assert
still saw 3), TestEmptyClearsAll passed and removed the file.
Route the walk through a package-level mountPoints variable and have
setupHomeTrash swap in an empty list, restoring it via t.Cleanup.
Production behavior is unchanged.
v0.0.27 builds unicode width LUTs in init() which slows every dms
invocation by ~15ms
Pinning to pre-0.0.27 makes dms ipc invocations even faster than
invoking the quuickshell binary itself after #3081
* feat(theme): support qtengine as a Qt platform theme
qtengine is a single platform theme plugin serving both Qt5 and Qt6 from one JSON config, so it needs neither the qt5ct/qt6ct split nor qt6ct-kde.
It reads a KDE .colors scheme by path, which the kcolorscheme template already generates, so no new matugen template is required.
DMS now merges theme.colorScheme and theme.iconTheme into $XDG_CONFIG_HOME/qtengine/config.json after each generate, on icon theme changes, and from Apply Qt Colors.
The merge preserves every other key,refuses to touch a config it cannot parse, and only names a color scheme that exists.
The write is atomic, and its mtime bump is what trips qtengine's watcher and repaints running apps.
QT_QPA_PLATFORMTHEME=qtengine is accepted without the missing-env-var toast, Apply Qt Colors no longer fails on a machine without qt5ct or qt6ct, and dms doctor reports whether the plugin is installed where Qt actually searches, per Qt major.
* fix(theme): resolve qtengine config via XDG, tighten doctor probing
---------
Co-authored-by: bbedward <[email protected]>
* feat(dock): attach bar to screen edge
* feat(dock): add bar length padding
* feat(dock): expand bar padding range
* refactor(dock): group bar corner settings
* fix(i18n): avoid attached term variant
* fix(settings): group corner style with goth corners
* fix(dankbar): attach flush bars to the screen edge
---------
Co-authored-by: bbedward <[email protected]>
* fix(frame): align connected chrome layer
Connected chrome now follows overlay modal/popout descriptors so the frame-rendered background stays on the same Wayland layer as the content. This currently moves the frame window as a whole, so hosted bars may follow the overlay layer while such surfaces are presented.
* fix(frame): keep overlay-opted surfaces out of connected chrome
---------
Co-authored-by: bbedward <[email protected]>
* feat(settings): persist named location for theme & gamma auto modes
* chore: bump dank-qml-common for location search support
* fix(settings): keep weather location name when search fills coordinates
* style(settings): unify manual coordinates layout with weather
- settings.json/session.json now store only values that differ from
spec defaults
- Machine-specific state moves out of settings.json into session.json
- Add dms backup create/restore: tar.gz of ~/.config/DankMaterialShell
fixes#3027
* feat(theme): add extract current theme to JSON file
Adds an 'Extract' button in Settings > Theme Color that exports the
current theme (generic/dynamic/custom/registry) to a dual-mode dark+light
JSON file via the DMS file save dialog.
* refactor(theme): simplify extract button to single DankButton
* refactor(theme): move extract button inside dynamic theme section
* fix(theme): address review feedback on extract theme save
* refactor: utilize dankActionButton
---------
handleSubscribe forwards every service except location, so neither
subscribe-all connections nor explicit location subscriptions ever
receive location events, and the shell's only source of coordinates
is a single getState pull that races the geoclue TimeThreshold and
loses. Wire location into the fan-out like loginctl/sysupdate:
initial state on subscribe, then manager pushes.
DankSlideout always held WlrKeyboardFocus.OnDemand regardless of modals,
causing Notepad to steal focus back when Launcher was opened.
ModalManager: clone currentModalsByScreen on mutation so QML bindings
re-evaluate when modals open/close.
DankSlideout: set keyboardFocus to None while a modal is active on the
same screen, letting the modal (launcher) claim focus exclusively.
* feat(tray): add IPC action to open tray menus
* tray: open IPC-requested menus on the focused screen only
---------
Co-authored-by: bbedward <[email protected]>
closeHoverSurfaces() called TrayMenuManager.closeAllMenus(), tearing down
every registered tray menu including ones opened by right-click. With
hoverPopouts enabled, moving the pointer after a right-click re-ran the
hit test and closed the menu before it could be reached.
Tag each menu with how it was opened and add closeHoverMenus(), which
only closes hover-opened menus. The hover controller now uses that.
Click-driven callers pass no argument, so byHover stays undefined and
`byHover === true` evaluates to false.
Fixes#2978
Claude-Session: https://claude.ai/code/session_01DYF9vdsaToU4Usu7kxPCNB
Co-authored-by: Claude Opus 5 (1M context) <[email protected]>
The current isIdle detection only works if the player has a title _and_ artist metadata,
which is not appropriate for all cases, e.g. MPV playing a video does not set artist.
In the current implementation, after a video stops playing in MPV, and the window closes
(but MPV and the MPRIS plugin server is still running as a background process), the playback controls remain
shown on the bar, and have non-functional controls.
It is sufficient to only check for the "Stopped" playing state. The MPRIS specification
states that Playback_Status==Stopped should be used to indicate that *NO* track is currently
playing. A compliant player should set that, and should also set the CanPlay/CanPause boolean
to false is not track is being played.
See: https://specifications.freedesktop.org/mpris/latest/Player_Interface.html#Enum:Playback_Status
* fix(mpris): enrich metadata across equivalent players
* fix(mpris): address review feedback on source stealing and title stripping
- _bestPlayingPlayer: when the active player is playing but not
controllable, only hand ownership to a controllable peer that passes
isSameTrack; unrelated players can no longer steal the active source.
- displayTrackTitle: keep the full trackTitle for display. Suffix
stripping now only applies to the isSameTrack matching key, and only
for a known app-name allowlist (YouTube, SoundCloud, browsers, etc.)
so legitimate titles containing ' | ' are never mangled.
Co-Authored-By: Claude Fable 5 <[email protected]>
---------
Port 1.5
Popouts sized from their content sometimes open too short and clip what
they render, then come back correct on the next open.
open() snapshots the surface geometry before setting shouldBeVisible,
and the onAligned{X,Y,Width,Height}Changed handlers are all gated on
that flag. Mapping contentWindow happens in between, and that's what
finally runs the layout Qt deferred while the window was hidden.
Content-sized popouts pick up their real height in that gap, so the
correction is dropped and the surface stays short. Re-snapshot once the
flag is set.
* feat: add QR Generator modal with auto-generate on input debounce
- QRGeneratorModal with dual-buffer sequential fade animation
- Auto-generate QR code on typing stop (200ms debounce)
- Clear text button on input field
- Escape key closes modal
- Register modal in PopoutService and AppSearchService
- Go backend for text QR code generation
* fix: add QR Generator toggle to Launcher DMS settings
* qr: also add qrg built in launcher plugin
---------
Co-authored-by: bbedward <[email protected]>
* feat(BatteryService): dismiss sticky critical battery notifications
If critical notifications do not timeout, dismiss the battery-related
ones automatically when battery is plugged in.
* fix(BatteryService): use material icons for battery alerts
Since the previously specified icons are not resolved, instead reuse
icons from DMS first-party plugin for critical battery alerts.
Reopening the autostart "Browse" picker crashed in
QQmlIncubatorPrivate::incubate. Drop the ListView model on hide()
and rebind it on show() so each open starts with a fresh
QQmlDelegateModel instead of reusing a stale incubation queue.
* feat(Themes): add support for GTK3 dynamic theming
Instead of overriding named colors of all GTK3 themes via the config
directory file, only override named colors for each variant of the
adw-gtk3 theme (light and dark), and let GTK3 apps change themes
depending on the value of org.gnome.desktop.interface color-scheme.
* fix(Themes): apply requested changes
- remove gtk3 override if it's a symlink
- remove call to gtk.sh with old arg 'assets-only'
- append key to dconf path to respect dconf write signature
* refactor(Themes): separate 'include'/'apply' logic for GTK3 dynamic theming
* revert(Themes): reset timer between color-scheme refresh
- rename DMS_ENABLE_GTK4_REFRESH to DMS_ENABLE_GTK_REFRESH
- revert a3fba8
* feat(Themes): add 'remove' mode to gtk.sh
Allow to remove all inclusions and overrides for GTK.
* fix(Themes): backup adw-gtk3 original css in 'apply' (not 'include')
* style(Themes): rename argument 'include' to 'patch' for clarity
* feat(settings): add option to suppress config reload toast (#2730)
Add a toggle in Settings → Compositor Layout to disable the 'config reloaded' toast for Niri and MangoWC compositors. Session setting showConfigReloadToast defaults to true.
* chore: update settings search index for config reload toast
* refactor(settings): rename 'Compositor Notifications' to 'Compositor Toasts'
* fix(settings): reuse existing I18n.tr('Notifications') translation
* Add option to ignore app-requested notification timeout
Notifications may include an expireTimeout hint that currently takes
precedence over the configured Low/Normal/Critical durations, so
popups can disappear earlier (or later) than expected with no
interaction involved. Add notificationIgnoreAppTimeout (default off,
matching the notification spec) to let users force their own
durations regardless of what the app requests.
* Update settings search index for new timeout override toggle
lock-before-suspend released the logind sleep delay inhibitor as soon as Qt
reported the lock rendered (afterAnimating/afterRendering). That signal fires
before the compositor has committed/presented the ext-session-lock surface, so
the machine could freeze ~1 frame later with the desktop still the last
presented frame. On resume the desktop was briefly visible before the lock
appeared (both suspend and hibernate).
Gate lockerReady on WlSessionLock.secure (the ext-session-lock `locked` event),
so the inhibitor is held until the compositor confirms the session is locked and
the desktop is hidden. LockSurface passes the WlSessionLock down as sessionLock;
the readiness check returns early until secure, and re-fires on secureChanged.
The Go-side fallback timer in loginctl monitor remains as the backstop.
Reproduced on niri / s2idle with 3 outputs: debug log showed lockerReady sent
18 ms before `PM: suspend entry`; with this change the desktop no longer flashes
on resume.
Signed-off-by: Thomas Kroll <[email protected]>
Co-authored-by: Claude Opus 4.8 <[email protected]>
* Add hotspot contract layer with capability-gated dispatch.
Establish the interface and manager plumbing for hotspot support so backends can opt in without expanding the core Backend contract.
Only backends that implement HotspotBackend get hotspot state propagated; others are forced to unsupported regardless of what they self-report.
* Add IPC handlers and API docs for hotspot actions.
Wire up configure/ start/ stop hotspot through the request router so the QML service layer can drive hotspot operations.
Bump API version to 27 and document the capability-gating contract clients should follow.
* Implement NetworkManager hotspot backend.
Implement HotspotBackend on NetworkManagerBackend with DMS-owned profile management, AP capability detection, and band validation.
Device resolution is deferred to StartHotspot when no device is specified, so profiles survive hardware changes.
* Isolate client Wi-Fi state from AP-mode connections.
Filter AP-mode profiles and access points out of all client Wi-Fi paths so the DMS hotspot (and user-created APs) never appear as saved networks, visible networks, or connected state.
This protects existing client behavior before hotspot controls are exposed in the UI.
* Prefer idle radios for automatic hotspot device selection.
* Add hotspot properties and methods to QML service layer.
Expose hotspot state and actions through DMSNetworkService, NetworkService, and LegacyNetworkService.
Capability is gated on API version and backend-reported support, not backend name checks, and stays stable when Wi-Fi radio is disabled.
* Add hotspot controls to Settings and Control Center.
Settings shows a hotspot setup card as a sibling in the Wi-Fi tab with SSID, password, device, band, save, and start/ stop controls.
Control Center shows a compact row that toggles a configured hotspot or routes to Settings for initial setup.
Both stay visible when Wi-Fi is disabled, explaining the requirement instead of hiding.
* Add translator context to hotspot strings.
* fix(spotlight): fix empty results on restoring last file search query
- When "remember last query" is enabled and the last query is file search, it shows "no results" until the query is manually modified in full launcher mode.
* fix(spotlight): properly restore last selected mode(tab) when enabled in full launcher mode
* fix(spotlight): allow file trigger in plugins mode(tab)
* fix(spotlight): restore previous mode when clearing '/' file search prefix
When the '/' prefix is removed from the query, the controller now correctly reverts to the previous mode instead of remaining stuck in "files" mode.
* fix(spotlight): code review fix
* feat(power): lower refresh rate on battery (#1203)
Add a Power & Sleep setting that lowers eligible displays to 60 Hz on battery and restores their previous mode on AC power.
Skip displays without an available 60 Hz mode, displays with a single refresh-rate option, current modes at or below 60 Hz, and VRR-enabled outputs.
Apply niri changes at runtime through output commands and use wlr-output-management as fallback.
* fix(quickshell): debounce display refresh sync
* chore: regenerate settings_search_index.json for pre-commit CI fix
* fix: translator context is missing for lowerDisplayRefreshRateOnBattery
* fix: activeDisplayProfileModes is missing from SettingsSpec
* feat: added guard to syncRefreshRates function
* refactor(display): move refresh sync listeners to display service
* fix(display): publish active profile modes after config write
* feat(display): persist previous refresh modes
* fix(display): route refresh mode writes through compositor backends
* fix(display): detect preferred niri refresh modes
* some small cleanups
---------
Co-authored-by: bbedward <[email protected]>
- Provide more status feedback
- Allow 2 fprint tries within 10 seconds for DMS-managed PAM
- Updated dms greeter docs
Note: distro-managed PAM takes precedence and may use different limits.
DMS auth changes remain conditional on greetd being installed.
Fixes#2853
Port 1.5
The open() method in both DankModalStandalone and DankModalConnected unconditionally set contentWindow.screen to the focused screen, ignoring the targetScreen property.
Fix: use root.targetScreen ?? CompositorService.getFocusedScreen() so the modal appears on the configured screen when set, and falls back to focused screen otherwise.
* core: fix security and concurrency issues found in backend audit
Security:
- privesc: pipe the sudo password via stdin (sudo -S) instead of
embedding it in the command string, so it no longer appears in argv
(readable by any local user via /proc/<pid>/cmdline or ps)
- greeter: tokenize a session .desktop Exec= line into argv and execve
directly instead of running it through /bin/sh -c, closing a command-
injection path via user-writable ~/.local/share/wayland-sessions
- plugins: reject path-separator/.. in plugin id/name before joining
into a filesystem path, closing an arbitrary-directory-delete in the
uninstall/update fallback
- keybinds/hyprland: always quote unrecognized bind actions/keys when
writing generated Lua; only re-emit genuine round-tripped custom Lua
verbatim (tracked via an explicit flag), closing a Lua-injection path
- desktop/mimeapps: reject newline/bracket in mime/desktop-id fields so
they can't inject fake sections into the shared mimeapps.list
Robustness / concurrency:
- server: recover panics in the request-dispatch path so one bad
handler can't crash the daemon and drop every client
- go-wayland: recover panics in the shared dispatch choke point so a
malformed compositor event can't crash CLI tools / the daemon
- server: per-connection D-Bus client ID instead of a shared constant,
fixing cross-client signal delivery and subscription teardown
- network: guard the NetworkManager device maps with a mutex (a
concurrent map read/write here is an unrecoverable fatal error)
- cups: close the event channel on Stop() so Unsubscribe() of the last
subscriber no longer deadlocks; allocate the fresh channel in Start()
- freedesktop: reuse the shared session conn for the settings watcher
and tear it down in Close(), fixing a per-Manager conn+goroutine leak
- clipboard: mutex-guard lazy dbusConn creation
- geolocation: use WithMatchMember for the GeoClue2 LocationUpdated
signal (was WithMatchSender with an interface.member string, so the
match never fired and live location updates never arrived)
- screenshot: set failed=true on buffer/pool creation errors so the
dispatch loop doesn't wait forever for a ready/failed that never comes
* apply code review comments
---------
Co-authored-by: bbedward <[email protected]>
* fix(dash): cursor not changing to pointer in Connected Mode (#2831)
Replaces the full-screen background dismissal MouseArea in
DankPopoutConnected.qml with four edge strips that exclude
the popup body. The full-screen MouseArea at z:-1 was
suppressing child cursorShape propagation on Wayland when
combined with the full-screen input mask.
Blame: the connected popout architecture itself (the issue
does not repro in Separate Mode where background dismissal
lives in a separate PanelWindow).
* fix(dash): resolve pointer cursor hover issue in connected mode
- Added UI component & popout settings to manage ignored packages
- Added CLI support available in danklinux docs
Fixes: #2827Closes: #2344, #1741
Port 1.5
Adds `dms ipc call settings dump`, returning the complete live
configuration via SettingsData.getCurrentSettingsJson() — the same
serialization used by the read-only banner's copy button.
Useful when settings.json is a read-only symlink (NixOS/home-manager):
changes made in the Settings UI only exist in the running shell's
memory, and this makes them retrievable from the CLI instead of only
through the clipboard button in the Settings modal.
Co-authored-by: Claude Fable 5 <[email protected]>
Port 1.5
* dms doctor: warn when XDG_MENU_PREFIX is missing with KDE apps
Under non-Plasma compositors (niri, Hyprland, Sway), Dolphin's
'Open with...' dialog appears empty because KService/KApplicationTrader
requires a populated XDG menu tree, which depends on XDG_MENU_PREFIX.
This check warns the user when kde-cli-tools (keditfiletype) is
installed but XDG_MENU_PREFIX is not set.
Ref: https://specifications.freedesktop.org/menu-spec/latest/
* fix(dms doctor): improve warning message for missing XDG_MENU_PREFIX
Port 1.5
---------
Co-authored-by: Matias <[email protected]>
Since Mango v0.15.0, disable_monitor sets only_sleep=0 which removes the
output from the layout entirely. On NVIDIA (proprietary driver), the DRM
handshake fails on re-enable, leaving a black screen recoverable only via
TTY switch.
sleep_monitor/wakeup_monitor set only_sleep=1, keeping the output in the
layout while powered off. This matches the pre-v0.15 behavior and fixes
DPMS wake on NVIDIA.
These commands are available since Mango v0.15.0 (commit b0326d7).
* fix(media): use system colours in player instead of album cover accent
Closes#2831
Removes ColorQuantizer-based album art accent extraction from
MediaAccentService. All accent properties now return Theme.primary
and Theme.onPrimary directly, so the Dank Dash player always
matches the system colour scheme.
Blame: ee6f7b47 (introduced MediaAccentService & ColorQuantizer)
d799175c (tweaked seekbar accent colours)
a62ae336 (further integrated accent into player/album art)
c44ffae7 (monochrome art edge case fixes)
* feat(media): add toggle for album art accent colours (#2831)
Adds mediaUseAlbumArtAccent setting (default: off) to Settings.
When enabled, MediaAccentService extracts accent from album art
via ColorQuantizer. When disabled, uses Theme.primary.
Toggle is in Settings > Media Player.
* fix: use american english spelling (colour -> color)
Port 1.5
When a sender calls CloseNotification (per freedesktop spec) for a
persistent notification (expire-timeout=0), the onDropped handler
removed the wrapper from internal arrays but did not dismiss the
visible popup. The popup remained on screen forever for non-critical
notifications like YubiKey touch prompts.
Set wrapper.popup=false in onDropped so the popup exits via the
normal signal chain. This is minimal; it does not change timeout
handling, queue management, or the isPersistent marker.
Also reverts the previous workaround that special-cased non-critical
expireTimeout=0 to use DMS timeout. Upstream spec-correct behavior
is to honor the sender timeout, and fix the dismissal bug instead.
Closes#2814
A zero-duration ViewTransition still engages ListView's transition
machinery but resolves within the same frame, so removed delegates are
never released and displaced items are never repositioned. On a filtered
ScriptModel that re-filters on every keystroke this leaves stale rows and
empty gaps behind, visible in any DankListView-backed list.
The shortest sub-duration (remove/add = expressiveDurations.fast =
base * 0.4) is coerced to an int, so it truncates to 0ms for any
animation base duration below 3: animation speed None (0), or a Custom
duration of 1-2ms (1*0.4=0.4->0, 2*0.4=0.8->0, only 3*0.4=1.2->1
survives). Presets (250/500/750) are unaffected, which is why this only
reproduces with animations disabled or a very small custom value.
Gate the four transitions to null whenever that shortest sub-duration
would truncate to 0, so the view takes the correct instant path instead
of a broken 0ms transition. Base >= 3 is unchanged.
This makes it easier for distros to use the Makefile when creating
packages. It enables us to specify a base destination directory to
install the project files to. E.g., on Solus, when creating eopkgs, files
must be installed to a special directory path, which becomes the package
files. I believe Fedora packages, and others, are the same.
Signed-off-by: Evan Maddock <[email protected]>
The notification popup's close button (X), action button clicks, and the
cardClick body's else branch all set notificationData.popup = false
directly. This relies on wrapperConn.onPopupChanged firing startExit(),
which can be interrupted by four races:
1. enterDelay (160 ms Timer) starts notificationData.timer after the
user clicks, on an already-orphan wrapper.
2. The dismiss Timer keeps running post-click and flips popup again.
3. wrapperConn.target is set imperatively in onNotificationDataChanged
(line 293) and may be stale after NotificationPopupManager._sync()
reorders wrappers.
4. exiting or _isDestroying stuck true gates wrapperConn.enabled.
Introduce a single helper dismissPopupReliably() that stops the timer,
sets popup = false, and kicks off startExit() via Qt.callLater as a
belt-and-suspenders fallback. Apply to the three vulnerable handlers.
This matches the existing upstream pattern used in the hover, contextMenu,
and Component.onDestruction paths (which all stop the timer before
mutating popup state).
Closes#2760
* refactor(battery): restore status list layout and add individual notification types
* settings/battery: fix RTL display of top tablular-like view
---------
Co-authored-by: bbedward <[email protected]>
this makes use of the existing getFilteredScreens function to allow
users to specify which displays the lockscreen should appear on. prior
to this change, it was "all" or just one.
* feat(plugins): add update all CLI flag and settings update dialog
* feat(plugins): add comparison diff URL support and update button styling
* feat(plugins): skip system plugins in bulk CLI update
* fix(plugins): remove check shorthand to resolve conflict with config flag
* feat(plugins): inline update dialog, show version tags, restructure buttons
- Defer updates when window state hasn't actually changed
- Cache frame-blocked state per screen instead of recalculating constantly
- keep popouts cache so reopening them is instant & clean up otherwise
* fix(spotlight): Fixes 'remember last query' being ignored in connected frame mode."
* fix(spotlight): select remembered query text instead of placing cursor at end
* feat(bar): separate workspace appearance for unfocused displays
Add a toggle in the Workspace Appearance card to style workspace
indicators independently on displays that aren't currently focused.
The card is split into Focused Display and Unfocused Display(s) tabs.
When the new master toggle is off (default), unfocused displays inherit
the focused settings, preserving existing behavior. When on, unfocused
displays use their own colors (focused/occupied/unfocused/urgent) and
focused border (enable, color, thickness).
WorkspaceSwitcher resolves the focused monitor per compositor and routes
color/border resolution through isFocusedMonitor/useUnfocusedAppearance.
* refactor(bar): address review feedback for unfocused workspace appearance
- Consolidate focused-monitor lookup into BarWidgetService.getFocusedScreenName(),
adding Mango support, and drop the duplicate compositor switches in
WorkspaceSwitcher (effectiveScreenName + isFocusedMonitor now use the helper).
- Extract the shared color and border options into reusable
WorkspaceAppearanceColorOptions and WorkspaceAppearanceBorderFields components
so the focused and unfocused tabs no longer duplicate the layout.
- Gate the unfocused-display options on BarWidgetService.focusedScreenDetectionSupported
and show an explanatory note on compositors without focus detection (e.g. labwc),
instead of silently no-opping.
- Add composite plugin type with multi-surface components field
- Add startupCheck manifest field and dependency gating docs
- Add IPC runtime plugin discovery commands (plugin-scan target)
- Add getPluginPath() to data persistence reference
- Document bar reveal visibility optimization for widget plugins
- Sync plugin-schema.json with source, add dependencies field
- Bump skill version to 1.1
* feat: parallax-scroll wallpaper
Add a `Scrolling` wallpaper fill mode that translates the wallpaper crop
with the active workspace — like Android home-screen parallax, but along
niri's vertical workspace axis.
The image is scaled to cover the screen on its non-scroll axis, and the
active workspace index drives a fractional offset into the cropped
overflow along the scroll axis. Scroll position is spring-animated
CPU-side and handed to a minimal single-texture shader as a UV offset.
Per-monitor scroll position is published into SessionData so the lock
screen renders the same crop as the active workspace, keeping visual
continuity across lock/unlock.
Two implementation details worth calling out for review:
- QSG_USE_SIMPLE_ANIMATION_DRIVER=1 is exported to the spawned quickshell
process. The default animation driver advances in fixed ~16ms steps,
capping the scroll at 60Hz and desyncing it from compositor motion on
high-refresh displays; the simple driver advances by real elapsed time,
restoring native-refresh pacing. Removing it visibly regresses to 60Hz.
- The wallpaper survives wl_output rebind cycles (e.g. OLED image-cleaning
on DPMS soft-off), which otherwise leave a stuck or void background.
Recovery re-anchors the scroll target on output-lifecycle signals,
rebuilds the ShaderEffect against the current render context, and
re-attaches the wallpaper-layer surface on unlock for parallax-active
monitors — guarded against lock state so the shader gets reliable frame
hints.
* simplify bindings and gate lock screen shader in a loader
---------
Co-authored-by: bbedward <[email protected]>
- Store a desktop session identity instead of relying on raw Exec commands
- Resolve the current session desktop file when auto-login launches
- Preserve legacy memory compatibility while ignoring stale lastSessionExec
- Add regression coverage for stale /nix/store session paths
- Autologin users should rerun the process
* feat(battery): add options to toggle percentage and remaining time
* fix(battery): add SettingsSpec entries and combine verticalDisplayText
* fix(battery): format vertical battery remaining time as HH\nMM and center alignment
* fix(battery): remove percent sign from vertical layout when time is enabled
* feat(wallpaper): support configurable background color
* chore(translations): update settings search index
* revert(wallpaper): keep Pad naming instead of Center
When pam_faillock locks the account (DMS authenticates through the system PAM
stack: /etc/pam.d/login -> system-auth), the lock screen kept showing the
generic "Incorrect password - try again" even though the real cause is a
lockout, so a correct password looks rejected and only a reboot (which clears
the tmpfs /run/faillock tally) appears to help. See #2647.
The previous onMessageChanged only matched the *English* faillock strings
("The account is locked ...") and then wiped that text again on the trailing
pam_unix "Password:" prompt. On a non-English system (e.g. German) the strings
never matched, so the lockout was never surfaced at all.
Detect the notice by position rather than by text: pam emits its informational
messages within an attempt before the password prompt. Collect every non-prompt
info message and, once the prompt arrives, surface the collected lines (minus
the prompt itself) as lockMessage. If the stack short-circuits without ever
prompting (e.g. pam_faillock preauth configured as requisite), the notice is
surfaced on completion instead. This is locale-independent. A per-attempt flag
keeps the message stable across repeated locked attempts and retires it when an
attempt completes without a lockout (faillock reset / unlock_time elapsed).
Fixes#2647
When a bar with background transparency + blur uses auto-hide, the
ext-background-effect-v1 blur region was only slid off-surface via the
reveal Translate, but the region object stayed non-empty.
Hyprland gates layer-surface blur on `!m_blurRegion.empty()`, so the
non-empty region keeps blur enabled; the renderer then intersects the
off-surface region with the surface box, the clip degenerates to empty,
and an empty clip is treated as "unclipped" — so the whole bar surface
box gets blurred, leaving a blurred strip where the hidden bar would be.
(niri clips correctly, so it never showed there.)
Gate the published blur region on `barRevealed`: tear it down (null)
whenever the bar is not currently shown, so the region is genuinely
empty and the compositor disables the effect. Fixes#2656.
* fix(ui): use primaryText instead of primary for text and icons displayed on primaryContainer background in the launcher
* fix(ui): use onPrimaryContainer instead of primaryText on primaryContainer background
* launcher pills: switch to buttonBg buttonText
---------
Co-authored-by: bbedward <[email protected]>
* feat: add battery settings tab and update battery widget interactions
* fix: import Quickshell.Io in BatteryTab.qml to fix Process type compilation error
* feat: move battery tab under media player and add notify when charge limit reached option
* chore: change default notification settings to false
* feat: move battery tab under Power & Security section in settings
* feat: add notification type button selection for battery alerts
DankDropdown popups opened and closed at a fixed speed regardless of the
configured Animation Duration. The Popup inherited Qt Material's default
enter/exit transitions, whose durations are hardcoded and never reference
Theme.shortDuration.
Override enter/exit with theme-driven transitions that keep the Material
grow/fade look (scale + opacity) but read their duration from
Theme.shortDuration, so every DankDropdown instance follows the
animation-speed setting.
Fixes#2659
A DPMS off/on cycle removes an output from Quickshell.screens and re-adds
it, which DMSShell's onScreensChanged cannot distinguish from a hotplug. It
fired triggerSurfaceRecovery() on every such event; on hardware where
recreating layer-shell surfaces re-wakes the just-powered-down output, this
drives an endless recovery storm that visibly power-cycles the monitor.
Route the screen-reconnect path through a 450 ms debounce (collapsing the
output-remove + re-add pair into a single pass) followed by a 4 s cooldown,
so repeated flaps trigger at most one recovery per window. Recovery still
runs once per resume, so the partial-DPMS-resume recovery added for #2579 is
preserved. The session-resume path runs its own recovery directly and now
clears any queued screen-reconnect recovery to avoid a redundant follow-up.
Fixes#2642
Decouple weather data fetching from reverse geocoding so that
weather loads as soon as coordinates are available, even when
Nominatim is unreachable (e.g. mainland China).
- Fetch Open-Meteo weather immediately once lat/lon are known.
- Resolve city name in parallel via Nominatim -> Photon -> BigDataCloud.
- If all reverse geocoding fails, keep displaying weather with a
placeholder city name.
- Skip reverse geocoding entirely when the user has configured a
city name.
- Fall back to ip-api.com when GeoClue2 is unavailable or returns
zero coordinates.
- Add request-generation tracking to discard stale geocoding
responses after location changes.
- Hold explicit WeatherService refs in bar widget and dashboard tab.
Co-authored-by: lingdiansr <[email protected]>
When a custom lock command is configured, Lock.lock() runs the command and
returns early without engaging WlSessionLock, so IdleService.isShellLocked
never transitions true->false. That transition is the only trigger that
dismisses a completed FadeToLockWindow, so the fully-faded black overlay stays
on screen and the desktop is unusable after re-login (regression from b8f4c35,
which added the _completed guard and tied dismissal solely to isShellLocked).
Add a dedicated dismissFadeToLock signal that the custom-lock branch emits
after launching the external locker, mirroring the existing fade signal wiring,
so the overlay is handed off and torn down. The built-in WlSessionLock path is
unchanged and still dismisses on unlock.
Fixes#2595
Commit e3dbaed started skipping all system sounds (notifications, volume,
power, login) whenever an MPRIS player is playing, with no way to opt out.
Users who rely on notification sounds while listening to music lost them
entirely.
Add a 'Mute During Playback' toggle (SettingsData.muteSoundsWhenMediaPlaying,
default true) so the current behaviour is preserved, but users can disable
it to restore audible sounds while media is playing. The media check is
wrapped in shouldMuteForMedia(), which gates on the new setting before
querying the active player.
Closes#2616
* feat(tailscale): add connect/disconnect/exit-node/LAN-access backend
The Tailscale backend previously exposed only read-only status
(tailscale.getStatus, tailscale.refresh). This adds write actions through the
existing tailscale.com/client/local integration:
- tailscale.connect / tailscale.disconnect (EditPrefs WantRunning)
- tailscale.setExitNode (EditPrefs ExitNodeID; empty id clears it and any
legacy ExitNodeIP, mirroring `tailscale set --exit-node`)
- tailscale.setAllowLanAccess (EditPrefs ExitNodeAllowLANAccess)
The manager's client interface gains GetPrefs/EditPrefs; fetchState merges
ExitNodeAllowLANAccess from prefs, and Peer exposes ExitNodeOption so the UI
can list exit-node-capable peers.
* feat(tailscale): expose the new actions in TailscaleService
Adds connectTailscale/disconnectTailscale, setExitNode/clearExitNode and
setAllowLanAccess wrappers, plus derived exitNodeOptions/currentExitNode and the
exitNodeAllowLanAccess state. Write-action errors surface via ToastService.
* feat(tailscale): add connection, exit-node and LAN-access controls to the widget
The control-center widget toggle was a no-op. It now connects/disconnects, and
the detail panel gains a connection status row with a connect/disconnect button,
an exit-node picker and a LAN-access toggle.
* fix(keybinds): record numpad keys as KP_* keysyms
The shortcut recorder passed only the Qt key code to xkbKeyFromQtKey and
dropped Qt.KeypadModifier. Since Qt reuses the same Qt::Key_* values for the
numpad and the main row / nav cluster, numpad presses collapsed onto their
twins: numpad-7 became "7" (NumLock on) or "Home" (NumLock off) instead of
"KP_7"/"KP_Home", numpad-+ became "Equal", numpad-* became "8", numpad Enter
became "Return". numpad-5 with NumLock off (Qt.Key_Clear) was missing from the
map entirely, so the capture was silently dropped.
niri and the other providers bind against the xkb KP_* keysym names, so these
tokens never matched the physical key.
Pass the keypad flag through to xkbKeyFromQtKey and map keypad presses to the
KP_* keysyms: KP_0..KP_9 for the NumLock-on digit codes, the navigation names
(KP_Home, KP_End, KP_Up, ...) for the NumLock-off codes, plus the operators
and KP_Enter. Main-row keys are unaffected because they never carry the keypad
modifier.
* fix(keybinds): ignore lock keys while capturing a shortcut
NumLock/CapsLock/ScrollLock are toggles, not useful bind targets. Pressing
NumLock to switch the numpad between its digit and navigation keysyms
(KP_7 vs KP_Home) was captured as the bind itself (e.g. "Super+Num_Lock").
Skip them in the recorder like the other pure modifier keys already are.
DMS reads the niri config with kdl-go, which rejects '_' as the first
character of a bare identifier ("unexpected character _") even though niri's
own parser and the KDL spec accept it. The common trigger is the
`_JAVA_AWT_WM_NONREPARENTING "1"` environment node (the standard Java /
tiling-WM fix). When the parse aborts, `dms keybinds show` returns nothing and
the Keyboard Shortcuts UI shows no binds at all.
Extend the existing preprocessor approach (the brace fix from #2230) with
quoteLeadingUnderscoreIdents, which double-quotes bare identifiers that begin
with '_' before the text reaches kdl-go. The scan is string/comment aware and
only touches a leading '_' at a token boundary, so mid-identifier underscores
(XDG_CURRENT_DESKTOP) and underscores inside strings/comments are left alone.
Token boundaries include the ends of block comments and KDL slashdash (/-), so
a node abutting a comment with no whitespace is handled too. This is safe
because the niri parser only dispatches on fixed node/section names that never
start with '_', so re-quoting such a name cannot change what DMS reads.
Refs #2230
- Add a full popout Notepad experience w/new layout settings
- New ability to choose the left or right side of the screen
- Add more safegaurds to preserve user data throughout
- New banner to show file reload/conflict handling
- New extensionless file support
- Polish settings with gap controls, compact buttons, and shortcuts help
* feat(media-control): support scroll and right-click on audio output devices in media popout
* feat(media-control): make device list volume scrolling optional
- Keep wallpaper surfaces persistent and remove `updatesEnabled` throttling that could leave wallpapers grey or frozen after DPMS, suspend, fullscreen, or output changes
Fixes#2612Fixes#2299Fixes#2272Fixes#2028
* feat(notifications): add user-configurable font size for summary and body in notification popups
* feat: add Unset for falling back to previous default values
* fix: prek hook errors
---------
Co-authored-by: Klesh Wong <[email protected]>
Adds a thin bar pinned to the bottom of the notification card that drains
full->empty over the auto-dismiss timer, as a visual countdown to
dismissal. Opt-in via notificationShowTimeoutBar (default off), with a
toggle in Settings > Notifications. Shown for any timed notification
(timer.interval > 0, including timed criticals); inset by the corner
radius, and frozen while hovered or during the exit animation. Plain
Rectangle - no offscreen textures or shader passes. A Connections on the
timer resets the bar on every (re)start, including the in-place restart
on a deduped notification.
Co-authored-by: bogdan-velicu <[email protected]>
Co-authored-by: Claude Opus 4.8 (1M context) <[email protected]>
* feat(plugins): expose IPC handlers for runtime plugin discovery
Follow-up to #1659. That issue landed hot-reload for settings.json via
FileView.watchChanges + a 1ms Timer to skirt the JSON parse race. It does
not cover plugin discovery in runtime: adding a new plugin directory to
~/.config/DankMaterialShell/plugins/ while the shell is running is not
consistently picked up by the existing FolderListModel watcher in
PluginService.qml, and there is no IPC handle for forcing a rescan from
outside the shell.
Adds an IpcHandler on PluginService with five small functions:
- scan(): wraps existing scanPlugins(), returns count snapshot
- rescan(pluginId): wraps existing forceRescanPlugin(id), validates id
- reload(pluginId): wraps existing reloadPlugin(id), validates id
- list(): newline-joined id\tloaded\ttype\tname for every known plugin
- status(pluginId): loaded\ttype\terror for one plugin
Scope intentionally small: no file-watcher changes, no new daemons, no
schema additions. Target string "plugins" does not collide with any
existing target in DMSShellIPC.qml.
Validation:
- qs ipc --pid <PID> call plugins list returns one row per known plugin
- qs ipc --pid <PID> call plugins scan returns SCAN_TRIGGERED with count
- qs ipc --pid <PID> call plugins rescan <id> returns RESCAN_TRIGGERED
- Empty-arg paths return ERROR strings instead of throwing
- git merge-tree against origin/master is clean
* hardening(plugins): fix 7 review findings in scan-ipc IPC handlers
Follow-up to commit 43603f56 which ported PR #2601 (AvengeMedia scan-ipc)
to the fork. The original port was functionally correct but had seven
review issues that would block upstream adoption. This patch addresses
each one with a minimal, focused change.
* B1 IPC target collision: renamed `target: "plugins"` to
`target: "plugin-scan"`. The original name collided with the
existing IpcHandler in DMSShellIPC.qml:1180 which already registers
enable/disable/toggle/list/status under "plugins". The split keeps
both APIs discoverable without one shadowing the other.
* H1 Fire-and-forget scan: documented that scan() returns the
pre-debounce count and that callers must poll list/status (or wait
~200ms) to observe the post-debounce state. A proper requestId +
await mechanism was considered and rejected for scope reasons.
* H2 TOCTOU in rescan(): the handler now reads availablePlugins[id]
inside forceRescanPlugin via the id string only — no captured
object reference. A parallel resyncDebounce tick can otherwise
mutate the entry between the read and the use.
* M1 list() cap: added a 256-entry cap and a leading header line
(`# count=N returned=M`) so callers can detect truncation. A
hostile / buggy plugin mass-creating entries could otherwise
allocate 80 KB+ per IPC call.
* M2 status() prefix: "unknown\t\t" became
`ERROR: unknown pluginId '...'` to match the rest of the
handlers' prefix convention. Empty trailing field means no error.
* M3 id sanitization: every handler that takes pluginId now
validates against `/^[a-zA-Z0-9_\-:]{1,64}$` before use. This
rejects shell-injection payloads ("foo\tmalicious") and prototype
pollution attempts ("__proto__", "constructor"). The list() and
status() handlers also sanitize \t/\n in name and error fields
so callers can rely on the TSV structure.
Verification: brace count balanced (252/252). Manual read of all
five handlers confirms no logic regression. QML runtime tests are
not part of the DMS test suite, so end-to-end validation requires
rebuilding the shell — deferred to the user.
Not pushed. Stage-local-first rule.
Co-Authored-By: Claude Opus 4.8 <[email protected]>
* refactor(plugins): strip inline comments per review feedback
Purian23 in PR #2611 review: 'let's address the amount of line
comments in the code, there's not a need for all of them to exist.'
Removed 48 comment lines. The substantive justification (why the
regex, why fire-and-forget, why re-read inside forceRescanPlugin,
why the 256 cap, why the target rename) now lives in the PR body
under 'Review-driven fixes in this iteration' and 'What changed'
where the reviewer already reads it.
No code logic changed. Brace count 252/252. Diff is -48/+0 on
quickshell/Services/PluginService.qml.
Co-Authored-By: Claude Opus 4.8 <[email protected]>
---------
* feat(quickshell): add local to-do planner / tasks to calendar overview card
* feat(quickshell): add auto-focus and task reordering support in calendar planner
* feat(quickshell): implement smooth drag-and-drop task reordering and inline editing
* fix(quickshell): resolve overlap and jitter in task drag-and-drop
* fix(quickshell): fix boundary swaps and prevent task list scrambling on reload
* fix(quickshell): resolve race, fix qml error, simplify dragging, and remove python dependency
* fix(quickshell): use Log service instead of console.warn in CalendarService
* style: format QML files w/qmlformat-qt6
---------
* fix: ignore keyboard shortcuts of disabled powermenu actions
* fix typo when checking for lock shortcut
* ignore shortcuts for hidden powermenu actions in grid navigation
* ignore keyboard shortcuts of disabled actions in lock power menu
* ignore keyboard shortcuts of disabled actions in lock power menu (list navigation)
The "power off monitors on lock" path relies on wake handlers
(MouseArea/Keys) at Lock.qml's root Scope, which sit outside the
WlSessionLock surface and never receive input while locked. The feature
only appeared to work on compositors that auto-restore output power on
input; compositors that fully tear down the output (e.g. MangoWC
disable_monitor) leave the screen off until blind-unlock.
Add a seat-level IdleMonitor (wlr idle-notify, surface-independent) in
IdleService that restores monitor power on any keyboard/pointer
activity. Gated on isShellLocked + monitorsOff + the setting, so it is
inert unless that path actually powered the monitors off.
Co-authored-by: Claude Opus 4.8 (1M context) <[email protected]>
- Read and convert external compositor rules into editable DMS rules
- Preserve niri multi-match rules and add match editor
- niri background-effect (blur/xray/noise/saturation) support
* feat(control-center): add widget config overlay with showMountPath toggle for DiskUsage
Introduces WidgetConfigOverlay and DiskUsageWidgetConfigMenu components, allowing
users to toggle mount path visibility per DiskUsage widget in edit mode
* refactor(control-center): use Theme.iconSizeLarge and Theme.fontSizeLarge for small tiles
Standardize SmallDiskUsageButton and SmallBatteryButton sizing with Theme.iconSizeLarge
and Theme.fontSizeLarge, and unify font weight to Font.Bold on both tile widgets.
* fix(control-center): adjust SmallDiskUsageButton font size based on showMountPath
* refactor(control-center): simplify DiskUsage config menu i18n strings
Remove the redundant "Disk Usage Widget" title and the toggle description
to reduce translatable strings
The networkd backend treated any link reporting Type=ether as a wired uplink.
Podman bridges and veth pairs report Type=ether, so they were classified as
ethernet: isWired() short-circuited on Type and never consulted looksVirtual(),
which also lacked a podman prefix.
The link map was also never pruned. Links discovered at enumeration or via
signals were kept forever, so torn-down container interfaces lingered as
routable and could win the wired-uplink slot over the real NIC -- leaving the
indicator showing WiFi while a wired connection was active and default-routed.
- isWired()/isWireless() exclude virtual interfaces before consulting Type, and
looksVirtual() now recognises podman.
- enumerateLinks() reconciles the cached map against ListLinks via syncLinks(),
pruning links that no longer appear so dead interfaces don't accumulate.
Currently, when switching dark/light theme in DMS, pywalfox's mode does not get updated. This commit adds `pywalfox light/dark` to the matugen post_hook to fix that.
* Fix gaps and overlaps when filtering clipboard history
* feat(Clipboard-Bar-Hist): Add search/filter to saved clipboard entries as well. Change title on toggle between recent/saved.
* keep Pinned/Saved icon highlighted when selected
* add back filter animations
* Implement snap state for list views based on animation settings
---------
Co-authored-by: purian23 <[email protected]>
* feat: add native powerprofile IPC target for power profiles management
* feat: show centered PowerProfileModal with 3 square buttons for powerprofile IPC toggle
* style: enhance PowerProfileModal size, icons, description, and keyboard hints
* feat: add Space key binding to select highlighted power profile
* feat(WidgetsTabSection.qml): added dynamic width and static padding for DiskUsage
* feat(DiskUsage.qml):added functionality for dynamic width and static padding, also changed spacing to work like in cpu and ram monitor. Now they look complete and same
* fix(DiskUsage): restore display modes & formatting
---------
Co-authored-by: purian23 <[email protected]>
Add applyStoredIconTheme() calls alongside existing applyStoredTheme()
calls in loadSettings(), onLoaded, and onLoadFailed, ensuring the stored
icon theme is synced to GTK/Qt/Cosmic configs on every startup and reload.
The applyStoredIconTheme() function and its _hooks entry already existed
in the codebase but were never invoked during initial settings loading.
Co-authored-by: lingdiansr <[email protected]>
Co-authored-by: Claude Opus 4.7 <[email protected]>
* fix(Mpris): exclude idle players from active-player selection
Add MprisController.isIdle() (player Stopped with empty title and
artist). _resolveActivePlayer excludes idle players from every
selection path, and re-resolves when the active player itself goes
idle. Existing triggers (availablePlayers change, isPlaying becoming
true) do not fire for a player merely stopping.
The fallback now gates on !isIdle(p) instead of p.canPlay. canPlay
describes whether Play() would succeed; !isIdle describes whether
there is anything to surface. canControl unchanged.
When no eligible player remains, activePlayer becomes null and
consumers that gate on it unload (the bar media widget via
WidgetHost; the dash via the companion fix).
* fix(DankDash): show no-player state when active player resolves to null
showNoPlayerNow gated on _noneAvailable (player count === 0) or
activePlayer being idle. Neither covers activePlayer being null
while players remain registered, which is now possible (an
always-on player sitting stopped with empty metadata).
Key showNoPlayerNow on !activePlayer; drop the unreachable
_trulyIdle.
* add(MprisController): add track artist change handling to active player resolution
---------
* feat: unify media controls dropdown interactions, hover behavior and cycle controls
- Implement hover-to-show and hover-to-hide for all media control dropdowns.
- Make clicking the Output Devices and Media Players buttons cycle through items when expanded.
- Always display the 'speaker' icon for Output Devices to maintain visual consistency.
- Bind dropdown player properties dynamically to fix list stale rendering states.
* fix(DankDash): use trackArtist property for artist label in MediaPlayerTab
* fix(DankDash): simplify active player label for consistency with output devices
* feat(DankDash): display volume levels for audio output devices in dropdown
* fix(DankDash): display Unknown Artist when artist is empty in player list
* feat(DankDash): add keyboard shortcuts for seeking, track cycling and playback control in Media popout
* feat(DankDash): change Up/Down arrow keys to adjust volume in Media popout
* feat(DankDash): auto-open volume dropdown overlay when using Up/Down shortcuts
* feat(DankDash): add Key M shortcut to toggle mute in Media popout
* fix(mpris): clamp minimum seek position to 0.1s to prevent browser player reset
* fix(mpris): cache stable length to prevent browser transient reset issues
* fix(mpris): persist activePlayerStableLength in MprisController singleton
* fix(mpris): resolve browser player album art with raw metadata and YouTube url fallbacks
* fix(mpris): resolve browser player album art with local caching and 16:9 youtube fallbacks
* style(mpris): trim trailing whitespace in TrackArtService
* fix(mpris): address code review feedback on remote caching, stale artwork, and hover state
* fix: secure curl commands and prevent premature dropdown overlays closing on button re-hover
The toast Rectangle uses `layer.enabled: true`, which renders to a
texture before compositing. With fractional implicit/content sizes
(derived from text and icon metrics), the cached texture was being
sampled with sub-pixel interpolation and the toast looked blurry
under fractional-scale-aware compositors (e.g., niri).
Wrap toastWidth/toastHeight and implicitWidth/implicitHeight with
Theme.px(value, dpr), matching the alignment NotificationPopup.qml
already applies to its surface.
* feat(clipboard): Add editing capability to clipboard entries
* Add split save menu for clipboard editor
* Add clipboard editor shortcuts and hints
* Show full clipboard text in editor
* feat(Clipboard): Revive ClipboardEditor PR
- Original PR #1916 by @nabaco
* fix(clipboard): restore Save button targets in editor
---------
Co-authored-by: Nachum Barcohen <[email protected]>
- Introduce multi-account greeter login with per-user theme previews
- Add `dms greeter sync --profile` for secondary users with or without sudo
- Add Manage greeter group membership from Settings UI → Users Tab
- Core user is logged in tty1 while user two is in tty3, you can now seamlessly switch bewteen them
New IPC options:
- `dms ipc call sessions list`
- `dms switch-user [target]`
- New Powermenu switch users option
Fixes#2354
Root cause (tray freeze): In clickThrough mode, the PanelWindow mask uses
sectionRect() with mapToItem() to compute input regions. After DPMS resume,
the PanelWindow is recreated with width=0, and mapToItem() returns wrong
positions. The right section's implicitWidth doesn't change after creation
(fixed-size tray icons), so the mask binding is never re-evaluated when the
compositor sets the actual screen width. Adding barWindow.width as a binding
dependency ensures the mask recalculates on resize.
Root cause (wallpaper loss): Wallpaper PanelWindows are recreated by Variants
during screen reconnection before the compositor finishes output initialization.
The wallpaper Image renders at 0x0 dimensions, resulting in a black screen.
Changes:
- DankBarWindow: add barWindow.width dependency to clickThrough mask bindings
- DMSShell: add surface recovery mechanism (screen reconnect + session resume)
with progressive 2-pass timer (800ms + 2800ms) that recreates bar, Frame,
wallpaper, and dock surfaces after the compositor is ready
- WlrOutputService: re-request output state on session resume
Pulses the WiFi and Bluetooth status icons while a connection is in
progress (lock screen, DankBar control center button, control center
compound pill). The pulse is implemented as a reusable Widgets/DankBlink
component, and the wifi-connecting condition is centralized as
NetworkService.isWifiConnecting.
Co-authored-by: Claude Opus 4.7 (1M context) <[email protected]>
* feat(Spotlight): fix submenu keyboard navigation
* feat(Launcher/Spotlight): disable persisting last mode when changed by triggers
* feat(Launcher/Spotlight): add option to disable last mode being persisted
* fix(Launcher/Spotlight): fix context menu keys navigation
* fix(NiriOverviewOverlay): fix context menu keys navigation and position
Sort networks by signal bucket of 25 with SSID as tiebreaker so minor
RSSI fluctuations between scans no longer reshuffle the list while the
user is selecting a network.
Co-authored-by: Claude Opus 4.7 (1M context) <[email protected]>
- New Spotlight toggle to show/hide chips, off by default
- Updated blur effects on all launcher inputs and footers
- Fixed previous queries resurfacing
- Upated Spotlight keyboard navigation
- Added functionality to show and shortcut to keybinds from the Launcher tab
- Update to add DMS Action keys in Keyboard Shortcuts
- Defaulted in niri/hyprland includes file as `Alt+Space`
- New (IPC): `dms ipc call spotlight-bar toggle`
- Slight UI update to follow user radius
Some monitors, especially cheaper models, are slow to power on after sleep and may not be present in Quickshell.screens
when onScreensChanged is triggered. This change adds a 3-second delay before updating currentOSDsByScreen to ensure all
screens are detected, mitigating the issue of OSDs not appearing on certain monitors after resume.
Co-authored-by: Klesh Wong <[email protected]>
* networkd: classify links by Type instead of name prefix
The systemd-networkd backend decided wifi-vs-ethernet by checking
whether the interface name started with "wlan" or "wlp". Anything
else (that was not on a small virtual-prefix denylist) was treated
as wired ethernet. That misclassified two common cases:
* Nebula tunnels (kernel name like "nebula.homelab", Type=none,
Kind=tun) showed up as a wired ethernet device — DMS rendered an
"Ethernet connected" indicator whenever the overlay was up, even
with no physical NIC plugged in.
* Renamed wifi interfaces (e.g. systemd link files that rename
wlan0 to a friendlier name like "wifi") were also miscategorised
as ethernet, because they no longer matched wlan*/wlp*.
networkd already publishes the real link kind in the JSON returned
by the per-link Describe method ("ether", "wlan", "loopback",
"none"). Fetch it during enumerateLinks, cache it on linkInfo, and
classify against that. The old prefix logic is kept as a fallback
for the case where Describe ever fails to populate Type.
The package-level looksVirtual() helper replaces the unexported
isVirtualInterface method so the new classification helpers and
their tests can use it without needing a live backend.
Tests cover both the Type-based and fallback paths, including the
Nebula-shaped Type=none/tun case that motivated this change.
* networkd: cache linkType across signal ticks and unit-test Describe fallback
enumerateLinks runs on every PropertiesChanged signal under
/org/freedesktop/network1, which fires on carrier flap, DHCP renew, and
each address change. The previous version rebuilt every linkInfo from
scratch on each tick, including a synchronous Describe D-Bus round-trip
per link, despite the link Type being fixed at netlink creation. Preserve
existing entries when the D-Bus path matches, refreshing only ifindex,
and only call fetchLinkType on a genuinely new entry. A link torn down
and re-created at a different path still triggers a refetch.
Extract parseDescribeType from fetchLinkType so the JSON failure path —
malformed payload, missing Type field, wrong type for Type — can be
exercised without a live D-Bus connection. The classifier's fallback to
name-prefix heuristics already had coverage; this locks in that the seam
between Describe and the classifier surfaces an empty string on every
failure mode rather than misclassifying a link.
- Replaced fullscreen hide/reveal toggles with Show Over Fullscreen layer toggles
- Added Launcher opt to Show Over Fullscreen setting
- Kept fullscreen stacking compositor-owned via top/overlay layer choices
- Fixed Hyrland Special Workspaces
- Updated DMS Advanced Configuration docs
- Note: We do not convert your existing conf configs to lua. This update only reflects DMS defaults state
- Updated README.md to reflect changes
- Updated Keyboard shortcut support
Senders that play their own audio for a notification can set the
standard org.freedesktop.Notifications "suppress-sound" boolean hint
to ask the server not to double up. NotificationService skipped its
sound only as a side effect of the dedup early-return (when an
identically-keyed popup was still visible), so transient notifications
double-sounded while lingering ones didn't — nondeterministic. Read
notif.hints["suppress-sound"] and gate the AudioService call on it.
Fixed a visual bug where the install button would overflow its container after a plugin was installed. This was caused by the width not re-evaluating correctly and StyledText's default elide/wrap properties interfering with implicit width calculations.
* feat(settings): Added Default Apps page to settings
Added a new page to settings. This page relies on xdg-mime and gio, so their existence is checked to show the page.
This logic and the mime type stuff was added to DesktopService.qml.
Slightly reordered the settings sidebar to include an Applications category
* fix(settings): read xdg-terminals.list directly
read the file directly instead of using xdg-terminal-exec which might not be installed
* feat: add Display Profiles builtin control center widget
Adds a new built-in widget that lets users switch between their configured
display profiles directly from the Control Center, without opening Settings.
- Widget shows the active profile name and cycles profiles on pill click
- Detail panel shows all profiles as a button group for direct selection
- Integrates with existing DisplayConfigState/SettingsData singletons
- Follows the same loader/instance pattern as VPN, CUPS, and Tailscale widgets
* fix: refine display profiles widget behavior
NetworkManager rejects AddConnection for 802-1x without a non-empty
identity, so the new API v7 agent-prompt flow never reaches the
SecretAgent for fresh enterprise connections. Fall back to the existing
modal-upfront path (already wired to ask for username + password via
requiresEnterprise) when modelData.enterprise, mirroring the legacy
DMSService.apiVersion < 7 branch.
Fixes#2358
Co-authored-by: Claude Opus 4.7 (1M context) <[email protected]>
* feat: add cycleProfile IPC command for display outputs
Adds `dms ipc outputs cycleProfile` which cycles through configured display
profiles in order, wrapping around from the last back to the first.
Useful for binding to a compositor keybind (e.g. Super+P in niri/Hyprland)
without needing an external wrapper script.
* feat: add Mod+P default keybind for cycling display profiles
Binds Super+P to `dms ipc outputs cycleProfile` in both the niri and
Hyprland default keybind configs. Mod+Shift+P is already reserved for
powering off monitors on both compositors, so Mod+P is a natural fit.
* fix: preserve pre-existing connection profiles on cancelled WiFi activation
When a pre-existing WiFi connection activation is cancelled, do not
remove the connection profile — only remove profiles for newly created
connections.
* feat: retrieve WiFi/VPN secrets from D-Bus secret service with unlock
Wireless, 802.1x, VPN, and WireGuard secrets are now looked up from
org.freedesktop.Secret before prompting the user. If the keyring is
locked, the unlock prompt is triggered via Prompt.Prompt() and the
lookup retried after the vault is unlocked.
* fix(ClipboardHistory): toggle button now toggles between saved and recents tabs
- Change tooltip text to reflect current tab state ('Recent' when on saved tab, 'Saved' when on recents tab)
- Previously always switched to saved tab; now toggles between 'saved' and 'recents'
* refactor(ClipboardHistory): remove redundant History button, keep single toggle button
The Saved button now toggles between saved/recents tabs, so the separate History
button is redundant and has been removed for a cleaner UI.
getPreferredBar("clockButtonRef") returns null when the clock widget
is removed from the bar. Fall back to getPreferredBar() (any bar) so
triggerDashTab can still open the popout with default positioning.
* fix(niri): properly close KDL output block when disabled
* feat(display): parse off directive and sync disabled state from compositor configs
* feat(display): visual treatment and canvas filtering for disabled outputs
* feat(display): eager auto-profile generation with debounced auto-select
* refactor(display): pass target profile ID to confirmChanges and remove dead code
* fix(display): make profile dropdown reactive by binding to property directly
* i18n(display): add Disabled translation term for output state
The dynamic-mode currentThemeData object at Theme.qml:447-468 omitted
tertiary entirely, even though matugen generates it and writes it to
the matugen colors JSON. As a result, Theme.tertiary returned undefined
in dynamic mode, and any QML binding through .r/.g/.b/.a defaulted to
white — visible as desaturated near-white highlights in the bar's
ChromeShader (which mixes Theme.tertiary at highlight peaks).
Two-line fix:
- Add tertiary to the dynamic-mode currentThemeData via getMatugenColor
- Add property color tertiary on the Theme singleton, falling back to
secondary if tertiary is absent (covers the stock-theme path where
tertiary is added separately via buildMatugenColorsFromTheme:1811)
Fallback hex #efb8c8 is the M3 baseline tertiary (light pink) for dark
mode, in case matugen output is missing.
Theme.tertiary was already referenced at LockScreenContent.qml:737 so
the property name is the right one.
Co-authored-by: Claude Opus 4.7 (1M context) <[email protected]>
hasFullscreenToplevelOnScreen's fast-path used `NiriService.currentOutput
=== screenName` as a proxy for "the active toplevel is on screenName".
Those are two independent state channels (niri's workspace-activated
events drive currentOutput; wlr-foreign-toplevel-management drives
ToplevelManager.activeToplevel) and they don't update atomically.
When focus crosses from a fullscreen toplevel on monitor A to a
non-fullscreen toplevel on monitor B, currentOutput flips to B before
activeToplevel updates away from A's still-fullscreen-and-activated
window. For one tick, B's bar evaluates `active.fullscreen &&
active.activated && currentOutput === B` → true, hides itself, then
flips back when activeToplevel finally updates. Visible as a one-frame
bar flicker on focus changes between monitors when one has a fullscreen
window.
Replace the proxy with _toplevelOnScreen(active, screenName), the same
helper the X11 fallback path uses 25 lines below. The check now
inspects the toplevel's actual outputs instead of trusting a separate
state signal, so the race can't fire.
The per-workspace loop below was already correct; it would catch any
real fullscreen+activated toplevel on the bar's workspace regardless
of focused output. The fast-path was redundant when the assertion
held and wrong when it didn't.
Co-authored-by: Claude Opus 4.7 (1M context) <[email protected]>
* feat(tailscale): add Tailscale control center widget
Full-stack Tailscale integration for DMS control center:
Backend (Go):
- Event-driven manager via WatchIPNBus (no polling)
- Reconnects with exponential backoff when tailscaled unavailable
- Typed conversion from ipnstate.Status to QML-friendly IPC types
- Testable via tailscaleClient interface with mock watcher
- Manager cleanup in cleanupManagers()
- 19 unit tests
Frontend (QML):
- TailscaleService with WebSocket subscription
- TailscaleWidget with peer list, filter chips, search
- Copy-to-clipboard for IPs and DNS names
- Daemon lifecycle handling (offline/stopped states)
Dependencies:
- Add tailscale.com v1.96.1 (official local API client)
- Bump Go to 1.26.1 (required by tailscale.com)
* cleanups
---------
Co-authored-by: bbedward <[email protected]>
- Introduces Standalone & Connected Modes
- Updated Animations & Motion effects for both modes
- Numerous QOL tweaks and updates throughout the system
- Highly inspired to the OG Caelestia Shell / @Soramanew
When quickshell crashes or is force-killed, the child systemd-inhibit
process (used as fallback when native IdleInhibitor is unavailable) gets
reparented to PID 1 and continues to block idle indefinitely. This causes
hypridle to ignore all idle timeout rules even though the idle inhibitor
widget appears inactive after restart.
Add a cleanup step during initialization that kills any orphaned
systemd-inhibit processes from previous sessions.
Move system update flow to GO, with a CLI (convenient AIO tool) and
server integration. All lifecycle, scheduling, execution occurs on
backend side.
Run some backends via pkexec, some via terminal like paru/yay.
Incorporate flatpak as an option to update.
Add terminal override setting in GUI, in addition to $TERMINAL env
variable.
fixes#2307fixes#822fixes#1102fixes#1812fixes#1087fixes#1743
The nixosModule/homeModule path previously called `buildDmsPkgs pkgs` but
internally referenced `self.packages.${system}.default`, which was
instantiated via `nixpkgs.legacyPackages`, an unoverlayed pkgs. That
meant downstream flakes couldn't reach through their own overlays to
the dms-shell build (e.g. to swap `kdePackages.sonnet` or trim perl
out of the aspell closure).
Extract the derivation as `mkDmsShell = pkgs: ...` at the top-level
`let`, and call it from both `packages.${system}.dms-shell` (for
direct consumers of the flake) and `buildDmsPkgs pkgs` (for module
consumers, which now pass in the system's overlayed pkgs).
Also re-checks overrideAttrs / .override still work: `mkDmsShell pkgs`
is the same `pkgs.lib.makeOverridable` wrapper as before, just
parameterized on the caller's pkgs instance.
Co-authored-by: Lucas <[email protected]>
* dock: add trash bin button
- icon reflects content- filled/empty
- multiple file manager support with nautilus as default, builtin as
fallback
- settingsspec at dock tab
- context menu
* fix: remove support for builtin filebrowser
needs specific adaptors at FB adhering the trash freedesktop spec
* fix: suppress auto-hide dock with trash context menu open
* feat: allow for custom file manager command
* feat: switch runner to proc.runcommand with toasts on command failures
* Close notification center after clicking action buttons
When clicking action buttons (e.g., "View", "Activate") in the
notification center, the action fires but the popout stays open. Since
the center is a layer-shell surface, it blocks focus changes on Wayland
compositors like niri, making the action appear to do nothing.
The keyboard navigation path already closes the center after invoking
actions; this brings the mouse click path in line.
Also fix closeNotificationCenter() in PopoutService to set
notificationHistoryVisible = false (matching PopoutManager._closePopout)
instead of calling close() directly, which left the visibility property
stale and caused the bell toggle to require two presses to reopen.
Fixes#2178
* Sync notificationHistoryVisible with shouldBeVisible
NotificationCenterPopout has its own notificationHistoryVisible property
that drives open/close, but the PopoutService public API (open, close,
toggle) calls DankPopout methods directly, bypassing that property. This
leaves notificationHistoryVisible stale, causing the bell toggle to
require two presses to reopen after a programmatic close.
Sync the property from onShouldBeVisibleChanged so any caller going
through open()/close() gets the state corrected automatically.
* services: add LabwcService with quit
labwc has a minimal IPC surface (no socket, no queries) but it does
expose `labwc --exit` as a clean shutdown path. Wrap that in a small
Singleton service following the same shape as DwlService/NiriService
so the compositor-specific dispatch in callers can stay uniform.
* session: dispatch labwc logout via LabwcService
CompositorService.isLabwc was detected but never dispatched in
_logout(); labwc sessions therefore fell through to the Hyprland
exit call, which silently no-ops under labwc. Users had to set
customPowerActionLogout to 'labwc --exit' as a workaround.
Add a labwc branch alongside the existing niri/dwl/sway branches
so the power menu logout works out of the box.
Adds a 'Monochrome Icons' toggle to the system tray widget context menu.
When enabled, all system tray icons are desaturated using MultiEffect,
giving a cleaner monochrome bar aesthetic that matches minimal themes.
The setting is per-user (settings.json), defaults to false to preserve
existing behavior.
The unconditional startup scan introduced duplicate tray icons on normal boot because apps were still registering their own SNI items when the scan ran.
Use CLOCK_BOOTTIME − CLOCK_MONOTONIC to detect whether the system has ever been suspended. The startup scan now only runs when the difference exceeds 5 s, meaning at least one suspend/resume cycle has occurred.
On a fresh boot the difference is ≈ 0 and the scan is skipped entirely.
* Fix ddc brightness not applying because process exits before debounce timer runs
* Added sync.WaitGroup to DDCBackend and use it instead of loop in wait logic, added timeout in case i2c hangs.
* go fmt
---------
Co-authored-by: bbedward <[email protected]>
outputs
fixes#2199 , possibly regresses #1235 - but I think the original bug
was lastAppliedTemp being incorrectly set, this allows compositors to
cache last applied gamma
gamma: add a bunch of defensive mechanisms for output changes
related to #2197
gamma: ensure gamma is re-evaluate on resume
fixes#1036
Screen sharing was not detected by PrivacyService on Niri because:
1. Niri creates the screencast as a Stream/Output/Video node, but
screensharingActive only checked PwNodeType.VideoSource nodes.
2. looksLikeScreencast() only inspected application.name and
node.name, missing Niri's node which has an empty application.name
but identifies itself via media.name (niri-screen-cast-src).
Add Stream/Output/Video to the checked media classes and include
media.name in the screencast heuristic. Also add a forward-compatible
check for NiriService.hasActiveCast for when Niri gains cast tracking
in its IPC.
Co-authored-by: Claude Opus 4.6 (1M context) <[email protected]>
* fix(system-update): open popout on first click
The SystemUpdate widget required two clicks to open its popout.
On the first click, the LazyLoader was activated but popoutTarget
(bound to the loader's item) was still null in the MouseArea handler,
so setTriggerPosition was never called. The popout's open() then
returned early because screen was unset.
Restructure the onClicked handler to call setTriggerPosition directly
on the loaded item (matching the pattern used by Clock, Clipboard, and
other bar widgets) and use PopoutManager.requestPopout() instead of
toggle() for consistent popout management.
Co-Authored-By: Claude Opus 4.6 (1M context) <[email protected]>
* fix(system-update): include AUR packages in update list
When paru or yay is the package manager, the update list only showed
official repo packages (via checkupdates or -Qu) while the upgrade
command (paru/yay -Syu) also processes AUR packages. This mismatch
meant AUR updates appeared as a surprise during the upgrade.
Combine the repo update listing with the AUR helper's -Qua flag so
both official and AUR packages are shown in the popout before the
user triggers the upgrade. The output format is identical for both
sources, so the existing parser works unchanged.
Co-Authored-By: Claude Opus 4.6 (1M context) <[email protected]>
---------
Co-authored-by: Claude Opus 4.6 (1M context) <[email protected]>
* added login sound functionality and settings entry
* Removed debug warning that was accidentally left in
* loginSound is off by default, and fixed toggle not working
* Prevent login sound from playing in the same session
---------
Co-authored-by: Iris <[email protected]>
* fix: add TrayRecoveryService with bidirectional SNI dedup (Go server)
Add TrayRecoveryService manager that re-registers lost tray icons after
resume from suspend via native DBus scanning in the Go server.
The service resolves every registered SNI item (both well-known names and
:1.xxx connection IDs) to a canonical connection ID, building a unified
registeredConnIDs set before either scan section runs. This prevents
duplicates in both directions:
- If an app registered via well-known name, the connection-ID section
skips its :1.xxx entry.
- If an app registered via connection ID, the well-known-name section
skips its well-known name (checked through registeredConnIDs).
- After successfully registering via well-known name, registeredConnIDs
is updated immediately so the connection-ID section won't probe the
same app in the same run.
A startup scan (3 s delay) covers the common case where the DMS process
is killed during suspend and restarted by systemd (Type=dbus), so the
loginctl PrepareForSleep watcher alone is not sufficient. The startup
scan is harmless on a normal boot — it finds all items already registered
and exits early.
Go port of quickshell commit 1470aa3.
* fix: 'interface{}' can be replaced by 'any'
* TrayRecoveryService: Remove objPath parameter from registerSNI
* feat: rewind to track start on previous when past 8 seconds
Adds MprisController.previousOrRewind() which rewinds the current track
to position 0 if more than 8 seconds in (with canSeek check), and falls
back to previous() otherwise — matching traditional media player behaviour.
All previous() call sites across Media.qml, MediaPlayerTab.qml,
MediaOverviewCard.qml, LockScreenContent.qml and DMSShellIPC.qml
are updated to use the new shared function.
Co-Authored-By: Claude Sonnet 4.6 <[email protected]>
* fix: poll position in MprisController for previousOrRewind accuracy
Without a polling timer, activePlayer.position is never updated in
contexts that don't display a seekbar (e.g. the DankBar widget), causing
the position > 8 check in previousOrRewind() to always see 0 and fall
through to previous() instead of rewinding.
Co-Authored-By: Claude Sonnet 4.6 <[email protected]>
---------
Co-authored-by: Claude Sonnet 4.6 <[email protected]>
* fix: use UnsetWorkspaceName for empty input in workspace rename
Previously, empty input would set workspace name to empty string,
causing issues with Niri's unique workspace name requirement.
Now uses UnsetWorkspaceName action when input is empty.
* prevent false failed to load config toast on niri validation
Move error toast logic from StdioCollector.onStreamFinished to Process.onExited
so it only displays when niri validate actually fails (non-zero exit code),
not when stderr outputs early progress messages during config processing.
Previously, empty input would set workspace name to empty string,
causing issues with Niri's unique workspace name requirement.
Now uses UnsetWorkspaceName action when input is empty.
- Add a neutral `dms auth sync` command and reuse the shared auth flow from:
- Settings auth toggle auto-apply
- `dms greeter sync`
- `dms greeter install`
- greeter auth cleanup paths
- Rework lockscreen PAM so DMS builds /etc/pam.d/dankshell from the system login stack, but removes fingerprint and U2F from that password path. Keep /etc/pam.d/dankshell-u2f separate.
- Preserve custom PAM files in place to avoid adding duplicate greeter auth when the distro already provides it, and keep NixOS on the non-writing path.
Niri already renders fullscreen windows above the top layer-shell layer
(see render_above_top_layer() in niri's scrolling.rs). Aside from
redundancy, this check also hides the bar for windowed-fullscreen
windows (aka fake fullscreen), since the Wayland protocol reports
identical state for both.
- Re-add loginConfigWatcher so installs can still fall through to
/etc/pam.d instead of the bundled PAM assets
- Add login-faillock bundled PAM asset at runtime. Use it as the bundled fallback when dankshell config is absent
- Fix invalid bare property writes (u2fPending, u2fState, unlockInProgress,
state) in Pam.qml
- Improve lockscreen auth feedback
* feat: add neovim-lualine template, set vim.o.background automatically based on dms light/dark mode
* feat(matugen): add option to follow dms background color or not on neovim
* chore: regenerate settings and translation index after merging master
* feat(filebrowser): add filebrowser video thumbnails display
- Find cached thumbnails first
- If not found, generate with ffmpegthumbnailer
- Fallback to placeholder icon if dependency not met
* fix(filebrowser): create thumbnail cache dir if not exists
* refactor(filebrowser): prefer using Paths lib
* fix(filebrowser): only check filetype once for each file
* fix(filebrowser): early test for thumbnails
* feat: add xdgCache path
The current template doesn't work for an OOTB config of pywalfox
without manual configuration. This commit fixes the colors to work
better with its defaults.
* feat: add FullscreenToplevel detection
For animating bar for fullscreen events
* fix: respect overview reveal settings
perform the overview check first
Plugin items can set _preScored to signal a priority boost (e.g. recently
used items). Previously _preScored was only respected when a search query
was active, so no-query default lists always fell back to typeBonus+frecency
scoring, making plugin-controlled ordering impossible.
Change the condition from:
if (query && item._preScored !== undefined)
to:
if (item._preScored !== undefined && (query || item._preScored > 900))
This respects _preScored in no-query mode only when the value exceeds 900
(the plugin typeBonus), which avoids changing behaviour for "all" mode items
whose _preScored is set to 900-j by the controller (≤ 900). Items without
_preScored set continue to use the existing typeBonus + frecency formula.
fzf.js relied on stable Array.sort to preserve score ordering, which is
not guaranteed in QML's JS engine. Results appeared in arbitrary order
with low-relevance matches above exact matches. The sort comparator now
explicitly sorts by score descending, with a length-based tiebreaker so
shorter matches rank first when scores are tied.
Also fixed Object.assign mutating the shared defaultOpts object, which
could cause options to leak between Finder instances.
DankDropdown's onOpened handler now reinitializes the search when previous
search text exists, fixing the empty results shown on reopen.
Added resetSearch() for consumers to clear search state externally.
* fix: restore lock screen U2F/fingerprint auth to working state
* fix(pam): Keep SettingsData as single source of truth for auth availability
- Restores SettingsData for fingerprint/U2F, keeping lock screen and New Greeter Settings UI in sync
---------
Co-authored-by: purian23 <[email protected]>
Fixed a QML property binding timing issue where dynamically created timers
and processes for per-monitor wallpaper cycling were being assigned to
properties and then immediately read back, which could return undefined
or stale values.
The fix stores the created object in a local variable before assigning
to the property map, ensuring a valid reference is always used.
Affected functions:
- startMonitorCycling() - timer creation
- cycleToNextWallpaper() - process creation
- cycleToPrevWallpaper() - process creation
After unlocking the screen (startup lock or wake from sleep), the desktop
showed Hyprland's background color instead of the wallpaper.
WallpaperBackground disables QML updates via updatesEnabled after a 1-second
settle timer. While WlSessionLock is active, Hyprland does not composite the
background layer, so when the lock is released it needs a fresh Wayland buffer
— but none is committed because the render loop is already paused.
The previous attempt used SessionService.sessionUnlocked, which is unreliable
for the startup lock case: DMSService is not yet connected when lock() is
called at startup, so notifyLoginctl is a no-op and the loginctl state never
transitions, meaning sessionUnlocked never fires.
Fix by tracking the shell lock state directly from Lock.qml's shouldLock via
a new IdleService.isShellLocked property. WallpaperBackground watches this and
re-enables rendering for 1 second on unlock, ensuring a fresh buffer is
committed to Wayland before the compositor resumes displaying the layer.
* Add grouped element reordering to control center setting popup.
Reorganize the control center widget menu into grouped rows and add drag handles for reordering.
Introduce controlCenterGroups to drive the grouped popup layout, along with dynamic content width calculation.
Disable dependent options when their parent icon is turned off, and refine DankToggle disabled colors to better distinguish checked and unchecked states.
* Apply Control Center group order to live widget rendering.
Apply persisted `controlCenterGroupOrder` to the actual control center button rendering path instead of only using it in the settings UI.
This refactors `ControlCenterButton.qml` to derive a normalized effective group order, build a small render model from that order, and use model-driven rendering for both vertical and horizontal layouts.
Highlights:
- add a default control center group order and normalize saved order data
- ignore unknown ids, deduplicate duplicates, and append missing known groups
- add shared group visibility helpers and derive a render model from them
- render both vertical and horizontal indicators from the effective order
- preserve existing icon, color, percent text, and visibility behavior
- keep the fallback settings icon outside persisted ordering
- reconnect cached interaction refs for audio, mic, and brightness to the real rendered group containers so wheel and right-click behavior still work
- clear and refresh interaction refs on orientation, visibility, and delegate lifecycle changes
- tighten horizontal composite group sizing by measuring actual rendered content, fixing extra spacing around the audio indicator
Also updates the settings widgets UI to persist and restore control center group ordering consistently with the live control center rendering.
With programs like rofi, pressing the tab key advances to the next item
in the list. This change makes the Launcher behave in the same way,
moving the action cycling to Ctrl+Tab (and Ctrl+Shift+Tab for reverse.
* Add tmux
* Add mux modal
* Restore the settings config version
* Revert typo
* Use DankModal for InputModal
* Simplify terminal flags
* use showWithOptions for inputModals instead
* Fix translation
* use Quickshell.env("TERMINAL") to choose terminal
* Fix typo
* Hide muxModal after creating new session
* Add mux check, moved exclusion to service, And use ScriptModel
* Revert unrelated change
* Add blank line
* fix(nix/greeter): skip invalid customThemeFile in preStart
Avoid attempting to copy a null/empty/missing customThemeFile path by validating the jq result and file existence before cp.
Update distro/nix/greeter.nix
Co-authored-by: Lucas <[email protected]>
* nix/greeter: update customTheme verification
---------
Co-authored-by: Lucas <[email protected]>
Co-authored-by: LuckShiba <[email protected]>
- Split auth capability state by lock screen and greeter
- Share detection between settings UI and lock runtime
- Broaden greeter PAM include detection across supported distros
The display config UI only applied changes for compositors with a
config-file backend (niri, hyprland, dwl). For any other compositor
that supports wlr-output-management-unstable-v1 the "Apply Changes"
button was silently a no-op.
Add WlrOutputService.applyOutputsConfig() as a high-level apply that
mirrors the generateOutputsConfig() pattern of the existing services
but applies directly via the protocol instead of writing a config file.
Route the default case in backendWriteOutputsConfig() to it.
This enables using dms-shell as a wayland compositor for emacs wayland
manager (ewm).
* Issue:(Settings)Switched Neovim Mutagen Theme To Default False
* also set to false in settingsData
- this is the case when file fails to parse
---------
Co-authored-by: bbedward <[email protected]>
Fixes two cases:
- some apps (e.g., Zen browser use the "—" character at the end of
webpage name)
- in compact mode, when app has only appName, and not window name, we
should display the appName to avoid empty title.
- Added pre-run checks for greeter and setup commands to enforce policy restrictions
- Created cli-policy.default.json to define blocked commands and user messages for immutable environments.
* Not everyone uses paru or yay on Arch: Support pacman command
* Handle sudo properly when using pacman
* Move pacman to bottom per Purian23
* Remote duplicate which -- thanks Purian23!
Supersedes #1878. Rather than duplicating the moddedAppId + file path
substitution pattern inline across 8 files, this introduces two
centralized functions in Paths.qml:
- resolveIconPath(iconName): for Quickshell.iconPath() callsites,
with DesktopService.resolveIconPath() fallback
- resolveIconUrl(iconName): for image://icon/ URL callsites
All consumer files now use one-line calls. When no substitutions are
configured, moddedAppId() returns the original name unchanged (zero
cost), so this has no impact on users who don't use the feature.
Affected components:
- AppIconRenderer (8 lines → 1)
- NotificationCard, NotificationPopup, HistoryNotificationCard
- DockContextMenu, AppsDockContextMenu
- LauncherContent, LauncherTab (×3)
Co-authored-by: odtgit <[email protected]>
Follow-up to #1867. The launcher's AppIconRenderer used its own
Quickshell.iconPath() call without going through appIdSubstitutions,
so PWA icons configured via regex file path rules were not resolved
in the app launcher.
Co-authored-by: odtgit <[email protected]>
- Added global toggles in the Themes tab
- Light color & directional user ovverides
- Independent shadow overrides per/bar
- Refactored various components to sync the updated designs
* feat: add setting for first day of the week
* fix: extract settings indices
* fix: formatting mistake
* fix(ui): add outline rectangle between settings and reorder settings
* fix: don't set firstDayOfWeek automatically to system's locale
Add a new "Add by Address" flow in the printer settings that allows
users to manually add printers by IP address or hostname, enabling
printing to devices not visible via mDNS/Avahi discovery (e.g.,
printers behind Tailscale subnet routers, VPNs, or across network
boundaries).
Go backend:
- New cups.testConnection IPC method that probes remote printers via
IPP Get-Printer-Attributes with /ipp/print then / fallback
- Input validation with host sanitization and protocol allowlist
- Auth-aware probing (HTTP 401/403 reported as reachable)
- lpadmin CLI fallback for CreatePrinter/DeletePrinter when
cups-pk-helper polkit authorization fails
QML frontend:
- "Add by Address" toggle alongside existing device discovery
- Manual entry form with host, port, protocol fields
- Test Connection button with loading state and result display
- Smart PPD auto-selection by probed makeModel with driverless fallback
- All strings use I18n.tr() with translator context
Includes 20+ unit tests covering validation, probe delegation, TLS
flag propagation, auth error detection, and handler routing.
Allow appIdSubstitutions to return absolute file paths (/, ~, file://)
that bypass Quickshell.iconPath theme lookup. This enables users to map
app IDs directly to icon files on disk via the existing substitution UI.
Fixes PWA icon resolution for Chrome, Chromium and Edge PWAs where
Qt's icon theme lookup fails to find icons installed to
~/.local/share/icons/hicolor/ by the browser.
Example substitutions (Settings → Running Apps → App ID Substitutions):
^msedge-_(.+)$ → ~/.local/share/icons/hicolor/128x128/apps/msedge-$1.png
^(chrome|msedge|chromium)-(.+)$ → ~/.local/share/icons/hicolor/128x128/apps/$1-$2.png
Tested with Chrome PWAs (YouTube, Twitch, ai-ta) and Edge PWAs
(Microsoft Teams, Outlook) on niri/Wayland.
Co-authored-by: odtgit <[email protected]>
* feat: switch auto location in weather widget to use GeoClue2 instead of simple IP check
* nix: enable GeoClue2 service by default
* lint: fix line endings
* fix: fall back to IP location if GeoClue is not available
* feat: add setting to change and hotreload locale
* fix: typo in component id
* feat: add persistent locale setting
* feat: wrap useLocale in a settings set hook, enable locale hotreload when editing settings file
* chore: update translation and settings file
* feat: enable fuzzy search in locale setting
* fix: regenerate translations with official plugins cloned
* fix: revert back to system's locale for displaying certain time formats
* feat: Add FIDO2/U2F security key support for lock screen
Adds hardware security key authentication (e.g. YubiKey) with two modes:
Alternative (OR) and Second Factor (AND). Includes settings UI, PAM
integration, availability detection, and proper state cleanup.
Also fixes persist:false properties being reset on settings file reload.
* feat: Add U2F pending timeout and Escape to cancel
Cancel U2F second factor after 30s or on Escape key press,
returning to password/fingerprint input.
* fix: U2F detection honors custom PAM override for non-default key paths
* feat(lockscreen): enable use of videos as screensaver in the lock screen
* reducing debug logs
* feature becomes available only when QtMultimedia is available
This fixes the problem that the system update terminal closes when the package manager encounters a problem (exit code != 0), allowing the user to understand the problem.
Signed-off-by: Jan Phillip Greimann <[email protected]>
- Implement deep search icon resolution in DesktopService with runtime caching.
- Update Paths.getAppIcon to utilize enhanced resolution for mismatched app IDs.
- Align Workspace Switcher fallback icons with AppsDock visual style.
- Synchronize fallback text logic between Switcher and Dock using app names.
- New background toggles
- New maxIcon & maxText widget sizes (global)
- Dedicated M3 padding slider
- New independent icon scale options
- Updated logic to improve performance on single & dual bar modes
* fix: add mockery v2 to nix flake pkgs
* feat: add requests for generating wifi qr codes as png files in /tmp, and to delete them later. only supports NetworkManager backend for now.
* feat: add modal for sharing wifi via qr code and saving the code as png file.
* fix: uncomment QR code file deletion
* network: light refactor and cleanup for QR code generation
---------
Co-authored-by: bbedward <[email protected]>
* clipboard: fix html elements get parsed in clipboard entry
* Revert "clipboard: fix html elements get parsed in clipboard entry"
This reverts commit 52b11eeb98.
* clipboard: fix html elements get parsed in clipboard entry
* clipboard: improve image thumbnail
- thumbnail image is now bigger
- circular mask has been replaced with rounded rectangular mask
* dock: fix dock still visible when there's no app
When switching tabs rapidly or closing multiple tabs, the taskbar shows
"ghost" workspaces — entries with no name, no coordinates, and no active
state. The ghosts appear at positions where workspaces were removed and
then recreated by the compositor.
When a compositor removes a workspace (sends `removed` event) and the
client calls Destroy(), the proxy is marked as zombie but stays in the
Context.objects map. For server-created objects (IDs >= 0xFF000000), the
server never sends `delete_id`, so the zombie proxy persists indefinitely.
When the compositor later creates a new workspace that gets a recycled
server object ID, GetProxy() returns the old zombie proxy. The dispatch
loop in GetDispatch() checks IsZombie() and silently drops ALL events
for zombie proxies — including property events (name, id, coordinates,
state, capabilities) intended for the new workspace. This causes the
ghost workspaces with empty properties in the UI.
Fix: check IsZombie() when handling `workspace` and `workspace_group`
events that carry a `new_id` argument. If the existing proxy is a
zombie, treat it as absent and create a fresh proxy via
registerServerProxy(), which replaces the zombie in the map. Subsequent
property events are then dispatched to the live proxy.
This directory contains agent skills following the [Agent Skills](https://agentskills.io) open standard - a portable, version-controlled format for giving AI agents specialized capabilities.
Each skill is a directory with a `SKILL.md` entrypoint, optional reference docs, scripts, and templates. Agents load skills progressively: metadata at startup, full instructions on activation, and supporting files on demand.
## Available Skills
| Skill | Description |
|-------|-------------|
| [dms-plugin-dev](dms-plugin-dev/) | Develop plugins for DankMaterialShell - covers all 4 plugin types (widget, daemon, launcher, desktop), manifest creation, QML components, settings UI, data persistence, theme integration, and PopoutService usage. |
## Installation
The `.agents/skills/` directory at the project root is the standard location defined by the agentskills.io spec. Many agents discover skills from this path automatically. Some agents use their own directory conventions and need a symlink or copy.
### Claude Code
Claude Code discovers skills from `.claude/skills/` (project-level) or `~/.claude/skills/` (personal). To make skills from `.agents/skills/` available, symlink them into the Claude Code skills directory:
After linking, the skill appears in Claude Code's `/` menu as `/dms-plugin-dev`, and Claude loads it automatically when you ask about DMS plugin development.
See the [Claude Code skills docs](https://code.claude.com/docs/en/skills) for more on skill configuration, invocation control, and frontmatter options.
### Cursor
Cursor discovers skills from `.cursor/skills/` in the project root:
See [Codex skills docs](https://developers.openai.com/codex/skills/) for details.
### Other Agents
The Agent Skills standard is supported by 30+ tools including Goose, Roo Code, JetBrains Junie, Amp, OpenCode, OpenHands, Kiro, and more. Most discover skills from a dot-directory at the project root (e.g., `.goose/skills/`, `.roo/skills/`). Some read `.agents/skills/` directly.
Check the [Agent Skills client showcase](https://agentskills.io/clients) for setup instructions specific to your agent.
**Long-running processes:** Use the `Process` QML component from `Quickshell.Io` with `StdioCollector`.
**Shell commands with pipes:**`["sh", "-c", "ps aux | grep foo"]`
**Do NOT use**`globalThis.clipboard` or browser JavaScript APIs - they don't exist in the QML runtime.
## Translations (Optional)
Ship per-locale files in a `translations/` directory; DMS loads the one matching the active locale and re-reads it on locale changes:
```
MyPlugin/
└── translations/
├── es.json
└── zh_CN.json
```
File names follow DMS locale naming (`es.json`, `pt.json`, `zh_CN.json`); resolution tries the exact tag, then the hyphenated form, then the bare language. No `en.json` - the QML terms are the English strings.
Each file maps the English term to its translation, using the term as its own context bucket:
```json
{
"Cycle Speed":{
"Cycle Speed":"Velocidad de ciclo"
}
}
```
In QML, call `I18n.trFor` with the plugin id as a literal string matching `plugin.json`:
```qml
text:I18n.trFor("myPlugin","Cycle Speed")
```
Lookup order: plugin file, then the global DMS catalog, then the English term. `ExampleEmojiPlugin` in `quickshell/PLUGINS/` demonstrates the full setup. Registry plugins can apply for central POEditor translation via the registry's CONTRIBUTING guide.
## Step 11: Validate and Test
1. Validate `plugin.json` against the schema at [assets/plugin-schema.json](assets/plugin-schema.json)
2. Run the shell with verbose output: `qs -v -p $CONFIGPATH/quickshell/dms/shell.qml`
3. Open Settings > Plugins > Scan for Plugins
4. Enable your plugin and add it to the DankBar layout
**Runtime plugin discovery via IPC:**
Plugins can be scanned, rescanned, and reloaded at runtime without restarting the shell:
```bash
dms ipc plugin-scan scan # Trigger a full rescan of all plugin directories
dms ipc plugin-scan rescan <id> # Force rescan of a specific plugin
dms ipc plugin-scan reload <id> # Force reload of a loaded plugin
dms ipc plugin-scan list # List all known plugins (TSV: id, loaded, type, name)
dms ipc plugin-scan status <id> # Get status of a specific plugin (TSV: loaded, type, error)
```
Plugin IDs are validated against `^[a-zA-Z0-9_\-:]{1,64}$`.
**Common issues:**
- Plugin not detected: check plugin.json syntax with `jq . plugin.json`
- Widget not showing: ensure it's enabled AND added to a DankBar section
- Settings error: verify `settings_write` permission is declared
- Data not persisting: check pluginService injection and permissions
- Startup check failing: check `pluginService.pluginLoadErrors` or run `dms ipc plugin-scan status <id>`
## Common Mistakes
1.**Missing `settings_write` permission** - Settings UI shows error without it
2.**Missing `property var popoutService: null`** - Must declare for injection to work
3.**Missing vertical bar pill** - Widget disappears when bar is on left/right edge
4.**Hardcoded colors** - Use `Theme.*` properties, not hex values
5.**Using `globalThis.clipboard`** - Does not exist; use `Quickshell.execDetached(["dms", "cl", "copy", text])`
6.**Wrong Theme property names** - `Theme.fontSizeS` does not exist, use `Theme.fontSizeSmall`
7.**Wrong import for Quickshell** - Use `import Quickshell` (not `import QtQuick` for execDetached)
8.**Forgetting `categories` in launcher items** - Items won't display without it
9.**Not handling null pluginService** - Always use optional chaining or null checks
10.**Using `PluginComponent` for launchers** - Launchers use plain `Item`, not `PluginComponent`
11.**Using `requires` instead of `dependencies`** - `requires` is deprecated; use `dependencies`
12.**Providing both `component` and `components`** - Use one or the other, not both
13.**Missing `trigger` on composite with launcher surface** - Still required when `components` has a `launcher` key
## Quick Reference: Imports
**Widget / Daemon:**
```qml
importQtQuick
importqs.Common
importqs.Widgets
importqs.Modules.Plugins
```
**Launcher:**
```qml
importQtQuick
importqs.Services
```
**Desktop:**
```qml
importQtQuick
importqs.Common
```
**For clipboard/exec:**`import Quickshell`
**For processes:**`import Quickshell.Io`
**For networking:**`import Quickshell.Networking`
**For toast notifications:** access `ToastService` from `qs.Services`
"description":"Relative path to main QML component file. Required unless 'components' is provided.",
"pattern":"^\\./.*\\.qml$"
},
"components":{
"type":"object",
"description":"Map of surface name to relative QML component path, for multi-surface (composite) plugins. Provide any subset of surfaces; each is loaded independently.",
"properties":{
"widget":{
"type":"string",
"description":"Bar/Control Center widget component (PluginComponent)",
"description":"Trigger string for launcher activation (required for launcher type)"
},
"icon":{
"type":"string",
"description":"Material Design icon name"
},
"settings":{
"type":"string",
"description":"Path to settings component QML file",
"pattern":"^\\./.*\\.qml$"
},
"startupCheck":{
"type":"string",
"description":"Path to a non-visual (QtObject) component exposing a check(done) function that gates activation. done(null) allows; done(error) blocks, where error is a string or { title, details }.",
"pattern":"^\\./.*\\.qml$"
},
"requires_dms":{
"type":"string",
"description":"Minimum DMS version requirement (e.g., '>=0.1.18', '>0.1.0')",
"pattern":"^(>=?|<=?|=|>|<)\\d+\\.\\d+\\.\\d+$"
},
"dependencies":{
"type":"array",
"description":"Array of required system tools/dependencies (registry metadata)",
"items":{
"type":"string"
}
},
"requires":{
"type":"array",
"description":"Deprecated alias for 'dependencies'.",
NOT persisted. Shared across all instances of a plugin. Use for cross-instance state synchronization (multi-monitor consistency, multi-instance widgets).
| User preferences (API keys, themes, intervals) | `savePluginData` / `loadPluginData` | Yes (settings.json) | Per plugin |
| Runtime state (history, cache, counters) | `savePluginState` / `loadPluginState` | Yes (state file) | Per plugin |
| Cross-instance sync (multi-monitor data) | `PluginGlobalVar` or `getGlobalVar`/`setGlobalVar` | No (runtime only) | All instances |
| Quick reactive reads from settings | `pluginData` property | N/A (read-only) | Per instance |
## Plugin Path
Retrieve a plugin's installation directory at runtime:
```qml
vardir=pluginService.getPluginPath(pluginId)
```
Returns the absolute path to the plugin's directory (e.g., `~/.config/DankMaterialShell/plugins/MyPlugin`), or an empty string if the plugin is not found. Useful for loading bundled assets (images, data files) relative to the plugin's location.
## Important Notes
1.**pluginData is reactive** - bindings update automatically when data changes
2.**Global vars are NOT persistent** - they reset when the shell restarts
3.**State vs Data** - data is for user-facing settings, state is for internal runtime data
4.**Null safety** - always check `pluginService` is not null before calling methods
5.**Signal namespacing** - global var signals include `pluginId` to filter for your plugin
6.**Performance** - global vars are efficient for frequent updates; settings writes are batched
| `editMode` | bool | `true` when user is dragging/resizing |
| `widgetWidth` | real | Current widget container width |
| `widgetHeight` | real | Current widget container height |
## Optional Properties
Define these on your root item to customize behavior:
| Property | Type | Default | Description |
|----------|------|---------|-------------|
| `minWidth` | real | 100 | Minimum allowed width during resize |
| `minHeight` | real | 100 | Minimum allowed height during resize |
## Position and Size Persistence
Position (`desktopX`, `desktopY`) and size (`desktopWidth`, `desktopHeight`) are automatically managed by the DesktopPluginWrapper. You do not need to handle persistence for positioning.
## Edit Mode
When `editMode` is true, the user is repositioning or resizing. Use this to:
- Show visual indicators (borders, handles)
- Disable interactive elements to prevent accidental actions
Launcher plugins extend the DMS launcher with custom searchable items and actions. They use trigger-based filtering and integrate directly into the app drawer.
## Base Component
Launchers use a plain `Item` (not PluginComponent):
| `author` | string | Creator name or email | Non-empty |
| `type` | string | Plugin type | One of: `widget`, `daemon`, `launcher`, `desktop`, `composite` |
| `capabilities` | array | Plugin capabilities | At least 1 string item |
One of `component` or `components` is required (not both):
| Field | Type | Description | Validation |
|-------|------|-------------|------------|
| `component` | string | Path to main QML file (single-surface plugins) | Must start with `./`, end with `.qml` |
| `components` | object | Map of surface name to QML path (multi-surface plugins) | At least 1 entry; keys: `widget`, `desktop`, `daemon`, `launcher` |
| `components` has `launcher` key | `trigger` | Same requirement applies to composite plugins with a launcher surface |
## Optional Fields
| Field | Type | Description |
|-------|------|-------------|
| `icon` | string | Material Design icon name (displayed in plugin list UI) |
| `settings` | string | Path to settings QML file (must start with `./`, end with `.qml`) |
| `startupCheck` | string | Path to a QtObject component that gates plugin activation via a `check(done)` function (must start with `./`, end with `.qml`). See Startup Check section below. |
| `process` | Execute system commands | No (not currently enforced) |
| `network` | Network access | No (not currently enforced) |
If your plugin has a `settings` component but does not declare `settings_write`, users will see an error instead of the settings UI.
## Capabilities
Capabilities are free-form strings that describe what the plugin does. Common values:
-`dankbar-widget` - general bar widget
-`control-center` - integrates with Control Center
-`monitoring` - system/service monitoring
-`launcher` - launcher search provider
-`desktop-widget` - desktop background widget
-`ai` - AI/LLM integration
-`slideout` - uses slideout panel
## Startup Check
The `startupCheck` field points to a non-visual `QtObject` component that gates plugin activation on dependency checks. The component must expose a `check(done)` function:
The schema allows additional properties (`"additionalProperties": true`), so plugins can include custom fields. Common custom fields seen in production plugins:
-`viewMode` - launcher display mode (`"tile"` for image grids)
-`viewModeEnforced` - lock launcher to specific view mode (`true`/`false`)
The `PopoutService` singleton lets plugins control all DMS popouts and modals. It is automatically injected into widget, daemon, and settings components.
## Setup
Declare the property in your component for injection to work:
```qml
propertyvarpopoutService:null
```
Without this declaration, injection fails with: `Cannot assign to non-existent property "popoutService"`
## Popouts (DankPopout-based)
| Component | Open | Close | Toggle |
|-----------|------|-------|--------|
| Control Center | `openControlCenter()` | `closeControlCenter()` | `toggleControlCenter()` |
| Notification Center | `openNotificationCenter()` | `closeNotificationCenter()` | `toggleNotificationCenter()` |
All plugin settings use the `PluginSettings` wrapper. Setting components auto-save on change and auto-load on creation.
## PluginSettings Wrapper
```qml
importQtQuick
importqs.Common
importqs.Widgets
importqs.Modules.Plugins
PluginSettings{
pluginId:"yourPlugin"// Required: must match plugin.json id
// Setting components go here
}
```
**Important:** The plugin must declare `"permissions": ["settings_write"]` in plugin.json for the settings UI to render. Without it, users see an error.
**PluginSettings provides to children:**
-`saveValue(key, value)` - save a setting value
-`loadValue(key, defaultValue)` - load a setting value
-`saveState(key, value)` - save plugin state (separate file)
-`loadState(key, defaultValue)` - load plugin state
| `ccWidgetSecondaryText` | string | Subtitle / status text |
| `ccWidgetIsActive` | bool | Active state (changes styling) |
**CC signals:**
| Signal | When fired |
|--------|-----------|
| `ccWidgetToggled()` | Icon area clicked |
| `ccWidgetExpanded()` | Expand area clicked (CompoundPill only) |
**CC sizing rules:**
- 25% width - SmallToggleButton (icon only)
- 50% width - ToggleButton (no detail) or CompoundPill (with ccDetailContent)
- Users can resize in CC edit mode
### Detail Content (CompoundPill)
Add an expandable panel below the CC widget:
```qml
ccDetailContent:Component{
Rectangle{
implicitHeight:200
color:Theme.surfaceContainerHigh
radius:Theme.cornerRadius
Column{
anchors.fill:parent
anchors.margins:Theme.spacingM
spacing:Theme.spacingS
// Detail UI here
}
}
}
```
## Visibility Control
Conditionally show/hide the bar pill:
```qml
PluginComponent{
visibilityCommand:"pgrep -x myapp"
visibilityInterval:5// seconds between checks
}
```
**Bar reveal optimization:** The visibility timer automatically pauses while the bar is hidden (auto-hide mode) and resumes checks when the bar is revealed. This is handled via the internal `_barRevealed` property - no plugin code needed. Plugins using `visibilityCommand` with `visibilityInterval` benefit from this automatically.
## Popout Namespace
For plugins with multiple popout instances, use `layerNamespacePlugin` to isolate popout state:
```qml
PluginComponent{
layerNamespacePlugin:true
}
```
## Reading Plugin Data
Access saved settings reactively via the injected `pluginData`:
Review pull request #${{ github.event.issue.number }} in ${{ github.repository }}, requested by @${{ github.event.comment.user.login }}:
---
${{ github.event.comment.body }}
---
Text after `/claude review` is an optional focus. Honor it but still report anything serious outside it.
The PR head is checked out. Use `gh pr view ${{ github.event.issue.number }}` and `gh pr diff ${{ github.event.issue.number }}`, read surrounding code as needed.
Steps:
1. Read CONTRIBUTING.md.
2. Read the PR title, description and diff.
3. Check existing comments so you don't repeat them: `gh api repos/${{ github.repository }}/pulls/${{ github.event.issue.number }}/comments` and `gh pr view ${{ github.event.issue.number }} --comments`. Skip findings already covered by a comment that still applies.
4. Verify each candidate finding against the code. Drop anything you can't confirm.
5. Post one inline comment per confirmed finding with `mcp__github_inline_comment__create_inline_comment` (`confirmed: true`). Say what's wrong, why, and the fix. Add a committable suggestion only if it alone fully fixes the issue.
6. Post one new summary comment with `gh pr comment ${{ github.event.issue.number }} --body ...`. Never edit earlier comments. Format:
## Claude review
<one line verdict>
- <finding, file:line> per inline comment posted this run
Checked: <one line>. Model: claude-opus-5.
No findings: "No issues found." plus what was checked.
Flag only issues introduced by this PR:
- Bugs: wrong logic, crashes, nil/undefined access, races, leaks (timers, connections, processes, handles), bad Go error handling, edge cases that will clearly be hit.
- QML: undefined refs, missing imports, binding loops, bindings that never update, handlers on the wrong object, raw Flickable/ListView/ScrollView instead of the Dank* wrappers.
- CONTRIBUTING.md rules: user-facing strings in `I18n.tr()` with translator context, reuse existing terms from quickshell/translations/en.json; en.json and template.json must not be touched in a PR; mocks under core/internal/mocks/ are regenerated not hand-edited; new dank-qml-common contract properties need the matching library change.
- Go: unformatted code, go.mod/go.sum out of sync with imports, tests that no longer compile.
Don't flag style or naming, pre-existing issues, anything a linter or compiler catches, speculative input-dependent problems, missing tests or docs, or refactor ideas. Not sure it's real: don't post it.
Review only. Never edit files, commit, push, approve or request changes. Post each finding once. Keep comments short.
entry:bash -c 'python3 quickshell/translations/extract_settings_index.py >/dev/null || exit 1; if ! git diff --exit-code -- quickshell/translations/settings_search_index.json; then echo "settings_search_index.json is out of date; run quickshell/translations/extract_settings_index.py and stage the result" >&2; exit 1; fi'
DMS is an open-source, MIT-licensed desktop shell for Wayland compositors on linux. Like KWin, gnome-shell, or cosmic-shell - but designed to work with many different compositors.
## Repo Structure:
There are 2 main apps in the repo, they are designed to operate together. Additionally there is a dank-qml-common submodule which contains common QML widgets, like a shared library, that is leveraged by other quickshell apps in the Dank Desktop suite.
There is a GO module (dankgo) that is used by core for certain common go functionality.
### 1- core:
Contains 2 apps:
1. cmd/dms - The parent process for quickshell (shell runner), and general purpose tools and utilities (such as screenshot, clipboard). It also runs the unix socket server the UI uses for functionality quickshell doesn't provide.
2. cmd/dankinstall - An entirely separate app, a TUI for quickly installing DMS
### 2- quickshell:
A quickshell application, the entirety of the UI for DMS. It uses quickshells own services supplemented by the core unix socket server for additional functionality.
- Follow each apps' own conventions. For example, QML widgets use Theme tokens instead of hardcoding colors, spacing, or other constants.
- Resource usage is extremely important to DMS. The shell runs 24/7 on every machine it's installed on - audit any change for idle CPU cost, extra processes, timers, and retained memory.
-`core` and `quickshell` are tightly coupled through the unix socket protocol. When you are changing one, be mindful of the consequences for the other.
- Use the Dank* wrappers in Widgets/ (DankListView, DankFlickable, and so on) instead of raw ListView/Flickable/ScrollView.
- All user-facing text goes through I18n.tr(). Prefer reusing existing catalog terms over adding new ones. Never edit the translation catalogs by hand, they are synced with POEditor.
- Do not start editing code in response to a question. We'll tell you when to edit code.
- Do not leave paragraphs of comments on top of the code. You should try to avoid them as much as possible with understandable function names and code. If they are necessary even then, make them concise. Remove such comments when you come by them in the codebase. Comments should always move with code, not be left behind.
- Use guard statement patterns in any code you write.
- Do not write any useless tests, tests should cover input output validation - not useless things like "main.go contains func main()"
- Do not edit generated code directly, mocks are generated with mockery. Not edited by hand
- If we are missing a glaring issue when we ask you to do something, do not hesitate to point it out.
- Reinvent the wheel but do not reinvent the car. If you are solving a simple problem do not introduce a library. If you are solving a complex but a common problem, there is likely a modern library for it, if so, use it.
- Never commit or push code unless explicitly asked to do so.
- Never make a PR unless explicitly asked to do so.
## Commit Messages
Commit messages start with the part of the system they touched, followed by a short lowercase explanation of the work:
network: report wifi state from the associated adapter
doctor: flag config files that fail to parse
The title should be concise. Description should explain the work in more detail (only if required) while still being concise. Use simple language, do not try to sound smart.
This file is more of a quick reference so I know what to account for before next releases.
# Next
- Go core migrated to dankgo (shared log/paths/errdefs/ipc/shellapp)
- Quickshell UI embedded in the dms binary, -c / DMS_SHELL_DIR override - breaking for setups relying on ~/.config/quickshell/dms auto-discovery (XDG search removed)
- release binaries + all -git packages build with withshell; sockets/pidfiles/env names unchanged, IPC APIVersion 28 wire-compatible
- -git packages (AUR dms-shell-git, fedora, opensuse, void, debian, ubuntu) no longer install /usr/share/quickshell/dms; dms greeter install falls back to the embedded UI when the greeter package is present
- ping over the socket returns {"pong":true} instead of "pong"
- startup: embedded UI resolution is keyed by build-time .dankrev (no per-start content hashing)
# 1.5.0
- Overhauled shadows
- App ID changed to com.danklinux.dms - breaking for window rules
@@ -6,6 +6,20 @@ To contribute fork this repository, make your changes, and open a pull request.
## Setup
Clone with submodules — the shared widget library ([dank-qml-common](https://github.com/AvengeMedia/dank-qml-common)) is vendored at `dank-qml-common/` and symlinked into `quickshell/DankCommon`:
To have `git pull` keep the submodule in sync automatically (moving it to the commit this repo points at, no separate `git submodule update` step), set:
```bash
git config submodule.recurse true
```
Install [prek](https://prek.j178.dev/) then activate pre-commit hooks:
```bash
@@ -22,12 +36,53 @@ nix develop
This will provide:
- Go 1.24 toolchain (go, gopls, delve, go-tools) and GNU Make
- Go 1.25+ toolchain (go, gopls, delve, go-tools) and GNU Make
- Quickshell and required QML packages
- Properly configured QML2_IMPORT_PATH
The dev shell automatically creates the `.qmlls.ini` file in the `quickshell/` directory.
## Building and running
The Quickshell UI is embedded into the `dms` binary at build time. `make build` copies `quickshell/` into `core/internal/shellembed/dist/` (generated, never committed) and compiles with the `withshell` tag. `make dev` builds without the tag — that binary carries no UI and requires an explicit config dir.
```bash
make build # embedded binary at core/bin/dms
make dev # untagged development build
make run # dev build, then launch against the live quickshell/ tree
```
The UI config dir resolves in order: `-c <dir>`, `DMS_SHELL_DIR`, the dir a running instance is using, then the embedded UI. Each candidate must contain `shell.qml`. `make run` uses `-c $(pwd)/quickshell`, so QML edits hot-reload from the working tree.
The Go core depends on [dankgo](https://github.com/AvengeMedia/dankgo) for logging, XDG paths, the IPC transport, and the quickshell process lifecycle. To develop against a local dankgo checkout, create a gitignored `go.work` at the repo root:
```
go 1.26.1
use (
./core
../dankgo
)
```
## Shared widgets (dank-qml-common)
Everything under `quickshell/DankCommon/` (core widgets, the file browser, scroll physics, bundled fonts) is shared across the DMS suite and lives in the `dank-qml-common` submodule. It is a normal git worktree:
1. Edit files under `dank-qml-common/` (or through the `quickshell/DankCommon` symlink — same files) and test in the running shell; hot reload works as usual. For isolated widget work, the library is its own runnable config with a gallery: `qs -c dank-qml-common`.
2. Commit and PR those changes in the `dank-qml-common` repo: `cd dank-qml-common && git switch -c my-change`, push, open the PR there.
3. Once merged, bump the pointer here: `make update-common` (updates the submodule and the nix flake input together), then commit alongside any DMS-side changes. If you only bump the submodule, CI syncs `flake.lock` to it automatically on master.
The submodule URL in `.gitmodules` is HTTPS so CI and anonymous clones keep working. To push over SSH instead of being prompted for credentials, add a push rewrite to your git config — fetches stay HTTPS, pushes use SSH:
Shared widgets read app-provided singletons (`Theme`, `SettingsData`, ...) through a documented contract — see the dank-qml-common README. If your change needs a new contract property, add it to the library's stub singletons in the same PR, then to `quickshell/Common/` here when you bump.
Files in `quickshell/Widgets/`, `quickshell/Common/`, and `quickshell/Modals/FileBrowser/` that moved to the library remain in place as thin wrappers, so `import qs.Widgets`, `qs.Common`, and `qs.Modals.FileBrowser` keep working for the shell and for plugins.
## VSCode Setup
This is a monorepo, the easiest thing to do is to open an editor in either `quickshell`, `core`, or both depending on which part of the project you are working on.
@@ -86,7 +141,9 @@ touch .qmlls.ini
4. Restart dms to generate the `.qmlls.ini` file
5.Make your changes, test, and open a pull request.
5.Run `make lint-qml` from the repo root to lint QML entrypoints (requires the `.qmlls.ini` generated above). The script needs the **Qt 6**`qmllint`; it checks `qmllint6`, Fedora's `qmllint-qt6`, `/usr/lib/qt6/bin/qmllint`, then `qmllint` in `PATH`. If your Qt 6 binary lives elsewhere, set `QMLLINT=/path/to/qmllint`.
6. Make your changes, test, and open a pull request.
### I18n/Localization
@@ -102,6 +159,10 @@ Text {
Preferably, try to keep new terms to a minimum and re-use existing terms where possible. See `quickshell/translations/en.json` for the list of existing terms. (This isn't always possible obviously, but instead of using `Auto-connect` you would use `Autoconnect` since it's already translated)
Don't re-extract the translations. `en.json` and `template.json` are synced with POEditor by a maintainer script, so running `extract_translations.py` in your PR just makes a diff that fights the next sync. Add your `I18n.tr()` calls and leave the catalogs alone. (`settings_search_index.json` is the exception, the pre-commit hook regenerates that one when you touch settings QML.)
Strings inside `quickshell/DankCommon/` are owned by the dank-qml-common repo but stay in the DMS POEditor project — extraction here deliberately skips them, and `scripts/i18nsync.py sync` uploads the union of app terms and the submodule's terms instead (common terms carry the `dank-qml-common` tag). On download the sync splits the exports: app translations go to `quickshell/translations/poexports/`, common translations go to `dank-qml-common/DankCommon/translations/poexports/` for you to commit in that repo and bump. At runtime `I18n` merges both catalogs (app terms win). Other apps (dankcalendar) keep their own POEditor projects and merge the `dank-qml-common`-tagged terms from the DMS project.
### GO (`core` directory)
1. Install the [Go Extension](https://code.visualstudio.com/docs/languages/go)
@@ -110,6 +171,21 @@ Preferably, try to keep new terms to a minimum and re-use existing terms where p
4. Run `go mod tidy`
5. Open pull request
golangci-lint runs as a pre-commit hook and covers `go vet`, so there's nothing separate to run. If you run `go vet ./...` yourself you'll see complaints about the generated mocks, ignore them, `core/.golangci.yml` excludes that directory.
#### Mocks
Test mocks under `core/internal/mocks/` are generated with [mockery](https://vektra.github.io/mockery/) v3. Don't edit them by hand, regenerate after changing a mocked interface:
```bash
cd core
go run github.com/vektra/mockery/v3@latest
# or with it installed (go install github.com/vektra/mockery/v3@latest):
mockery
```
`core/.mockery.yml` lists every mocked interface and where its mock goes (e.g. `network.Backend` -> `internal/mocks/network/mock_Backend.go`). To mock a new interface, add it there under its package and regenerate.
## Pull request
Include screenshots/video if applicable in your pull request if applicable, to visualize what your change is affecting.
[>)](https://aur.archlinux.org/packages/dms-shell-git)
@@ -83,7 +83,7 @@ One command installs DMS and all dependencies on Arch, Fedora, Debian, Ubuntu, o
Wallpaper-based color schemes that automatically theme GTK, Qt, terminals, editors (vscode, vscodium), and more using [matugen](https://github.com/InioX/matugen) and dank16.
**System Monitoring**
Real-time CPU, RAM, GPU metrics and temperatures with [dgop](https://github.com/AvengeMedia/dgop). Process list with search and management.
Real-time CPU, RAM, GPU metrics and temperatures built into the dms daemon (powered by [dgop](https://github.com/AvengeMedia/dgop)). Process list with search and management.
**Powerful Launcher**
Spotlight-style search for applications, files ([dsearch](https://github.com/AvengeMedia/danksearch)), emojis, running windows, calculator, and commands. Extensible with plugins.
@@ -98,10 +98,12 @@ Notification center with grouping, rich text support, and keyboard navigation.
MPRIS player controls, calendar sync, weather widgets, and clipboard history with image previews.
**Session Management**
Lock screen, idle detection, auto-lock/suspend with separate AC/battery settings, and greeter support.
Lock screen, idle detection, auto-lock/suspend with separate AC/battery settings, and a settings front-end for [dank-greeter](https://github.com/AvengeMedia/dank-greeter).
**Plugin System**
Extend functionality with the [plugin registry](https://plugins.danklinux.com).
Extend functionality with the [plugin registry](https://plugins.danklinux.com). DMS keeps
`~/.config/DankMaterialShell/plugins.lock.json` synchronized with managed plugin installs and
their exact Git commits, so the same plugins can be reproduced on another machine.
DMS is one piece of the suite. The rest lives in its own repos:
- [dank-greeter](https://github.com/AvengeMedia/dank-greeter) - greetd login screen with the Dank Material aesthetic. The Greeter tab in DMS settings is the front-end for it.
- [dankcalendar](https://github.com/AvengeMedia/dankcalendar) - Local, Google, Microsoft, and CalDAV calendars for the dank desktop.
- [dgop](https://github.com/AvengeMedia/dgop) - System monitoring TUI and Go library; its library powers the process list and dashboard widgets inside the dms daemon.
- [dsearch](https://github.com/AvengeMedia/danksearch) - Fast file search that powers file results in the launcher.
- [dank-qml-common](https://github.com/AvengeMedia/dank-qml-common) - Shared QML widgets and components used by DMS, dank-greeter, and dankcalendar.
- [dankgo](https://github.com/AvengeMedia/dankgo) - Common Go modules behind the single binary apps.
## Development
See component-specific documentation:
@@ -188,7 +203,7 @@ For documentation contributions, see [DankLinux-Docs](https://github.com/AvengeM
## Star History
[](https://www.star-history.com/#AvengeMedia/DankMaterialShell&type=date&legend=top-left)
[](https://star-history.dera.page/#AvengeMedia/DankMaterialShell&type=date&legend=top-left)
@@ -10,7 +10,7 @@ Go-based backend for DankMaterialShell providing system integration, IPC, and in
Command-line interface and daemon for shell management and system control.
**dankinstall**
Distribution-aware installer with TUI for deploying DMS and compositor configurations on Arch, Fedora, Debian, Ubuntu, openSUSE, and Gentoo.
Distribution-aware installer for deploying DMS and compositor configurations on Arch, Fedora, Debian, Ubuntu, openSUSE, Gentoo, and Void. Supports both an interactive TUI and a headless (unattended) mode via CLI flags.
## System Integration
@@ -74,11 +74,11 @@ Custom IPC via unix socket (JSON API) for shell communication.
-`dms run [-d]` - Start shell (optionally as daemon)
-`dms greeter` - Deprecated; forwards to the standalone [dms-greeter](https://github.com/AvengeMedia/dank-greeter) binary
### Color Picker
@@ -96,7 +96,7 @@ The on-screen preview displays the selected format. JSON output includes hex, RG
## Building
Requires Go 1.24+
Requires Go 1.25+
**Development build:**
@@ -109,7 +109,7 @@ make test # Run tests
**Distribution build:**
```bash
make dist # Build without update/greeter features
make dist # Build without update features
```
Produces `bin/dms-linux-amd64` and `bin/dms-linux-arm64`
@@ -132,9 +132,11 @@ This runs gofmt, golangci-lint, tests, and builds before each commit when `core/
**Regenerating Wayland Protocol Bindings:**
Protocol XML files live in `internal/proto/xml/`. Bindings are generated with the [dankgo](https://github.com/AvengeMedia/dankgo) `go-wayland-scanner`:
```bash
go install github.com/rajveermalviya/go-wayland/cmd/go-wayland-scanner@latest
Headless mode requires cached credentials or a passwordless rule for your privilege escalation tool (sudo, doas, or run0). With sudo, run `sudo -v` first:
| `--exclude-deps <name,...>` | | Skip specific dependencies |
| `--replace-configs <name,...>` | | Replace specific configuration files (mutually exclusive with `--replace-configs-all`) |
| `--replace-configs-all` | | Replace all configuration files (mutually exclusive with `--replace-configs`) |
| `--yes` | `-y` | Required for headless mode - confirms installation without interactive prompts |
Headless mode requires `--yes` to proceed; without it, the installer exits with an error.
Configuration files are not replaced by default unless `--replace-configs` or `--replace-configs-all` is specified.
Optional packages (`dms-greeter`, `danksearch`, `dankcalendar`) are disabled by default; enable them with their dedicated flags, `--include-deps`, or `--all-features`.
When no flags are provided, `dankinstall` launches the interactive TUI.
### Headless mode validation rules
Headless mode activates when `--compositor` or `--term` is provided.
- Both `--compositor` and `--term` are required; providing only one results in an error.
- Headless-only flags (`--privesc`, `--git`, `--git-deps`, `--all-features`, `--include-deps`, `--exclude-deps`, `--replace-configs`, `--replace-configs-all`, `--yes`, `--danksearch`, `--dankcalendar`, `--dms-greeter`) are rejected in TUI mode.
- Positional arguments are not accepted.
### Log file location
`dankinstall` writes logs to `/tmp` by default.
Set the `DANKINSTALL_LOG_DIR` environment variable to override the log directory.
Short:"Generate the lock-screen PAM config from the system auth stack",
Long:"Resolve the distribution's PAM auth stack into a self-contained lock-screen config under the user state directory.\n"+
"Runs unprivileged (reads /etc/pam.d, writes to the user's state dir) and is used by the shell as a fallback when /etc/pam.d/dankshell is not managed.\n"+
Short:"List candidate lock-screen PAM services available on this system",
Long:"Enumerate the lock-screen PAM services that exist on this system and report their resolved auth stack (whether it has an auth directive and whether fingerprint/U2F modules appear inline).",
Short:"Validate a PAM service file for use by the DMS lock screen",
Long:"Validate one PAM service (by --service NAME or --path /abs/file) for use as the DMS lock-screen password or dedicated U2F stack. Exits 1 when the file is not usable.",
Run:func(cmd*cobra.Command,args[]string){
path,_:=cmd.Flags().GetString("path")
service,_:=cmd.Flags().GetString("service")
purpose,_:=cmd.Flags().GetString("purpose")
asJSON,_:=cmd.Flags().GetBool("json")
if(path=="")==(service==""){
log.Fatalf("Error: exactly one of --path or --service is required")
}
ifpurpose!="password"&&purpose!="u2f"{
log.Fatalf("Error: --purpose must be password or u2f")
Long:"Install greetd and configure it to use DMS as the greeter interface",
Run:func(cmd*cobra.Command,args[]string){
iferr:=installGreeter();err!=nil{
log.Fatalf("Error installing greeter: %v",err)
Use:"greeter",
Short:"Deprecated: moved to the standalone dms-greeter binary",
Long:"The greeter has moved to the standalone dms-greeter package.\nThis command forwards to 'dms-greeter' when it is installed.",
DisableFlagParsing:true,
SilenceUsage:true,
RunE:func(_*cobra.Command,args[]string)error{
binary,err:=exec.LookPath("dms-greeter")
iferr!=nil{
returnfmt.Errorf("'dms greeter' has moved to the standalone dms-greeter package.\n %s",greeterPackageInstallHint())
}
ifisLegacyWrapperScript(binary){
returnfmt.Errorf("'dms greeter' has moved to the standalone dms-greeter package; %s is the old wrapper script.\n %s",binary,greeterPackageInstallHint())
}
fmt.Fprintln(os.Stderr,"warning: 'dms greeter' is deprecated; use 'dms-greeter' directly")
@@ -46,18 +47,30 @@ var keybindsSetCmd = &cobra.Command{
Short:"Set a keybind override",
Long:"Create or update a keybind override for the specified provider",
Args:cobra.ExactArgs(3),
Run:runKeybindsSet,
Run:runKeybindsEdit,
}
varkeybindsRemoveCmd=&cobra.Command{
Use:"remove <provider> <key>",
Short:"Remove a keybind override",
Long:"Remove a keybind override from the specified provider",
Short:"Remove a keybind",
Long:"Remove a keybind. For Hyprland this writes a negative override to dms/binds-user.lua so the key stays unbound across DMS updates. For other providers it deletes the entry from the managed file.",
Args:cobra.ExactArgs(2),
Run:runKeybindsRemove,
Run:runKeybindsEdit,
}
varkeybindsResetCmd=&cobra.Command{
Use:"reset <provider> <key>",
Short:"Reset a keybind override to its DMS default",
Long:"Drop the user override for the given key so the DMS default re-applies. For providers without a separate default file (Niri, MangoWC) this is equivalent to remove.",
Args:cobra.ExactArgs(2),
Run:runKeybindsEdit,
}
funcinit(){
keybindsCmd.PersistentFlags().String("expected-generation","","Configuration generation observed when editing Aqueous bindings")
Long:"Manage the registries DMS fetches plugins and themes from. The official registry is always active; additional registries can be added by name and git URL.",
}
varregistryListCmd=&cobra.Command{
Use:"list",
Short:"List configured registries",
Run:func(cmd*cobra.Command,args[]string){
for_,s:=rangeregistries.Load(afero.NewOsFs()){
suffix:=""
ifs.Official(){
suffix=" (official)"
}
fmt.Printf("%s%s\n %s\n",s.Name,suffix,s.URL)
}
},
}
varregistryAddCmd=&cobra.Command{
Use:"add <name> <url>",
Short:"Add a registry",
Long:"Add a registry by name and git URL. The repository must contain a plugins/ or themes/ directory in the registry format.",
log.Debug("Using config from active session state file: %s",statePath)
configPath=statePath
log.Debug("Using config from: %s",configPath)
returnnil
}
os.Remove(configStateFile)
}
}
log.Debug("No custom path or active session, searching default XDG locations...")
varerrerror
configPath,err=config.LocateDMSConfig()
iferr!=nil{
returnerr
}
log.Debug("Using config from: %s",configPath)
returnnil
rootCmd.PersistentFlags().StringVarP(shellApp.CustomConfigVar(),"config","c","","Path to a UI config dir (containing shell.qml) to use instead of the embedded UI (env: DMS_SHELL_DIR)")
@@ -91,10 +124,33 @@ If no previous region exists, falls back to interactive selection.`,
varssWindowCmd=&cobra.Command{
Use:"window",
Short:"Capture the focused window",
Long:`Capture the currently focused window. Supported on Hyprland and DWL.`,
Long:`Capture the currently focused window. Supported on Hyprland, Mango, niri, and Aqueous. Aqueous requires a running DMS shell and crops the output including borders and overlapping windows.`,
Run:runScreenshotWindow,
}
varssScrollIntervalint
varssScrollCmd=&cobra.Command{
Use:"scroll",
Short:"Capture a scrolling region stitched into one tall image",
Long:`Select a region, then scroll the content beneath with the mouse wheel or
touchpad while frames are captured and stitched vertically. Finish with the
on-screen done button; cancel with the cancel button. Enter and Esc work
everywhere: most compositors hold the keyboard on the overlay (keyboard
scrolling does not reach the app there), while Hyprland leaves the keyboard
with the application — keyboard scrolling works, and Enter/Esc act through
temporary global binds for the session. The cursor is never included in
frames.
Frames are stitched continuously while scrolling, and revisited content is
never duplicated — scrolling up past the starting point extends the image
upward. Content jumped past faster than capture can follow is skipped rather
than stitched incorrectly.
Rotated outputs are not supported.`,
Run:runScreenshotScroll,
}
varssListCmd=&cobra.Command{
Use:"list",
Short:"List available outputs",
@@ -105,13 +161,14 @@ var notifyActionCmd = &cobra.Command{
Use:"notify-action",
Hidden:true,
Run:func(cmd*cobra.Command,args[]string){
screenshot.RunNotifyActionListener(args)
notify.RunActionListener(args)
},
}
funcinit(){
screenshotCmd.PersistentFlags().StringVarP(&ssOutputName,"output","o","","Output name for 'output' mode")
screenshotCmd.PersistentFlags().BoolVar(&ssIncludeCursor,"cursor",false,"Include cursor in screenshot")
screenshotCmd.PersistentFlags().StringVar(&ssSeat,"seat","","Seat for Aqueous capture (required with multiple seats)")
screenshotCmd.PersistentFlags().StringVar(&ssCursor,"cursor","off","Include cursor in screenshot (on/off)")
screenshotCmd.PersistentFlags().StringVarP(&ssFormat,"format","f","png","Output format (png, jpg, ppm)")
screenshotCmd.PersistentFlags().BoolVar(&ssNoClipboard,"no-clipboard",false,"Don't copy to clipboard")
screenshotCmd.PersistentFlags().BoolVar(&ssNoFile,"no-file",false,"Don't save to file")
screenshotCmd.PersistentFlags().BoolVar(&ssNoNotify,"no-notify",false,"Don't show notification")
screenshotCmd.PersistentFlags().BoolVar(&ssNoConfirm,"no-confirm",false,"Region mode: capture on mouse release without Enter/Space confirmation")
screenshotCmd.PersistentFlags().BoolVar(&ssReset,"reset",false,"Reset saved last-region preselection before capturing")
screenshotCmd.PersistentFlags().BoolVar(&ssStdout,"stdout",false,"Output image to stdout (for piping to swappy, etc.)")
screenshotCmd.PersistentFlags().BoolVar(&ssJSON,"json",false,"Print capture metadata as JSON")
screenshotCmd.PersistentFlags().BoolVarP(&ssGeometry,"geometry","g",false,"Print selected region geometry (X,Y WxH) to stdout without capturing an image")
screenshotCmd.PersistentFlags().BoolVar(&ssAllowMulti,"allow-multiple",false,"Open a selector even when another one is already open")
screenshotCmd.PersistentFlags().StringVar(&ssHUD,"hud","auto","HUD overlay scale in region selector (auto, off/0, or integer scale 1-4)")
ssScrollCmd.Flags().IntVar(&ssScrollInterval,"interval",45,"Capture interval in milliseconds (30-1000)")
Loaded 100 of 1867 files, more files were not shown because too many files have changed in this diff.
Show more
Reference in new issue
Block a user
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.