load_local_machines_from_config() iterated a reference to the live
m_local_machines map while erase_local_machine() erased from it for
printers without access rights, invalidating the range-for iterator
(use-after-free on ++it). Iterate a copy instead, as the code did
before commit 5028a5000e.
* Call proper EXT variant of the function if framebuffer type is EXT
* Update glad with `GL_EXT_framebuffer_blit` and `GL_EXT_framebuffer_multisample` extensions support
Give previously-bare panels a keyed background so the existing dark-UI walk
can theme them. Runtime-created widgets (HMS items, device firmware/nozzle
panels) route their whole subtree through one UpdateDarkUIWin(this) call
instead of per-widget darkModeColorFor, which also themes their child text
and follows live light/dark switches. Extruder-card chips re-apply their
colours from Sidebar::sys_color_changed so a live switch updates them too.
The stored plug-in identity is now the AA.BB.CC series (matching BambuStudio)
rather than the full build version, so the meaningless 4th build digit stops
driving config, the whitelist, filenames, and the version selector. This fixes
the macOS "can't switch to an older build" bug: the cloud endpoint is
series-keyed and only ever serves a series' newest build, so downgrade-by-
download was impossible and the download silently adopted the latest.
A startup migration normalizes an existing full-version config and file name to
the series form with no re-download, the loader resolves a bare series to the
newest same-series build on disk, and user-provided custom-named plug-ins
(libbambu_networking_02.08.01_custom.*) are still enumerated and loaded.
Discovered builds were inserted positionally behind the whitelist entry they
anchored to, so an OTA-installed 02.08.01.53 listed below 02.08.01.52. The list
is now built by appending and sorting once, which also drops the per-entry
insertion scans.
"(installed)" tested whether the library was present on disk, so it marked every
version ever selected - switching leaves the previous library in place. The flag
is now is_loaded, resolved against the plug-in actually loaded, and the two
combo populators share one label helper.
restart_networking() rebuilds m_agent with a null printer agent, so the bare
m_agent->start_discovery() no-oped (NetworkAgent::start_discovery returns false when
m_printer_agent is null) and LAN discovery stayed dead until the user next changed a
preset. Call switch_printer_agent() instead - it installs the printer agent for the
active preset and then starts discovery, mirroring startup.
When the network plugin is not yet installed at startup, on_init_network builds
the DeviceManager without an agent, so its constructor skips loading the persisted
LAN printers. After the plugin is installed and the network stack hot-reloads, the
manager is reused via set_agent(), which never loaded them - so previously paired
printers stayed missing from the device list until an app restart. Load them once a
real agent first arrives.
- Allow mapping to EMPTY trays only from the multi-machine send page and keep
both panels pickable in the LEFT_AND_RIGHT view
- Guard the error-dialog cloud snapshot against stale callbacks and fall back
to the local illustration on timeout
- Parse the ipcam storage-check ack and axis/chamber pushes defensively
- Strip fan-control telemetry, initialize the upgrade error code, restore the
.json filter in the model-id scan
- Fix best-position popup tray lookup, gradient placement, and colour-list
ownership
- Cover switch binding sets and invalid-track transients in DevMapping tests
Trim the version whitelist to the latest series plus the pinned legacy build,
and reject out-of-series configured versions at startup, compatibility check,
and load failure - falling back to the latest installed build or the clean
re-download flow so the config never keeps pointing at an unsupported build.
Adapt the speed dial's ActionRegistry to the collapsed
get_plugin_capability(PluginCapabilityId) overload, and restore the
script success/skipped status message the dialog lost when its
PluginScriptRunner refactor was superseded by ActionRegistry.
The preset option was plugin_preference_overrides while the GUI field type
that renders it was GUIType::plugin_config, for one and the same thing.
Settle on "config": the store, the dialog and the Python hooks all say config
already, so renaming that way touches 4 files instead of the whole plugin
subsystem and the public plugin API.
Keep the _overrides suffix — the option is the preset's override layer over
the base PluginConfig store, a distinction EffectiveCapabilityConfig tracks.
Also wrap the printer tab's group heading in L(); it was the only one of the
three missing it, and was therefore untranslatable.
Replace hand-rolled nozzle type comparison + Hybrid hack with
BBS-style NozzleGroupInfo comparison in check_ams_status_impl.
Previous approach: direct nozzle_volume_type == printer_flow_type
with a Hybrid tolerance lambda. This either suppressed the dialog
entirely (Hybrid always matched) or showed it on every Preview switch.
New approach (matching BBS):
- Build preset NozzleGroupInfo from extruder_nozzle_stats config
- For Hybrid presets: expand into per-type counts (Std#N, HF#M)
- nozzle_count==0 (never synced): dialog appears for first sync
- nozzle_count>0 (after sync): compare with printer GetNozzleGroups()
- Counts match → dialog suppressed on Prepare↔Preview switches
Safe for all multi-extruder printers (H2C, H2D): non-Hybrid presets
use single NozzleGroupInfo per extruder. Single-extruder printers
exit at is_multi_extruders() guard before reaching this code.
Reference to BBS: BambuStudio/src/slic3r/GUI/Plater.cpp
is_extruder_stat_synced() line 16642
The merge kept this branch's PluginConfig design, which deletes
PluginDescriptor::settings, get_plugin_settings() and ctx.params, but left
references to them behind: the slic3rutils target did not build, and the
bindings test still asserted the removed ctx.params attribute.
Port the two settings tests onto PluginConfig instead of dropping them. They
guard a field bug where a cloud-metadata refresh wiped a plugin's settings and
it silently ran on its own defaults, so the equivalent properties are still
worth pinning: that a stored config survives the refresh, and that an edited
config reaches the plugin through a real dispatch.
Also defer PluginsConfigDialog's web commands off the webview script-message
callback, as PluginsDialog already does. Its remove_preset_override handler put
a modal wxMessageBox on that stack, which is the GTK crash class fixed in
b779a7bfed/f2ccbfc8b5 for the sibling dialog.
