SPE-1072
Working but flipped normals with the interior.
Testing on treefrog passed
Oversampling for hollowed mesh should not be less than 3x
Flip back normals after simplify and remove redundant test code.
* Add Linux openvdb integration
* Add Mac openvdb integration and enable in ALL
* Create openvdb sandbox to test integration.
* Additional fixes in the patches
* Remove slabasebed sandbox as it has no relevance now
* Provide FindOpenVDB module and fix build issues
* Remove unused libnest2d files. Make it use the global build script targets.
* Modify FindTBB to address multi-config builds and take care of __TBB_NO_IMPLICIT_LINKAGE
* Move FindNLopt to project common cmake module dir
* Rename libnest.hpp to nester.hpp and libnest.h to libnest.hpp
* Clean up common test suite build scripts
producing better shortest path estimate than the "closest next neighbor"
heuristics. The new greedy algorithm utilizes KD tree for closest
end point search, and builds a graph to detect loops.
PerimeterGenerator newly uses the optimized TSP algorithm.
ExtrusionEntity has been refactored / simplified.
All requested config values are written into SL1 ini file inside the zip
* TIme.hpp and Time.cpp is now part of libslic3r instead of libslic3r_gui
* Updated time manipulation function: separate timestamp_local_str and timestamp_utc_str
* timestamp_utc_str is used in header_slic3r_generated(). Gcode now contains UTC timestamps
Source files are checked using a small utility in src/build-utils
This is done to prevent bugs in build and localization caused
by weird non-UTF-8 encodings interpreted by MSVC in terms of local codepages
rather than UTF-8.
A static symbol Slic3r::SEMVER is introduced, which holds
the running slicer's Semver object.
This is mainly done to make testing updater behaviour
_much_ easier. Additionaly to cleanup some questionable code
(Semver was being parsed multiple times / in multiple places
in the frontend.)
- wipe tower respects max volumetric flow, slowing down with PVA etc is not hardcoded anymore
- wipe tower is now allowed for multiple-extruder printers. single extruder stuff is not used in that case (ramming, cooling, etc.)
- start/end filament and toolchange custom gcodes are now inserted differently than before - see 41164a9
- some refactoring, e.g. the abstract WipeTower class was eradicated
Integration of the "cereal" serialization library.
Serialization / deserialization of the DynamicConfig / DynamicPrintConfig.
DynamicPrintConfig serializes ordinal identifiers instead
of the option key strings to conserve space.
Fixing dep build script on Windows and removing some warnings.
Use bundled igl by default.
Not building with the dependency scripts if not explicitly stated. This way, it will stay in
Fix the libigl patch to include C source files in header only mode.
Cleanup, fix build on windows and add test for rotcalipers.
Try to fix compilation on windows
With updates from libnest2d
Another build fix.
Clean up and add comments.
adding rotcalipers test and some cleanup
Trying to fix on OSX
Fix rotcalipers array indexing
Get rid of boost convex hull.
Adding helper function 'remove_collinear_points'
Importing new libnest2d upgrades.
Disable using __int128 in NFP on OSX
https://cgit.freedesktop.org/mesa/glu/
The library is stable (it should be, since it is in use since 1994),
but it is not thread safe: Its callbacks do not receive any pointer
to a context.
Removed libslic3r from the default include paths for all modules but
libslic3r. Now headers from libslic3r need to be included with an
explicit path (libslic3r/libslic3r.h etc)
Split the localization macros into I18N.{cpp,h}
