Address SoftFever review item 1 on #13752: CFS sync now lives in
CrealityPrintAgent::fetch_filament_info, matching the MoonrakerPrinterAgent
/ QidiPrinterAgent / SnapmakerPrinterAgent shape. The standard
Sidebar::sync_ams_list
-> DeviceManager::get_selected_machine()
-> Sidebar::load_ams_list(obj)
-> Sidebar::build_filament_ams_list(obj)
-> agent->fetch_filament_info(dev_id)
-> build_ams_payload(box_count, max_lane_index, trays)
path now applies to CrealityPrint hosts identically to the other
Klipper-flavoured agents. CrealityPrintAgent inherits
MoonrakerPrinterAgent, so the inherited connect_printer() ->
announce_printhost_device() -> SSDP callback already triggers
MachineObject creation in DeviceManager's localMachineList for
K-series LAN hosts; the Plater-side special case was simply bypassing
get_selected_machine() before it could fire.
Removed:
* The if (host_type == htCrealityPrint) block in
Sidebar::sync_ams_list (and the CrealityPrintAgent.hpp include it
required).
* The static CrealityPrintAgent::sync_filaments_into_ams_list()
helper that wrote directly to PresetBundle::filament_ams_list,
plus the CFSAmsListResult status struct that surfaced dialog text
back to the Sidebar.
CrealityPrintAgent::fetch_filament_info itself is unchanged - it was
already calling the inherited build_ams_payload() correctly.
Net diff: 227 deletions / 4 insertions across 3 files. No behaviour
change. Discovery + WS protocol parsing unchanged.
Root-cause + fix for the K2 Plus sparse-slot / wrong-slot-mapping bug
reported by multiple testers on Reddit and PR #13744 (DaviBe92, Psych0SW,
swilsonnc, TrainAss, Gullible-Price-4257).
DaviBe92 supplied a raw /box/getRealBoxesInfo payload from a K2 Plus
running printer FW 1.1.5.2 / CFS 1.2.2 that shows the K2 Plus uses
THREE state values for slots:
state: 0 → empty
state: 1 → loaded AND currently selected as the active spool
state: 2 → loaded but not currently selected
K2 (base) and K2 Pro firmwares — confirmed against the maintainer's
test printer — only use 0/1. Our parser had assumed the 0/1 form and
filtered with `if (state != 1) continue`, dropping every state=2 slot.
Symptoms this explains:
* DaviBe92: 3 spools loaded, only 1 displayed (the state=1 slot).
* Psych0SW / swilsonnc: 2 of 4 slots returned, "2nd shows 3rd's data"
— the parser dropped slots with state=2, leaving a sparse map that
PresetBundle::sync_ams_list then packs into consecutive UI trays.
* TrainAss / Gullible-Price-4257: "No loaded slots detected" — likely
the same root cause when zero slots happen to be state=1.
Fix: treat any non-zero state as loaded. Belt-and-braces: also skip
entries that are blanked-out (vendor and type both empty) regardless of
state, in case a future firmware uses yet another encoding for empty.
No change required for K2 / K2 Pro behaviour — they already only emit
state=0 (empty) or state=1 (loaded), and the new filter accepts both.
Per @SoftFever review on #13752, printer-specific filament sync logic
belongs in the agent rather than in Sidebar. This consolidates the
previously-duplicated code so all CFS-specific work lives in
CrealityPrintAgent.
Changes:
- New: CrealityPrintAgent::sync_filaments_into_ams_list() — static method
that builds a CrealityPrint host from a printer_cfg, queries CFS slots,
populates PresetBundle::filament_ams_list, and triggers sync_ams_list().
GUI-free; returns a result struct (Status + counts + detail) so the
caller decides what dialog to show.
- New: nested CFSAmsListResult struct describing the five possible
outcomes (Success / NotCfsCapable / QueryFailed / EmptySlots / NoMatches).
- Removed: Sidebar::sync_filaments_from_creality_cfs() entirely (its body
is now the agent method).
- Plater.hpp loses the declaration; Plater.cpp dispatches to the agent
inline within sync_ams_list() and owns only the dialog + post-sync UI
refresh (combo updates, layout, preset selection, persistence).
Two CFS-related entry points on the agent now coexist:
- fetch_filament_info() — agent-driven path; publishes via AmsTrayData
and build_ams_payload(). Active when a MachineObject is bound (BBL
concept, not currently created for Creality LAN hosts).
- sync_filaments_into_ams_list() — explicit-pull path used today by the
Sidebar's "Sync filaments" button until the K-series MachineObject
work catches up.
No user-visible behaviour change — same end-to-end flow, the data work
just lives in the agent now.
Per @SoftFever review on #13752, third-party vendored libraries belong in
deps_src/ alongside expat, imgui, hidapi, etc.
- All 5 files (mdns.{h,c}, cxmdns.{h,cpp}, NOTICE.md) move from
src/slic3r/Utils/mdns/ to deps_src/mdns/.
- deps_src/mdns/CMakeLists.txt builds mdns as a STATIC library and scopes
the MSVC Iphlpapi/Ws2_32 link requirement to that target instead of
libslic3r_gui's global MSVC block.
- deps_src/CMakeLists.txt gains add_subdirectory(mdns).
- src/slic3r/CMakeLists.txt drops the inline source listings and links
libslic3r_gui against the new mdns target; MSVC block keeps only
Setupapi.lib.
- src/slic3r/Utils/CrealityHostDiscovery.cpp #include updated to use the
include dir exposed by the new mdns target.
Verified by a clean Linux build (orca-slicer links successfully).
When a user adds a Creality K-series printer (host_type=crealityprint)
and clicks the existing "Browse" button in the Physical Printer dialog,
dispatch to a new CrealityDiscoveryDialog that finds K2 / K2 Plus /
K2 Pro printers on the LAN automatically. For other host types the
button keeps its existing BonjourDialog behaviour.
CrealityHostDiscovery (src/slic3r/Utils/CrealityHostDiscovery.{hpp,cpp})
Wraps the vendored cxmdns wrapper from the previous commit:
static std::vector<CrealityHost> scan(bool probe_info = true);
Calls cxnet::syncDiscoveryService({"Creality", "creality"}) to find
K-series printers via DNS-SD, dedupes by IP, then optionally HTTP
GETs http://<ip>/info on each match to fetch the model code
(F008 = K2 Plus, F012 = K2 Pro, F021 = K2) and MAC. Returns enriched
{ip, hostname, model_code, model_name, mac, cfs_capable} entries.
CrealityDiscoveryDialog (src/slic3r/GUI/CrealityDiscoveryDialog.{hpp,cpp})
Modal dialog with a wxListView showing Model / Hostname / IP per
discovered host. Runs CrealityHostDiscovery::scan() synchronously
with wxBusyCursor + wxWindowDisabler (5-10s total wait). User picks
one, dialog returns the IP via selected_ip().
PhysicalPrinterDialog (src/slic3r/GUI/PhysicalPrinterDialog.cpp)
The "Browse" button's click handler now reads host_type from the
edited config. If htCrealityPrint, opens CrealityDiscoveryDialog
and writes "http://<ip>" into the print_host field. Otherwise the
existing BonjourDialog path runs unchanged -- no behaviour change
for OctoPrint / Moonraker / Klipper users.
No new UI surface: one existing button now does the right thing per
host_type, mirroring how Creality Print discovers its own printers.
Drop a public-domain mDNS / DNS-SD lookup library into the tree at
src/slic3r/Utils/mdns/. Two pieces:
mdns.{h,c} -- public-domain library by Mattias Jansson from
https://github.com/mjansson/mdns
cxmdns.{h,cpp} -- C++ wrapper from CrealityOfficial/CrealityPrint
v7.1.1 (AGPL-3.0, compatible with OrcaSlicer's
AGPL-3.0)
cxmdns exposes one function:
std::vector<machine_info> cxnet::syncDiscoveryService(
const std::vector<std::string>& prefix);
It sends a DNS-SD meta-discovery query (_services._dns-sd._udp.local.),
listens ~5 seconds, and returns {ip, service_name} for every service
announcement whose name contains any of the given prefixes.
Motivation: Creality K-series firmware announces each printer under a
per-device-unique service type _Creality-<MAC-derived-hex>._udp.local.,
so OrcaSlicer's existing Bonjour code (which queries fixed service
names like _octoprint._tcp) cannot find them. The DNS-SD meta-browser
approach implemented here is the standard way to discover services
when you do not know their exact names in advance.
mdns.c calls GetAdaptersAddresses (iphlpapi) and Winsock2 functions,
neither of which were on libslic3r_gui's MSVC link line; both are
added here so the vendored sources compile and link standalone.
Attribution captured in src/slic3r/Utils/mdns/NOTICE.md. No callers
yet; the K-series discovery class + UI lands in the next commit.
The matcher tiebreaker previously preferred user-edited filament
presets over system bases on a tied score. For a K2 owner who has
a custom copy of Creality Generic PLA @K2-all called eg
Creality Hyper PLA @K2 (mine), the matcher scored both that copy
and the shipped brand-specific Hyper PLA @Creality K2 0.4 nozzle
at 30, then tiebreak picked the user copy. User copies inherit
filament_id from their parent -- in this case the generic PLA
GFL99 -- so the returned id pointed at Generic PLA, not at
Hyper PLA brand-specific id (01001). PresetBundle::sync_ams_list
then resolved by id back to Creality Generic PLA @K2-all, visibly
losing the brand on every sync.
Flip the tiebreaker to prefer system over user. The shipped
brand-specific preset always wins now and sync_ams_list lands on
the right slot label.
Drop the post-sync user-override step from the sidebar path that
was layered on to compensate -- silently substituting the user
local tuning is the wrong default for an upstream-shipped
feature; users who want their local tuning on a synced slot still
get to it via the existing combo dropdown.
The agent fetch_filament_info() path does not fire for Creality
K-series hosts because Sidebar::build_filament_ams_list()
short-circuits when no MachineObject is bound. MachineObject is a
BBL cloud-connected-printer concept that does not apply to LAN
Moonraker-style hosts like the K2 -- the AMS-sync icon click was a
no-op for them.
Mirror what Creality Print own slicer does (explicit Auto Mapping
button bypassing the BBL plumbing): when the user clicks the
existing AMS-sync icon and host_type=crealityprint, dispatch to a
new Sidebar::sync_filaments_from_creality_cfs() that reads the
active printer host config, confirms the printer is a CFS-capable
K-series board, queries boxsInfo over the printer WS on port 9999,
scores each loaded slot against the user filament presets and
builds a filament_ams_list entry with the matched filament_id,
colour and slot indices, then routes through
PresetBundle::sync_ams_list so the filament combo widgets get the
same rebuild as BBL printers and runs the BBL post-sync refresh
sequence (on_filament_count_change + combo update + select_preset
+ export_selections + update_dynamic_filament_list).
No new UI surface -- the existing AMS-sync icon does the right
thing per host_type. Match-and-resolve logic is hoisted out of the
agent anonymous namespace into public statics so the sidebar can
call it without duplicating scoring rules.
K-series printers (K2, K2 Plus, K2 Pro) ship with Mainsail on port
4408. Port 80 hosts only the Creality control/upload API, which
returns 404 for unknown paths and renders as a blank/404 page in
Orca Device tab.
Override CrealityPrint::get_print_host_webui() to default to
http://<host>:4408/ when the user has not explicitly set
print_host_webui, giving K-series owners a complete printer
dashboard in the Device tab out of the box.
Score visible compatible filament presets against the CFS spool
(vendor, brand_name, type) tuple to pick the right preset:
+20 preset name contains the brand_name as a substring
(eg Hyper PLA in Hyper PLA @Creality K2 0.4 nozzle)
+10 preset name contains the vendor substring (eg Creality)
Requires preset.filament_type to equal the spool base type so a
PETG preset is never auto-picked for a PLA spool. Falls back to
filaments.filament_id_by_type(base_type) when nothing scores.
Considers both base/system presets and user-derived copies -- K2
owners frequently keep tweaked copies of system presets (per-spool
PA, temps), so filtering to bases-only would skip exactly the
presets users care about most.
Subclass of MoonrakerPrinterAgent for K-series Creality printers
(K2, K2 Plus, K2 Pro) with CFS support. Registers in
NetworkAgentFactory under host_type=crealityprint. Overrides
fetch_filament_info() to defer to base when the host is not a
CFS-capable K-series board, otherwise query the boxsInfo WebSocket
on port 9999, parse the box hierarchy into CFSSlot[], and publish
each loaded slot as an AmsTrayData entry via build_ams_payload()
so they surface in Orca filament UI.
boxsInfo schema reference (verified against K2 Combo F021 firmware
v1.1.260206):
materialBoxs[].materials[] fields: id, state, vendor, type, name,
color, pressure, rfid, percent. state=1 means loaded; box.type=0
is a CFS unit, type=1 is the external spool holder (handled by
the upload dialog).
Accepted CFS boxes are renumbered sequentially since the K2 raw
box.id has gaps for the external spool holder. Model detection is
delegated to CrealityPrint::supports_multi_color_print() added in
PR #13291.
When the spool holder is selected for printing, use the opGcodeFile
command with enableSelfTest instead of colorMatch + multiColorPrint.
This matches the protocol used by CrealityPrint desktop for spool
holder prints.
- Detect spool holder mode: any colorMatch entry with box_id 0
- Spool holder: send opGcodeFile + enableSelfTest (single command)
- CFS: send colorMatch + multiColorPrint (existing behavior)
Signed-off-by: Igor Mammedov <niallain@gmail.com>
Add model_name() to map firmware model codes (F008, F012, F021)
to human-readable names (K2 Plus, K2 Pro, K2). Update the send
dialog init() to detect multi-color support and show a group box
with the printer name.
Signed-off-by: Igor Mammedov <niallain@gmail.com>
For supported models, start_print() sends colorMatch (filament-to-slot
mapping) followed by multiColorPrint (with optional calibration via
enableSelfTest) instead of the legacy opGcodeFile command. The dialog
passes color mapping and calibration settings through extended_info.
Signed-off-by: Igor Mammedov <niallain@gmail.com>
---
v4:
- Skip "path" form field in multipart upload for multi-color printers
(extra field breaks K2 Plus firmware parser)
- Use fire-and-forget ws.write() for colorMatch and multiColorPrint
instead of ws_send_and_read() (printer sends response asynchronously)
Add query_boxes_info() to discover loaded materials in the CFS
(Creality Filament System) via websocket. Returns the boxsInfo
JSON with slot details (type, color, vendor, temperature range).
Signed-off-by: Igor Mammedov <niallain@gmail.com>
Add a reusable websocket helper that sends a JSON command and loops
reads until finding a response containing the expected key. This
handles the printer's unsolicited status messages that arrive on
connect before the actual response. Returns empty string on timeout
instead of throwing.
Signed-off-by: Igor Mammedov <niallain@gmail.com>
Consolidate websocket connection setup into a reusable helper.
Migrate from tcp::socket to beast::tcp_stream for timeout support.
Set SO_RCVTIMEO read timeout (3s) and connect timeout (5s).
Refactor query_boxes_info() and start_print() to use ws_connect().
Signed-off-by: Igor Mammedov <niallain@gmail.com>
Parse the model field from the /info JSON response to enable
model-specific features. Add supports_multi_color_print() which
returns true for K2-platform printers (K2 Plus, K2 Pro, K2).
Signed-off-by: Igor Mammedov <niallain@gmail.com>
Change start_print() from void to bool with a wxString& msg
out-parameter, matching the error handling pattern used by Duet,
MKS, ESP3D and Flashforge. On failure, the error message is now
reported to the user via error_fn instead of being silently
swallowed. Use BOOST_LOG_TRIVIAL for logging instead of std::cerr.
Signed-off-by: Igor Mammedov <niallain@gmail.com>
Move duplicate get_host_from_url() implementations from ElegooLink and
OctoPrint into Http as a static method using the curl_url API. This
eliminates code duplication and provides a single reliable URL host
extraction utility for all print host implementations.
Signed-off-by: Igor Mammedov <niallain@gmail.com>
get_host_from_url() returns host:port which may cause
boost::asio::ip::make_address() to fail when a port is present,
bypassing the direct IP upload path and falling through to DNS
resolution via upload_inner_with_host(). Use get_host_from_url_no_port()
to extract just the host.
Signed-off-by: Igor Mammedov <niallain@gmail.com>
[How to Download Pull Requests Artifacts for
Testing](https://www.orcaslicer.com/wiki/how_to_download_pr_artifacts)
This can be enabled in Preference->Developer->Keep painted feature after
mesh change.
<img width="731" height="633" alt="init"
src="https://github.com/user-attachments/assets/8b195486-538e-4eda-9e77-bfdf1a794306"
/>
TODO:
- [ ] Bug fixes
- [ ] Make it faster
- Keep painting after other mesh operations such as reload from
disk/simplify/boolean operation etc:
- [x] Planar cut
- [x] Dovetail cut
- [x] Cut with part assigned to other side
- [x] Split to parts
- [x] Split to objects
- [x] Mesh boolean gizmo
- [x] Mesh boolean in right click
- [x] Reload from disk/replace stl (won't work well if mesh changed too
much)
- [x] Fix model
- [x] Simplify/smooth (this two won't work well due to too much mesh
changes)
- [x] Add options in settings since I think this will be experimental
for a long time until being tested by a lot of ppl
* Remove unused `get_printer_preset` calls.
* `Plater::check_printer_initialized` should skip nozzle flow type checks if nozzle is unknown.
Also remove unnecessary `is_multi_extruders` check.
* Remove unused `CalibUtils::is_same_nozzle_diameters()` fn.
* Simplify `CalibUtils::check_printable_status_before_cali`.
The single-arity can delegate to the vectorized arity.
* Add `DevExtruderSystem::NozzleDiameterMatchesOrUnknown` to simply checks.
* Update `CalibrationPresetPage::update_sync_button_status()` to use `NozzleDiameterMatchesOrUnknown`.
Simplify logic by iterating over each extruder and checking for diameter and volume type match.
The previous code had several mistakes (from what I could tell):
- `curr_obj->is_multi_extruders()` doesn't imply exactly 2 extruders
- the single/multi branch served no purpose
- the single branch failed to check the volume type
* Specify `std::fabs` and add explicit import.
Ref: https://github.com/OrcaSlicer/OrcaSlicer/pull/13330#discussion_r3133613736
Not sure how idiomatic this is in C++ / OrcaSlicer codebase, but CoPilot suggested it and it seems reasonable.
---------
Co-authored-by: SoftFever <softfeverever@gmail.com>
* Support file uploads and the device details page for CC2 printers.
* Resolved build issues for Linux and macOS.
* 1. Added `ElegooPrinterWebViewHandler` to handle WebUI messages for Elegoo printers. Other printers will keep the current behavior.
2. Added a static `get_print_host_webui` method in `PrintHost` to retrieve the printer WebUI URL.
* Improved timeout handling for CC2 file upload and SN info APIs.
---------
Co-authored-by: SoftFever <softfeverever@gmail.com>
Two related fixes for LAN printing with newer Bambu networking plugins:
1. FileTransferUtils: guard against missing ft_tunnel_* symbols
The eMMC tunnel constructor calls ft_tunnel_create / ft_tunnel_set_status_cb
without checking if the plugin exported them. Older plugins (e.g.
01.10.01.01) don't have these symbols, so the calls dereference null and
crash when sending a print. Now throw a clear exception so callers can
fall back gracefully.
2. PrintJob: disable eMMC print path by default
Plugin 02.03.00.62's eMMC tunnel code hangs indefinitely at the upload
phase (30%) on Bambu H2D, blocked inside the plugin waiting on a future
that never resolves. Cancel doesn't work because the plugin doesn't
check cancel_fn during the wait. The plain FTP path works reliably.
Default to try_emmc_print=false; opt-in via AppConfig setting
"disable_emmc_print" = "0". Also wrap the eMMC tunnel creation in
try/catch so a missing-plugin exception cleanly falls back to FTP
instead of killing the worker thread.
Co-authored-by: hyiger <hyiger@gmail.com>
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Co-authored-by: SoftFever <softfeverever@gmail.com>
* Add get_json_string_field helper
Introduce get_json_string_field in OrcaCloudServiceAgent.cpp to safely extract string fields from JSON objects.
* Add resolve_display_name helper
Introduce resolve_display_name to normalize provider metadata labels for the UI. The helper returns the first non-empty value from display_name, nickname, full_name, name, falling back to username, resolving human-facing label across varying provider payloads.
* Replace safe_str anonymous function
Replace get_json_string_field for better readability.
* Replace resolution flow for nickname with function
Consolidate both flows into one function for easier maintenance and more consistency.
* Update OrcaCloudServiceAgent.hpp
* Add OrcaCloudServiceAgent display name tests
Add unit tests verifying OrcaCloudServiceAgent resolves a user's display name from various session JSON shapes.
* Add OrcaCloud sync platform and preset bundle sharing system
Introduce OrcaCloud, a cloud sync platform for user presets, alongside
a preset bundle system that enables sharing printer/filament/process
profiles as local exportable bundles or subscribed cloud bundles.
OrcaCloud platform:
- Auth to Orca Cloud
- Encrypted token storage (file-based or system keychain)
- User preset sync with
- Profile migration from default/bambu folders on first login
- Homepage integration with entrance to cloud.orcaslicer.com
Preset bundles:
- Local bundle import/export with bundle_structure.json metadata
- Subscribed cloud bundles with version-based update checking
- Thread-safe concurrent bundle access with read-write mutex
- Canonical bundle preset naming (_local/<id>/... and _subscribed/<id>/...)
- Bundle presets are read-only; grouped under subheaders in combo boxes
- PresetBundleDialog with auto-sync toggle, refresh, update notifications
- Hyperlinked bundle names to cloud bundle pages
Co-authored-by: Sabriel Koh <sabrielkcr@gmail.com>
Co-authored-by: Derrick <derrick992110@gmail.com>
Co-authored-by: Mykola Nahirnyi <mnahirnyi@amcbridge.com>
Co-authored-by: Ian Chua <iancrb00@gmail.com>
Co-authored-by: Draginraptor <draginraptor@gmail.com>
Co-authored-by: ExPikaPaka <112851715+ExPikaPaka@users.noreply.github.com>
Co-authored-by: Ian Bassi <ian.bassi@outlook.com>
Co-authored-by: Ocraftyone <Ocraftyone@users.noreply.github.com>
Co-authored-by: yw4z <ywsyildiz@gmail.com>
Co-authored-by: peterm-m <101202951+peterm-m@users.noreply.github.com>
* Fixed an issue on Windows it failed to login Orca Cloud with Google account
Reduce the size of current and new JSONs by standardising them with 1 tab indentation instead of 4 spaces.
This effectively reduces the size by almost 20 MB.
| Current | New |
|---|---|
| 85.2 Mib | 67 Mib |
Used [JQ](https://jqlang.org/) `--tab` to automatically format every current JSON.
> [!NOTE]
> Some profiles had the arrays on the same line, but those created by Orca were in the standard format (each object below the previous one). In some cases, this increases the number of tabs due to the new lines, but the increase is negligible, and this way both the base profiles and those created by Orca maintain the same style.
* Add runtime display backend detection for Wayland support
Add LinuxDisplayBackend utility to detect X11 vs Wayland at runtime
using GDK_IS_X11_DISPLAY / GDK_IS_WAYLAND_DISPLAY macros. This is
the foundation for removing the forced GDK_BACKEND=x11 and enabling
native Wayland support.
- New files: LinuxDisplayBackend.hpp/.cpp with get_linux_display_backend(),
is_running_on_wayland(), and is_running_on_x11()
- Propagate wxHAVE_GDK_X11 / wxHAVE_GDK_WAYLAND from FindGTK3.cmake
as compile definitions to libslic3r_gui
- No-op on non-Linux platforms (returns Unknown / false)
* Fix Phase 1 code quality: pragma once, source ordering, static cache
* Make X11 initialization conditional for Wayland support
Remove the unconditional GDK_BACKEND=x11 force that blocked native
Wayland. Replace with conditional logic:
- EGL safety fallback: re-force X11 only when wxUSE_GLCANVAS_EGL is
off and WAYLAND_DISPLAY is set, with a warning log
- XInitThreads() only called when DISPLAY is set (X11 in use)
- __GLX_VENDOR_LIBRARY_NAME only set when DISPLAY is present (GLX-specific)
- WEBKIT_DISABLE_COMPOSITING_MODE only set under XWayland (both
DISPLAY and WAYLAND_DISPLAY present)
- Guard X11/Xlib.h include with __has_include for robustness
- Restore display validation to accept either DISPLAY or WAYLAND_DISPLAY
This is Phase 2 of the Wayland support plan.
* Fix Phase 2: safer EGL macro check, add clarifying comments
* Add GLAD2 library and replace GLEW linkage in build system
Set up GLAD2 as a static library to replace GLEW for OpenGL loading.
GLAD2 supports both GLX and EGL, which is required for Wayland support.
- Create src/glad/ with pre-generated GLAD2 sources (GL 4.6 compat)
- Add src/glad/CMakeLists.txt building glad as a static library
- Wire glad into src/CMakeLists.txt before libvgcode
- Modify libvgcode to use shared glad for GL path (keeps local copy
only for GLES2/Emscripten) to avoid duplicate symbol conflicts
- Replace GLEW::GLEW with glad in libslic3r_gui link libraries
Note: GLEW is kept in deps for OpenCSG. Code migration from GL/glew.h
to glad/gl.h headers will follow in Phase 3B+3C.
* Fix Phase 3A+3D: libvgcode GLAD include, dead files, dlopen dep, OpenGL link var
* Migrate from GLEW to GLAD: replace headers and API calls across codebase
Replace all #include <GL/glew.h> with <glad/gl.h> across 49 source files.
Migrate GLEW API calls to GLAD equivalents:
- glewInit/glewExperimental -> gladLoaderLoadGL()
- GLEW_EXT_* / GLEW_ARB_* extension checks -> GLAD_GL_EXT_* / GLAD_GL_ARB_*
- Remove GLEW-specific EGL/GLX mismatch #error guards (not needed with GLAD)
- Replace unavailable EXT symbols with core GL equivalents in
GLCanvas3D.cpp (GL_MAX_SAMPLES, glRenderbufferStorageMultisample,
glBlitFramebuffer, GL_READ/DRAW_FRAMEBUFFER)
- Update log messages from glewInit to gladLoadGL
* Fix Phase 3B+3C: remove GLEW find, clean EXT symbols, update attribution
- Remove find_package(GLEW) block from root CMakeLists.txt since GLEW
is no longer linked by any main application code
- Remove "glew" from SLIC3R_STATIC option description
- Replace all remaining EXT framebuffer symbols with core equivalents
in render_thumbnail_framebuffer_ext and _rectangular_selection_picking_pass
- Update AboutDialog credits from GLEW to GLAD
* Enable EGL in wxWidgets and add runtime GLX/EGL selection for Wayland
- Set wxUSE_GLCANVAS_EGL=ON in wxWidgets build and Flatpak manifest
- Add PreferGLX() call on X11 sessions for driver compatibility
- Remove Phase 2 safety fallback (EGL is now always compiled in)
- Guard SwapBuffers against hidden canvases to prevent Wayland stalls
* Fix Phase 4: move PreferGLX to app startup, fix FPS counter guard
Move wxGLCanvas::PreferGLX() from OpenGLManager::create_wxglcanvas()
(static initializer) to GUI_App::on_init_inner() before any wxGLCanvas
is constructed. This prevents a race where SkipPartCanvas could trigger
wxGLBackend::Init() before the GLX preference is set. The new location
also adds explicit is_running_on_wayland() detection with a warning for
unknown backends.
Move increment_fps_counter() inside the IsShownOnScreen() guard so FPS
is only counted when a frame is actually swapped.
* Update GLFW from 3.3.7 to 3.4 for runtime Wayland/X11 backend selection
Replace the compile-time GLFW_USE_WAYLAND flag (which locked to a single
backend) with GLFW 3.4's GLFW_BUILD_WAYLAND + GLFW_BUILD_X11 flags that
build both backends and auto-select at runtime based on the available
display server. This enables the CLI thumbnail renderer to work on both
Wayland and X11 sessions without separate builds.
* wayland: Fix UI call sites that rely on global screen coordinates
On Wayland, wxGetMousePosition() returns (0,0) and SetPosition() is a
no-op for top-level windows. Fix the highest-impact call sites:
- GLCanvas3D: Use cached m_mouse.position from event handlers instead
of wxGetMousePosition() + ScreenToClient() in get_local_mouse_position()
- Plater: Use event-relative coords via ClientToScreen(e.GetPosition())
instead of wxGetMousePosition() in 3 leave-window handlers
- BBLTopbar: Use event.GetPosition() and FindToolByPosition() directly
in mouse handlers instead of wxGetMousePosition()/FindToolByCurrentPosition()
- Search: Use focus-based dismiss logic on Wayland instead of
wxGetMousePosition()-based rect checks in SearchDialog and
SearchObjectDialog
- GUI_App: Skip SetPosition() in window_pos_restore() on Wayland where
it is a no-op; still restore size and maximize state
- Button: Position tooltip relative to button widget via ClientToScreen
instead of wxGetMousePosition()
* Fix SearchDialog Wayland dismiss: guard against search_line focus
* flatpak: Add Wayland socket permission for native Wayland support
* spec
* Fix crash on Wayland when wxWidgets lacks EGL support
Restore the safety fallback that forces GDK_BACKEND=x11 when wxWidgets
was not built with wxUSE_GLCANVAS_EGL=ON. Without this, the GLX backend
tries to access a non-existent X11 display on native Wayland, crashing
in wxGLCanvas::IsDisplaySupported() with SIGSEGV at offset 0xe4.
Also add a defense-in-depth guard in detect_multisample() that skips
the IsDisplaySupported call entirely on Wayland without EGL.
Root cause: deps/wxWidgets must be rebuilt after enabling EGL. The
compile-time check in OrcaSlicer.cpp detects the mismatch and falls
back safely.
* Fix EGL detection: use wxHAS_EGL instead of wxUSE_GLCANVAS_EGL
wxUSE_GLCANVAS_EGL is a CMake build option, NOT a C++ preprocessor
macro. The actual macro defined in wxWidgets setup.h is wxHAS_EGL.
All compile-time EGL checks were using the wrong macro, causing
the safety fallback to always trigger even with a properly built
EGL-enabled wxWidgets.
* Fix GL function pointers invalidated on Wayland/EGL
gladLoaderLoadGL() dlopen's libGL.so.1 to resolve GL function pointers
via dlsym, then immediately dlclose's the handle. On X11/GLX this is
fine because the GLX context keeps libGL.so mapped. On Wayland/EGL,
nothing else holds libGL.so open, so dlclose unmaps it and all function
pointers become dangling — causing SIGSEGV on the first GL call.
Fix: on Wayland, use gladLoadGL(eglGetProcAddress) which resolves
function pointers through the EGL loader without opening/closing
libGL.so.
* fix crash on start and various rendering issues
* fix crash on close
* small refactor
* move GPU selection to desktop file
* clean up a bit
* clean up more
* fix appimage error
The Qidi Max 4 doesn't report that it's a Qidi Max 4 via /server/info
like the other Qidi printers apparently do. This means that filament
matching is a fool's erand because you'll never actually match anything.
this adds a fallback where we just use the configured machine type. I'm
not sure why we wouldn't just do that all the time, but I wanted to
change as little as possible.
Happy Hare now writes mmu lane data to the moonraker db similar to AFC, so we can normalize to the one implementation. For now, i've just re-ordered it so that the moonraker db is checked first. I want this because I've actually updated the happy hare data to include more data than we had access to via the "mmu" object. For happy hare we now have access to vendor name, which we can eventually use to further fine tune the auto-matched filament preset.
# Description
Esthetic filaments such as wood-infused, matte and marble filaments will now be handled by the Snapmaker printer agent so that they could be matched to more appropriate print profiles instead of generic PLA.
