### Problem
Waybar segfaults on startup when `niri/workspaces` is the only configured
module (default options). Right after the first `WorkspacesChanged` event:
Thread 1 "waybar" received signal SIGSEGV
#0 gtk_label_set_markup ()
#1 waybar::modules::niri::Workspaces::doUpdate () at src/modules/niri/workspaces.cpp:106
### Root cause
[doUpdate()](cci:1://file:///home/lj/Downloads/Waybar/src/modules/niri/window.cpp:27:0-90:1) set the label markup via:
static_cast<Gtk::Label*>(button.get_children()[0])->set_markup(name);
The button's child *is* a valid `GtkLabel`, but gtkmm's `get_children()`
returns it wrapped as a generic `Gtk::Widget` (confirmed: `dynamic_cast`
to `Gtk::Label*` yields `nullptr`). The unchecked `static_cast` then
performs an invalid downcast, producing a corrupt pointer whose `gobj()`
is garbage (`0x1`), so `gtk_label_set_markup()` dereferences it and crashes.
This is reliably triggered when no other module has instantiated a
`Gtk::Label` yet (so the `Gtk::Label` wrapper isn't registered), which is
exactly the case for a `niri/workspaces`-only bar. The same idiom exists in
`sway/workspaces` and `wayfire/workspaces`; it's masked there because typical
configs include other label-using modules.
### Testing
- Reproduced the crash on 0.15.0 / current `master` with a minimal
`"modules-left": ["niri/workspaces"]` config under niri.
- After the fix: no crash; three workspace buttons render with correct
names/labels (`niri-workspace-1/2/3`).
When Host::proxyReady fails to create the SnWatcher proxy (e.g. because
the Watcher has not finished exporting /StatusNotifierWatcher yet), the
cancellable is left set, causing nameAppeared to early-return on every
subsequent event (see the // TODO marker). The Host is then stuck
without a watcher, and the tray module reports
'No such object path /StatusNotifierWatcher' until the bar is fully
restarted.
Clear the cancellable on non-CANCELLED errors and schedule a single
delayed retry of nameAppeared. The guard `watcher_ != nullptr` skips
the retry if a parallel call already succeeded.
Closes#3468
Fix an issue in the MPD module where text failed to render
when raw tags containing markup characters were truncated
mid-sequence after sanitization,
resulting in fragmented XML entities (such as `&` cut into `&am`).
Resolve this by shifting the order of operations to
truncate the raw strings *before* running `sanitize_string`.
To keep `setLabel` decoupled and clean,
we introduce helper methods (`getArtistStr`, `getAlbumArtistStr`,
`getAlbumStr`, `getTitleStr`) inside the `MPD` class,
matching the architectural style of MPRIS.
Additionally:
- Introduce support for the customizable `ellipsis` config option
(defaulting to `…`).
- Switch the truncation logic of the MPD module to be
visual-width-aware using the newly extracted
`waybar::util::utf8_truncate`.
- Decouple the tooltip text from the main label's truncated text.
The tooltip now displays un-truncated, complete metadata
(properly escaped), providing a significantly better user
experience.
Fix an issue where the mpris module's tooltip failed to render
when track metadata contained unescaped XML/HTML markup characters.
This occurred because
the mpris module formatted raw strings into the tooltip template
before passing the final string directly to `set_tooltip_markup()`,
triggering GTK/Pango parsing warnings and rendering failures.
Resolve this by wrapping tooltip metadata fields
in `Glib::Markup::escape_text` before formatting,
ensuring Pango-compliant strings are always delivered to the GTK
tooltip markup.
Move the UTF-8 visual-width measurement and truncation helper functions
from the mpris module to a common utility
(include/util/utf8_string.hpp and src/util/utf8_string.cpp).
This decouples string truncation and width measurement
from the mpris module, allowing other modules (like mpd) to reuse.
The helper function `utf8_truncate` (formerly `truncate` in mpris)
and `utf8_width` is exported under the `waybar::util` namespace,
while the low-level `measure_and_truncate` (formerly `utf8_truncate`)
is encapsulated in an anonymous namespace in src/util/utf8_string.cpp
to avoid unnecessary API exposure.
No functional changes were made to the mpris module's behavior.
The handleScroll method unconditionally dispatched workspace cycling commands
and never checked for custom on-scroll-up/on-scroll-down config values.
Register the scroll handler when custom scroll commands are configured (even
without enable-bar-scroll), and delegate to AModule::handleScroll so user
commands are executed, matching the pattern used by other modules.
The taskbar currently requests enough width for every task title, which can push right-side modules off-screen when many windows are open. Allow task buttons to share the taskbar allocation and ellipsize title labels instead of forcing the whole module to grow.
This keeps neighboring modules visible while preserving the existing taskbar click actions and formatting behavior.
Co-authored-by: Cursor <cursoragent@cursor.com>
- Add "muted" class for styling the muted state.
- Add "zero-on-mute" boolean option to control slider position
when muted.
- Add "unmute-on-volume-change" boolean option to control whether
to automatically unmute when the volume changes.
devices) so transient events stop blanking layout_. Refresh layouts_map_
on "added"/"xkb_keymap" input events and union layouts across all
keyboards so new devices contribute their layouts. Release mutex_ before
the refresh sendCmd to avoid self-deadlock from the synchronous
signal_cmd emit.
Tray::update() already had logic to hide the module when all child items were passive, but it was only re-run on item add/remove — never on a status change. When the last visible item transitioned to Passive, the item hid itself but the now-empty module remained visible.
Connect Tray to each item's Gtk::EventBox signal_show/signal_hide so update() runs on every visibility transition, and simplify update() to check child->get_visible() directly instead of inspecting the `passive` CSS class. The Tray-level `show-passive-items` read becomes redundant since Item already honours it when deciding its own visibility; remove it along with the now-unused Tray::show_passive_ member.
Fixes: #3721
Hyprland 0.54 replaced the text-based dispatch socket protocol with a
Lua-based one. Commands like "dispatch workspace 1" are now interpreted
as invalid Lua (return hl.dispatch(workspace 1)), breaking workspace
clicks and scroll navigation.
Add IPC::dispatch() that probes the running Hyprland on first call and
routes commands through the new hl.dsp Lua API when the Lua protocol is
detected, falling back to the old text format otherwise.
Only delete the corresponding address type (IPv4 or IPv6) when an event
about a specific type (AF_INET or AF_INET6) is received
This fixes situations where only one type of the address is deleted (and
possibly added again) but Waybar still thinks the interface is in
"linked" (no IP) state.
When netlink events (link/addr/route changes) fire between timer
intervals, dp.emit() triggers update() which consumes the byte delta
and resets bandwidth_down_total_. A subsequent timer update sees
near-zero delta, displaying very small bandwidth.
Cache the last computed bandwidth values and skip recalculation
when update() is called within half the interval. Event-driven
updates reuse the cached values instead.
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
When an interface is not specified for the network module, we parse the
routing table to look for default routes.
We have defined a default route to:
- have a gateway specified, and
- have no destination specified, or have an all-zero destination.
Previous versions of Waybar had the second condition inverted,
causing it to incorrectly pick interfaces are used to route a
subnet/single host.
For example, with the following routing table, we should pick `eth0` to
show information about, not `wg0`.
```
ip -4 route
default via 192.168.252.1 dev eth0 proto dhcp src 192.168.252.200 metric 100
192.168.252.0/24 dev eth0 proto kernel scope link src 192.168.252.200 metric 100
192.168.2.254 via 192.168.1.1 dev wg0 proto static metric 50
192.168.1.0/24 dev wg0 proto static scope link metric 50
192.168.1.0/24 dev wg0 proto kernel scope link src 192.168.1.254 metric 50
```
The window module registers itself with the Hyprland IPC singleton at
the start of its constructor, before calling update(). If update()
throws an exception (e.g. from an invalid format string), the object is
destroyed without the destructor running, leaving a dangling pointer in
the IPC callback list. When the IPC thread receives an event, it
attempts to call onEvent() on this invalid memory, causing a crash.
Moving the update() call before IPC registration ensures that any
initialization errors occur before the pointer is shared. If the
configuration is invalid, the module fails to construct and is
gracefully disabled by the factory without leaving a "landmine" in the
background IPC thread.
Fixes: #4923
Signed-off-by: Emir Baha Yıldırım <jayshozie@gmail.com>
Jason Chiu