* Add safety checks and add logic to avoid creating degenerate polygons in contour_extrusion_path
* fix comment
* fix(zaa): cast from slice_z upward only, guard normal on miss
* remove old code
---------
Co-authored-by: Aleksandr Dobkin <alex@dobk.in>
Co-authored-by: Rodrigo Faselli <162915171+RF47@users.noreply.github.com>
* fix: manual filament change not emitting MANUAL_TOOL_CHANGE for BBL printers
BBL-specific M1020 S handling in toolchange() bypassed toolchange_prefix(),so manual_filament_change had no effect — M1020 was always emitted regardless of the setting. Move the BBL logic into toolchange_prefix() so all callers respect the option consistently.
Fixes#11795
* add comment
Fix extrusion of some support layers at wrong Z height
Fixes the issue that PR #12736 (github.com/OrcaSlicer/OrcaSlicer/pull/12736) reverted the changes of PR #13327 (github.com/OrcaSlicer/OrcaSlicer/pull/13327)
This PR supersedes #12225, which originally proposed this feature but
appears inactive.
The feature originated from work I developed earlier in
[BambuStudio-ZAA](https://github.com/adob/BambuStudio-ZAA), a private
fork of Bambu Studio
Compared to #12225, I updated the implementation for current upstream
and fixed the following issues:
- fixed broken tests
- removed references to nonplanar directory
Reviewers may want to compare against #12225 for earlier
discussion/context.
## Summary
Port of **Z Anti-Aliasing (ZAA)** from
[BambuStudio-ZAA](https://github.com/adob/BambuStudio-ZAA) to
OrcaSlicer.
ZAA eliminates visible stair-stepping on curved and sloped top surfaces
by raycasting each extrusion point against the original 3D mesh and
micro-adjusting its Z height to follow the actual surface geometry. The
result is visibly smoother domes, chamfers, and shallow slopes — without
post-processing.
## How It Works
1. The slicer runs normally, then a **posContouring** step processes
each layer
2. `ContourZ.cpp` raycasts every extrusion point vertically against the
source mesh
3. Each point's Z is adjusted to the mesh intersection, converting flat
`Polyline` paths into `Polyline3` paths with per-point Z coordinates
4. The G-code writer emits the adjusted Z values, so the printer follows
the true surface
## Configuration
Five new settings under **Print Settings > Quality**:
| Setting | Type | Default | Description |
|---------|------|---------|-------------|
| `zaa_enabled` | bool | off | Master enable/disable switch |
| `zaa_min_z` | float | 0.06 mm | Minimum Z layer height; controls
slicing plane offset |
| `zaa_minimize_perimeter_height` | float | 35° | Reduce perimeter
heights on slopes below this angle (0 = disabled) |
| `zaa_dont_alternate_fill_direction` | bool | off | Keep fill direction
consistent instead of alternating |
| `zaa_region_disable` | bool | off | Disable ZAA for a specific print
region/material |
## Key Changes
- **Core algorithm**: New `src/libslic3r/ContourZ.cpp` (~330 lines) —
raycasting engine
- **3D geometry**: `Point3`, `Line3`, `Polyline3`, `MultiPoint3` extend
existing 2D types
- **Arc fitting**: Templated to work with both 2D and 3D geometry
- **Pipeline**: `ExtrusionPath::polyline` changed from `Polyline` to
`Polyline3`; new `posContouring` step in `PrintObject.cpp`
- **G-code**: `GCode.cpp` writes per-point Z when `path.z_contoured` is
set
- **UI**: ZAA settings exposed in Print Settings > Quality panel
- **Documentation**: `docs/ZAA.md` with usage and implementation details
57 files changed, ~1800 insertions, ~200 deletions.
## Test Plan
- [ ] Load a model with curved top surfaces (sphere, dome, chamfered
box)
- [ ] Enable **Z contouring** in Print Settings > Quality
- [ ] Slice and verify G-code has varying Z values within contoured
layers
- [ ] Build on macOS (verified), test on Linux and Windows
`check_outside` (placement check) calls `ensure_imex_zones()` to make sure
zone geometry is current before deciding if an instance is in-bounds. In
the GUI path that's fine, but the CLI / headless 3MF-load path also reaches
this through `PartPlateList::load_from_3mf_structure -> reload_all_objects
-> add_instance -> check_outside`, and CLI mode has no GUI_App initialized.
`build_imex_cache_key` and `calc_imex_zones` both dereference
`wxGetApp().preset_bundle` — without a GUI_App, `wxGetApp()` returns memory
that segfaults on member access, killing the slicer with SIGSEGV before
any G-code is produced.
Latent since 461c69c83e (Apr 9), surfaced now that upstream's main carries
the headless regression-test CI step (#13353) that exercises CLI slicing
on every PR build.
Fix: short-circuit `ensure_imex_zones()` when `m_plater` is null (already
the GUI/CLI marker used by `calc_imex_ghosts`). Also tighten the existing
`build_imex_cache_key` null check to consult `m_plater` first as defence
in depth, so the function stays safe if reached from another headless
caller.
Verified locally against the upstream regression suite — klipper /
p1s_multicolor / toolchanger_4_color all slice cleanly within the
20% baseline tolerance.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
Noisyfox