viewit/OrcaSlicer-KX
5
4
Fork 1
Code Issues 2 Pull Requests Actions Packages Projects Releases 3 Wiki Activity

(author: ratdoux) support more than 16 colors in filament region painting data

Browse Source 
partial commit ported from OrcaSlicer-FullSpectrum to maintain data format compatibility. original commit hash: dd994574, "added support for >16 colors in color picker" by ratdoux
This commit is contained in:
sentientstardust
2026-05-03 01:36:06 +01:00
parent 9923664212
commit 5d93d333dd
2 changed files with 128 additions and 37 deletions
Download Patch File Download Diff File Expand all files Collapse all files

149
src/libslic3r/TriangleSelector.cpp
View File

@@ -699,11 +699,14 @@ int TriangleSelector::triangle_midpoint(const Triangle &tr, int vertexi, int ver
// If itriangle == -1 or if the side sharing (vertexi, vertexj) is not split, return -1.
int TriangleSelector::triangle_midpoint(int itriangle, int vertexi, int vertexj) const
{
return itriangle == -1 ? -1 : this->triangle_midpoint(m_triangles[itriangle], vertexi, vertexj);
return itriangle < 0 || itriangle >= int(m_triangles.size()) ? -1 : this->triangle_midpoint(m_triangles[itriangle], vertexi, vertexj);
}
int TriangleSelector::triangle_midpoint_or_allocate(int itriangle, int vertexi, int vertexj)
{
if (vertexi < 0 || vertexj < 0 || vertexi >= int(m_vertices.size()) || vertexj >= int(m_vertices.size()))
return -1;
int midpoint = this->triangle_midpoint(itriangle, vertexi, vertexj);
if (midpoint == -1) {
Vec3f c = 0.5f * (m_vertices[vertexi].v + m_vertices[vertexj].v);
@@ -1361,14 +1364,22 @@ void TriangleSelector::perform_split(int facet_idx, const Vec3i32 &neighbors, En
for (int j=0, idx = tr.special_side(); j<3; ++j, idx = next_idx_modulo(idx, 3))
verts_idxs.push_back(tr.verts_idxs[idx]);
auto get_alloc_vertex = [this, &neighbors, &verts_idxs](int edge, int i1, int i2) -> int {
return this->triangle_midpoint_or_allocate(neighbors(edge), verts_idxs[i1], verts_idxs[i2]);
bool split_failed = false;
auto get_alloc_vertex = [this, &neighbors, &verts_idxs, &split_failed](int edge, int i1, int i2) -> int {
const int vertex = this->triangle_midpoint_or_allocate(neighbors(edge), verts_idxs[i1], verts_idxs[i2]);
if (vertex < 0)
split_failed = true;
return vertex;
};
int ichild = 0;
switch (tr.number_of_split_sides()) {
case 1:
verts_idxs.insert(verts_idxs.begin()+2, get_alloc_vertex(next_idx_modulo(tr.special_side(), 3), 2, 1));
if (split_failed) {
tr.set_division(0, 0);
return;
}
tr.children[ichild ++] = push_triangle(verts_idxs[0], verts_idxs[1], verts_idxs[2], tr.source_triangle, old_state);
tr.children[ichild ] = push_triangle(verts_idxs[2], verts_idxs[3], verts_idxs[0], tr.source_triangle, old_state);
break;
@@ -1376,6 +1387,10 @@ void TriangleSelector::perform_split(int facet_idx, const Vec3i32 &neighbors, En
case 2:
verts_idxs.insert(verts_idxs.begin()+1, get_alloc_vertex(tr.special_side(), 1, 0));
verts_idxs.insert(verts_idxs.begin()+4, get_alloc_vertex(prev_idx_modulo(tr.special_side(), 3), 0, 3));
if (split_failed) {
tr.set_division(0, 0);
return;
}
tr.children[ichild ++] = push_triangle(verts_idxs[0], verts_idxs[1], verts_idxs[4], tr.source_triangle, old_state);
tr.children[ichild ++] = push_triangle(verts_idxs[1], verts_idxs[2], verts_idxs[4], tr.source_triangle, old_state);
tr.children[ichild ] = push_triangle(verts_idxs[2], verts_idxs[3], verts_idxs[4], tr.source_triangle, old_state);
@@ -1386,6 +1401,10 @@ void TriangleSelector::perform_split(int facet_idx, const Vec3i32 &neighbors, En
verts_idxs.insert(verts_idxs.begin()+1, get_alloc_vertex(0, 1, 0));
verts_idxs.insert(verts_idxs.begin()+3, get_alloc_vertex(1, 3, 2));
verts_idxs.insert(verts_idxs.begin()+5, get_alloc_vertex(2, 0, 4));
if (split_failed) {
tr.set_division(0, 0);
return;
}
tr.children[ichild ++] = push_triangle(verts_idxs[0], verts_idxs[1], verts_idxs[5], tr.source_triangle, old_state);
tr.children[ichild ++] = push_triangle(verts_idxs[1], verts_idxs[2], verts_idxs[3], tr.source_triangle, old_state);
tr.children[ichild ++] = push_triangle(verts_idxs[3], verts_idxs[4], verts_idxs[5], tr.source_triangle, old_state);
@@ -1448,7 +1467,12 @@ void TriangleSelector::get_facets(std::vector<indexed_triangle_set>& facets_per_
{
facets_per_type.clear();
for (int type = (int)EnforcerBlockerType::NONE; type <= (int)EnforcerBlockerType::ExtruderMax; type++) {
int max_state = int(EnforcerBlockerType::NONE);
for (const Triangle &tr : m_triangles)
if (tr.valid() && !tr.is_split())
max_state = std::max(max_state, int(tr.get_state()));
for (int type = int(EnforcerBlockerType::NONE); type <= max_state; ++type) {
facets_per_type.emplace_back();
indexed_triangle_set& its = facets_per_type.back();
std::vector<int> vertex_map(m_vertices.size(), -1);
@@ -1668,9 +1692,10 @@ void TriangleSelector::get_seed_fill_contour_recursive(const int facet_idx, cons
TriangleSelector::TriangleSplittingData TriangleSelector::serialize() const {
// Each original triangle of the mesh is assigned a number encoding its state
// or how it is split. Each triangle is encoded by 4 bits (xxyy) or 8 bits (zzzzxxyy):
// or how it is split. Each triangle is encoded by 4 bits (xxyy) or by
// 4 bits plus one or more extension nibbles:
// leaf triangle: xx = EnforcerBlockerType (Only values 0, 1, and 2. Value 3 is used as an indicator for additional 4 bits.), yy = 0
// leaf triangle: xx = 0b11, yy = 0b00, zzzz = EnforcerBlockerType (subtracted by 3)
// leaf triangle: xx = 0b11, yy = 0b00, zzzz... = EnforcerBlockerType (subtracted by 3) in base-15 chunks
// non-leaf: xx = special side, yy = number of split sides
// These are bitwise appended and formed into one 64-bit integer.
@@ -1713,14 +1738,18 @@ TriangleSelector::TriangleSplittingData TriangleSelector::serialize() const {
data.used_states[n] = true;
if (n >= 3) {
assert(n <= 16);
if (n <= 16) {
// Store "11" plus 4 bits of (n-3).
data.bitstream.insert(data.bitstream.end(), { true, true });
n -= 3;
for (size_t bit_idx = 0; bit_idx < 4; ++bit_idx)
data.bitstream.push_back(n & (uint64_t(0b0001) << bit_idx));
// Store "11" plus one or more 4-bit chunks of (n - 3), where
// 0b1111 indicates that another chunk follows.
data.bitstream.insert(data.bitstream.end(), { true, true });
n -= 3;
while (n >= 15) {
for (size_t bit_idx = 0; bit_idx < 4; ++bit_idx) {
data.bitstream.push_back(uint64_t(0b1111) & (uint64_t(0b0001) << bit_idx));
}
n -= 15;
}
for (size_t bit_idx = 0; bit_idx < 4; ++bit_idx)
data.bitstream.push_back(n & (uint64_t(0b0001) << bit_idx));
} else {
// Simple case, compatible with PrusaSlicer 2.3.1 and older for storing paint on supports and seams.
// Store 2 bits of n.
@@ -1746,17 +1775,18 @@ TriangleSelector::TriangleSplittingData TriangleSelector::serialize() const {
return out.data;
}
void TriangleSelector::deserialize(const TriangleSplittingData &data,
void TriangleSelector::deserialize(const TriangleSplittingData& data,
bool needs_reset,
EnforcerBlockerType max_ebt,
EnforcerBlockerType to_delete_filament,
EnforcerBlockerType replace_filament)
EnforcerBlockerType replace_filament,
const EnforcerBlockerStateMap* state_map)
{
if (needs_reset)
reset(); // dump any current state
for (auto [triangle_id, ibit] : data.triangles_to_split) {
if (triangle_id >= int(m_triangles.size())) {
BOOST_LOG_TRIVIAL(info) << __FUNCTION__ << "array bound:error:triangle_id >= int(m_triangles.size())";
if (triangle_id < 0 || triangle_id >= int(m_triangles.size()) || ibit < 0 || ibit >= int(data.bitstream.size())) {
BOOST_LOG_TRIVIAL(info) << __FUNCTION__ << "array bound:error:triangle_id or bitstream index is out of range";
return;
}
}
@@ -1781,7 +1811,12 @@ void TriangleSelector::deserialize(const TriangleSplittingData &data,
for (auto [triangle_id, ibit] : data.triangles_to_split) {
assert(triangle_id < int(m_triangles.size()));
assert(ibit < int(data.bitstream.size()));
auto next_nibble = [&data, &ibit = ibit]() {
bool bitstream_valid = true;
auto next_nibble = [&data, &ibit = ibit, &bitstream_valid]() {
if (ibit < 0 || ibit + 4 > int(data.bitstream.size())) {
bitstream_valid = false;
return 0;
}
int n = 0;
for (int i = 0; i < 4; ++ i)
n |= data.bitstream[ibit ++] << i;
@@ -1792,17 +1827,39 @@ void TriangleSelector::deserialize(const TriangleSplittingData &data,
while (true) {
// Read next triangle info.
int code = next_nibble();
if (!bitstream_valid)
return;
int num_of_split_sides = code & 0b11;
int num_of_children = num_of_split_sides == 0 ? 0 : num_of_split_sides + 1;
bool is_split = num_of_children != 0;
// Only valid if not is_split. Value of the second nibble was subtracted by 3, so it is added back.
auto state = is_split ? EnforcerBlockerType::NONE : EnforcerBlockerType((code & 0b1100) == 0b1100 ? next_nibble() + 3 : code >> 2);
auto state = EnforcerBlockerType::NONE;
if (!is_split) {
if ((code & 0b1100) == 0b1100) {
int next_code = next_nibble();
if (!bitstream_valid)
return;
int extension_count = 0;
while (next_code == 0b1111) {
++extension_count;
next_code = next_nibble();
if (!bitstream_valid)
return;
}
state = EnforcerBlockerType(next_code + 15 * extension_count + 3);
} else {
state = EnforcerBlockerType(code >> 2);
}
}
// BBS
if (state == to_delete_filament)
state = replace_filament;
else if (to_delete_filament != EnforcerBlockerType::NONE && state != EnforcerBlockerType::NONE) {
state = state > to_delete_filament ? EnforcerBlockerType((int)state - 1) : state;
if (state_map != nullptr && state != EnforcerBlockerType::NONE) {
const size_t state_idx = static_cast<size_t>(state);
state = state_idx < state_map->size() ? (*state_map)[state_idx] : EnforcerBlockerType::NONE;
} else {
if (state == to_delete_filament)
state = replace_filament;
else if (to_delete_filament != EnforcerBlockerType::NONE && state != EnforcerBlockerType::NONE) {
state = state > to_delete_filament ? EnforcerBlockerType((int)state - 1) : state;
}
}
if (state > max_ebt) {
@@ -1812,6 +1869,10 @@ void TriangleSelector::deserialize(const TriangleSplittingData &data,
// Only valid if is_split.
int special_side = code >> 2;
if (is_split &&
(special_side < 0 || special_side >= 3 ||
(num_of_split_sides != 1 && num_of_split_sides != 2 && special_side != 0)))
return;
// Take care of the first iteration separately, so handling of the others is simpler.
if (parents.empty()) {
@@ -1822,6 +1883,8 @@ void TriangleSelector::deserialize(const TriangleSplittingData &data,
parents.push_back({triangle_id, neighbors, 0, num_of_children});
m_triangles[triangle_id].set_division(num_of_split_sides, special_side);
perform_split(triangle_id, neighbors, EnforcerBlockerType::NONE);
if (!m_triangles[triangle_id].is_split())
return;
continue;
} else {
// root is not split. just set the state and that's it.
@@ -1840,13 +1903,20 @@ void TriangleSelector::deserialize(const TriangleSplittingData &data,
int child_idx = last.total_children - last.processed_children - 1;
Vec3i32 neighbors = this->child_neighbors(tr, last.neighbors, child_idx);
int this_idx = tr.children[child_idx];
if (this_idx < 0 || this_idx >= int(m_triangles.size()))
return;
m_triangles[this_idx].set_division(num_of_split_sides, special_side);
perform_split(this_idx, neighbors, EnforcerBlockerType::NONE);
if (!m_triangles[this_idx].is_split())
return;
parents.push_back({this_idx, neighbors, 0, num_of_children});
} else {
// this triangle belongs to last split one
int child_idx = last.total_children - last.processed_children - 1;
m_triangles[m_triangles[last.facet_id].children[child_idx]].set_state(state);
const int this_idx = m_triangles[last.facet_id].children[child_idx];
if (this_idx < 0 || this_idx >= int(m_triangles.size()))
return;
m_triangles[this_idx].set_state(state);
++last.processed_children;
}
@@ -1893,7 +1963,16 @@ void TriangleSelector::TriangleSplittingData::update_used_states(const size_t bi
if (const bool is_split = (code & 0b11) != 0; is_split)
continue;
const uint8_t facet_state = (code & 0b1100) == 0b1100 ? read_next_nibble() + 3 : code >> 2;
size_t facet_state = code >> 2;
if ((code & 0b1100) == 0b1100) {
size_t extension_count = 0;
size_t next_code = read_next_nibble();
while (next_code == 0b1111) {
++extension_count;
next_code = read_next_nibble();
}
facet_state = next_code + 15 * extension_count + 3;
}
assert(facet_state < this->used_states.size());
if (facet_state >= this->used_states.size())
continue;
@@ -1923,9 +2002,19 @@ bool TriangleSelector::has_facets(const TriangleSplittingData &data, const Enfor
auto num_children_or_state = [&next_nibble]() -> int {
int code = next_nibble();
int num_of_split_sides = code & 0b11;
return num_of_split_sides == 0 ?
((code & 0b1100) == 0b1100 ? next_nibble() + 3 : code >> 2) :
- num_of_split_sides - 1;
if (num_of_split_sides != 0)
return - num_of_split_sides - 1;
if ((code & 0b1100) != 0b1100)
return code >> 2;
int extension_count = 0;
int next_code = next_nibble();
while (next_code == 0b1111) {
++extension_count;
next_code = next_nibble();
}
return next_code + 15 * extension_count + 3;
};
int state = num_children_or_state();

16
src/libslic3r/TriangleSelector.hpp
View File

@@ -10,14 +10,15 @@
namespace Slic3r {
enum class EnforcerBlockerType : int8_t {
enum class EnforcerBlockerType : int16_t {
// Maximum is 3. The value is serialized in TriangleSelector into 2 bits.
NONE = 0,
ENFORCER = 1,
BLOCKER = 2,
// For the fuzzy skin, we use just two values (NONE and FUZZY_SKIN).
FUZZY_SKIN = ENFORCER,
// Maximum is 15. The value is serialized in TriangleSelector into 6 bits using a 2 bit prefix code.
// Extruder states are serialized using a 2-bit prefix plus one or more 4-bit nibbles.
// This allows more than 16 painted states while keeping backward compatibility.
Extruder1 = ENFORCER,
Extruder2 = BLOCKER,
Extruder3,
@@ -34,7 +35,7 @@ enum class EnforcerBlockerType : int8_t {
Extruder14,
Extruder15,
Extruder16,
ExtruderMax = Extruder16
ExtruderMax = 255
};
// Type alias for the state mapping array to improve code readability
@@ -363,10 +364,11 @@ public:
// Load serialized data. Assumes that correct mesh is loaded.
void deserialize(const TriangleSplittingData& data,
bool needs_reset = true,
EnforcerBlockerType max_ebt = EnforcerBlockerType::ExtruderMax,
EnforcerBlockerType to_delete_filament = EnforcerBlockerType::NONE,
EnforcerBlockerType replace_filament = EnforcerBlockerType::NONE);
bool needs_reset = true,
EnforcerBlockerType max_ebt = EnforcerBlockerType::ExtruderMax,
EnforcerBlockerType to_delete_filament = EnforcerBlockerType::NONE,
EnforcerBlockerType replace_filament = EnforcerBlockerType::NONE,
const EnforcerBlockerStateMap *state_map = nullptr);
// Extract all used facet states from the given TriangleSplittingData.
static std::vector<EnforcerBlockerType> extract_used_facet_states(const TriangleSplittingData &data);