Fix of re-slicing with multiple regions.

This is a fix of a bug, which was in Slic3r forever, where raw slices
were not cached, but recalculated from classified regions, where
merging the regions did not produce the original contour reliably.
Fixes [2.3.0-beta2] Odd bad slicing related to infill (?) percentage #5407
This commit is contained in:
Vojtech Bubnik 2020-12-11 12:21:07 +01:00
parent cc7dbf8925
commit a2959ec944
5 changed files with 38 additions and 20 deletions

View file

@ -28,6 +28,10 @@ public:
// collection of surfaces generated by slicing the original geometry
// divided by type top/bottom/internal
SurfaceCollection slices;
// Backed up slices before they are split into top/bottom/internal.
// Only backed up for multi-region layers or layers with elephant foot compensation.
//FIXME Review whether not to simplify the code by keeping the raw_slices all the time.
ExPolygons raw_slices;
// collection of extrusion paths/loops filling gaps
// These fills are generated by the perimeter generator.
@ -125,8 +129,10 @@ public:
// Test whether whether there are any slices assigned to this layer.
bool empty() const;
void make_slices();
// Merge typed slices into untyped slices. This method is used to revert the effects of detect_surfaces_type() called for posPrepareInfill.
void merge_slices();
// Backup and restore raw sliced regions if needed.
//FIXME Review whether not to simplify the code by keeping the raw_slices all the time.
void backup_untyped_slices();
void restore_untyped_slices();
// Slices merged into islands, to be used by the elephant foot compensation to trim the individual surfaces with the shrunk merged slices.
ExPolygons merged(float offset) const;
template <class T> bool any_internal_region_slice_contains(const T &item) const {