with their original names. Using weird shorthands makes the code
unreadable for anyone but the original author.
template<class T> using shptr = std::shared_ptr<T>;
template<class T> using uqptr = std::unique_ptr<T>;
template<class T> using wkptr = std::weak_ptr<T>;
There was a bug for more than 4 years, which likely triggers now more
often with the introduction of raft_contact_distance parameter,
which is usually significantly smaller than support_material_contact_distance.
There were no support towers built at the raft for contact layers
below the print_z of the first object layer.
1) Added parsing of 3MF PrusaSlicer generator semantic version
2) For 3MFs generated by >= "2.4.0-alpha1" and < "2.4.0-alpha3",
remove unreferenced vertices on loading. This should shrink the incorrectly
generated 3MFs back.
3) Added check for an empty mesh, which could have crashed PrusaSlicer on 3MF parsing.
This is a regression wrt. PrusaSlicer 2.4.0-alpha1 due to the G-code
processing optimization and parallelization.
Related to GCode Viewer changes files modified date in windows #5079
Zero support base spacing is newly allowed, switching from
the "support base" infill pattern to rectilinar infill pattern.
Why someone would want to use a solid infill for support base eludes me,
but it is simpler to support it instead of working out some rules on
minimum support base density.
Fixes Support patern spacing set to zero causing crash #6989
so that a mesh with incorrectly oriented faces will not be considered
a manifold.
Added assert for 3D convex hulls constructed by QHull. They shall be
manifold (however sometimes they are not).
TriangleMesh newly only holds indexed_triangle_set and
TriangleMeshStats. TriangleMeshStats contains an excerpt of stl_stats.
TriangleMeshStats are updated when initializing with indexed_triangle_set.
Admesh triangle mesh fixing is newly only used when loading an STL.
AMF / 3MF / OBJ file formats are already indexed triangle sets, thus
they are no more converted to admesh stl_file format, nor fixed
through admesh repair machinery. When importing AMF / 3MF / OBJ files,
volume is calculated and if negative, all faces are flipped. Also
a bounding box and number of open edges is calculated.
Implemented its_number_of_patches(), its_num_open_edges()
Optimized its_split(), its_is_splittable() using a visitor pattern.
Reworked QHull integration into TriangleMesh:
1) Face normals were not right.
2) Indexed triangle set is newly emitted instead of duplicating
vertices for each face.
Fixed cut_mesh(): Orient the triangulated faces correctly.
stl_stats are newly only accessed by TriangleMesh::stats(),
most of the direct access to TriangleMesh::stl is gone with the exception
of parsing input files (3MF, AMF, obj).
Sometimes Clipper produces a polyline with more than 2 points when
clipping a line with a polygon or a set of polygons. We hope the intermediate
points are collinear with the line, so we may just ignore them.
Exporting G-code on a worker thread did not work correctly as the worker
threads were using user's locale, not "C" locale.
The "C" locale is newly enforced to TBB worker threads by
name_tbb_thread_pool_threads_set_locale()
WIP to G-code export parallelization through pipelining:
Decoupled CoolingBuffer from GCode / GCodeWriter, ready to be
pipelined on a different thread.
WIP to G-code export parallelization through pipelining:
GCodeProcessor is called during the G-code export,
the G-code is no more reopened and re-read, but it is pipelined
from the G-code generator.
GCodeViewer no more parses G-code just to extract line end positions.
Removed start_mapping_gcode_window(), void stop_mapping_gcode_window(),
they are no more needed.
This is a regression to a late PrusaSlicer 2.4.0-alpha0 change
8dfc0422a8
Faster and hopefully more reliable projection of paint-on support
blockers and enforcers on a sliced mesh.
Previous d89f01c717 did not fix it.
This is a regression to a late PrusaSlicer 2.4.0-alpha0 change
8dfc0422a8
Faster and hopefully more reliable projection of paint-on support
blockers and enforcers on a sliced mesh.
1) Activate installed filament or SLA material profile after update_compatible(),
so that the compatiblity and visibility flags of presets are updated.
2) Only activate the first newly installed filament / SLA material profile
if the active printer did not change. This also means that if no filament
profile was active before Wizard was open or it became incompatible with
the newly installed Printer profile, the default filament profile assigned
to the activated Printer is activated preferably, which may or may not
be one of the newly installed filament profiles.
object layer over raft interface:
"first_layer_speed_over_raft", "first_layer_acceleration_over_raft".
Fixes I have a question about the speed of the first layer after the raft. #6623
Fixes Layer After Raft Is Not Considered First Layer! #6166
1) Changed the name of the variable "brim_offset" to "brim_separation"
for clarity.
2) Added legacy conversion after loading an old 3MF that does not define
then new "brim_separation" variable: The "brim_separation" is being
filled in with the "elefant_foot_compensation" value to produce
equal brim separation to the old PrusaSlicer that saved that 3MF file.
The triangle-ray intersection function used a hard coded epsilon,
which did not work for triangle meshes, that were either too small
or too large. Newly the epsilon may be provided to the AABBTreeIndirect
search functions externally and IndexedMesh calculates a suitable
epsilon on demand from an average triangle mesh edge length.
6b03b66167
Renamed the new "printhost_ignore_check" option to "printhost_ssl_ignore_revoke"
Improved the Physical Printers dialog in regard to the new option checkbox
(added tooltip to the checkbox, moved it to the end of options).
Disabled the host_xxx options at the command line interface, they no
more work after these options were separated to Physical Printers profiles.
Little refactoring of Http.cpp/hpp, OctoPrint.cpp/hpp
Private local variables prefixed with m_, some renaming for clarity.
list of modified options.
Optimization of DynamicConfig::equals(), ::diff(), ::equal()
to iterate over the two compared std::map trees instead of
first generating a list of keys and then searching for each key
in the respective map.
Optimization of PresetCollection::current_is_dirty() and ::saved_is_dirty()
to call DynamicConfig::equals() instead of ::diff().
of a printer from MMU to non-MMU (when number of extruders changes).
Also slightly optimized in case a MMU-painted object is sliced in
single extruder mode.
In case the configuration snapshot cannot be taken before installing
configuration updates (because the current configuration state
is invalid), ask user whether to continue or abort.
Vojtech Bubnik