Don't merge classified pieces of layerm->slices, but use the non-split
islands of a layer. For a single region print, these shall be equal.
Added Slic3r-console.ico windows icon for the console Slic3r.
This is a copy of the big icon, with the big images stripped off.
These could be calculated from the fill areas if needed.
On the other side, the non-classified (non-split) fill areas are stored
now for use in the "ensure vertical wall thickness" feature,
also the non-split fill areas are re-used when recalculating the infills.
This is safer than trying to stitch the fill region together from the
classified fragments.
Modified the "ensure vertical wall thickness" feature to use the non-split
fill areas instead of perimeter areas for the calculation
of non-supported regions. This is cheaper as the fill areas contain
roughly half the edges.
Visual Studio 2013. Added multiple compile time checks for data
sizes and alignment. The library STL import is not big endian safe, so
added a test for endianity, modified STL export to a faster little endian only.
slightly by inflating the projected top/bottom/bottom bridge surfaces
before they are added into a surface. This ensures, that the possible
projected infill areas merge with the perimeter supporting areas,
but the perimeter supporting areas will not be inflated on their own,
if there is no touching projection of a top/bottom/bottom bridge
surface.
Fixed some cracks in the fill surfaces created by rounding all surfaces inside detect_surface_type().
Fixed https://github.com/prusa3d/Slic3r/issues/12
Bridging-Angle not optimal
Extended the "Ensure veritcal wall thickness" mode (merged with the original discover_horizontal_shells function), but this a work in progress. Already Slic3r with "ensure vertical wall thickness" produces less spurious infills inside solids.
will force the perl confess / croak called from a C++ code to
print an error message and exit. There is a bug in some Strawberry
Perl instalations leading to an interpreter hang up and an exit
with a reasonable message is better than that.
direction over multiple regions. This allows a single bridge to be
drawn over holes, which are too close to each other to allow
for separate bridges.
Fixes Bridging-Angle not optimal
https://github.com/prusa3d/Slic3r/issues/12
Re-allowed adaptive infill line width for solid infills. The adaptive
infill line width works in some circumstances, see Issue #15,
but the original implementation often changed the line width too
aggressively. The current implementation limits the line width change
to 20%.
Fixes Gaps between infill and perimeter leads to errors in laydown on following layer
https://github.com/prusa3d/Slic3r/issues/15
to return a string with an error message instead. This was necessary
to avoid a hang-up on some Strawberry Perl distributions, when
a perl "croak" function is called after a C++ exception is caught.
A to_polygons(SurfacePtrs &&) method does not make sense as
the ownership of the Surfaces stored in the pointer array is not known.
Thanks to @flannelhead for precisely pinpointing this issue.
Replaced eval { die } construct with a bool return value indicating
success or failure of an automatic arrangement of parts on the print bed.
Don't know exactly what is happening here, but throwing a "die" inside
a XS function and then catching it inside an eval {} block is suspcious.
into polygons_covered_by_width() and polygons_covered_by_spacing().
Bugfix of ExtrusionLoop::split_at(const Point &point),
where the split ExtrusionPaths were not initialised correctly.
Removed dependency on Perl Math::PlanePath module.
Fixed compilation with Visual Studio and SLIC3R_DEBUG: Visual Studio older than 2015 does not support the prinf type specifier %zu. Use %Iu instead.
C++11 move semantics enabled.
Added "filament_max_volumetric_speed", a cap on the maximum volumetric
extrusion role, filament specific. This is very useful when mixing
rigid filament with a soft filament.
Extended the import / export of multi-string values into configuration
values, including the test cases. Multi-line strings will be enclosed
into quotes, quotes escaped using a C-style escape sequences. Single
word strings could still be stored without quotes.
Initial implementation of the "infill link maximum distance" feature.
Parts of the perimeter connecting two infill lines will be dropped,
if longer than a given threshold.
are thick enough to be actually filled by the rectilinear fill.
Otherwise gaps would be left in the sparse infill, causing missing
bond to the perimeters.
Removed the deprecated VibrationLimit feature.
Added triangle infill.
The Prusa3D fork of Slic3r has been marked as "Slic3r Prusa Edition"
with menus pointing to the prusa3d/slic3r github release page
and Prusa3D drivers downloads page.
Made sure the C++ fillers are instantiated at the worker threads,
where there are being released.
Extended the FillRectilinear2 to calculate the contour / line intersection
with exact arithmetics, improved robustness and added error handling
and error reporting, if the contours to be filled are not correct.
Added a safe variant of offset(const Slic3r::ExPolygon...), which offsets each loop separately.
New functions "remove_sticks" to remove zero area parts of polygons.
New functions "remove_small" and "remove_degenerate" for polygon clean up.
Extended the C++ supports, those are not finalized yet though.
Also a bug has been fixed for zero interface layers. Before
slic3r would put infinite number of interface layers over top surfaces,
if the number of interface layers was set to zero.
these regions are grown to anchor the bridge lines to the bottom surface.
The grown regions may overlap. In that case the regions are now merged
before the bridging direction is calculated for the merged region.
Updated xs/Build.PL to understand BOOST_LIBRARY_PATH and
BOOST_INCLUDE_PATH environment variables. This way one may easily
switch between various boost builds.
Some minor tweeks were done to make Slic3r compile with
Visual Studio 2013.
of virtual inheritance. Note that an invocation of ConfigBase::optptr()
is routed to FullPrintConfig::optptr() for all classes of the FullPrintConfig
hierarchy. FullPrintConfig::optptr() in turn invokes optptr()
of PrintObjectConfig, PrintRegionConfig, PrintConfig and HostConfig.
Due to the use of virtual inheritance, this all happens, when
PrintObjectConfig gets constructed as part of FullPrintConfig, but
at that time PrintRegionConfig, PrintConfig and HostConfig are not
constructed yet. Accessing them at that time leads to crashes,
when compiled with Visual Studio 2013 compiler. For some reason
the code generated by gcc does not crash, but I believe the behavior
is undefined and it is better to be fixed anyway.
The patch solves the problem by calling set_defaults() by the topmost
object, which not only fixes the crashes, but also avoids repeated
initialization.
The recent GCode writer changes which put the speed changes on a line of
their own have caused the layer time slowdown to be ignored by the regex
in CoolingBuffer.pm.
Fixes: #3134
