The new line collision search uses BBox() to check for colliding
objects. BBox in the SHAPE_LINE_CHAIN did not include width as the
chains were assumed to be zero-width. This is not the case for
PNS::LINE elements.
We mostly don't notice this because DRC checks for SEGMENT collisions
but it becomes obvious/annoying when we cannot place a track for unknown
reasons and the snap-back doesn't take line width into account.
Fixes#3776 | https://gitlab.com/kicad/code/kicad/issues/3776
(cherry picked from commit fe15511d38)
1) Intermediate states might be self-intersecting, and we shouldn't
be policing our users on what order to do things in
2) The polygon might already be self-intersecting, at which point we're
preventing the user from fixing it.
Also includes better const management for SHAPE_POLY_SET API.
Fixes: lp:1833831
* https://bugs.launchpad.net/kicad/+bug/1833831
(cherry picked from commit a3c74051c2)
Sometimes we want to inflate a polygon without adding rounded edges.
This add the option using the jtMiter setting.
This is used in the Eagle parser to expand the Eagle zones for KiCad.
Eagle Zones are drawn on the polygon edge, so they extend out from the
outline. KiCad zones are drawn inside the polygon. We need to both
increase the zone size and decrease the minimum pen width to account for
this.
Fixes: lp:1817312
* https://bugs.launchpad.net/kicad/+bug/1817312
(cherry picked from commit 490c805319)
Libpolygon can provide its own includes via target_include_dirs PUBLIC.
This means any linking targets do not need to specifiy them manually.
As common requires polygon, the polygon dep is also now no longer
required downstream of libcommon, as it's transisitvely implied
by libcommon's target_link_libraries.
This resolves a circular dependency previously detected and also
simplifies CMakeLists.
ClipperLib::Path is std::vector<IntPoint>. Iterating this with
"for( auto point : path)" could result in 'n' IntPoint copy-constructions.
It does seem GCC 8, at least, manages to optimise these constructions
out.
Replace with the "standard" for( const auto& point : path) idiom.
When removing steiner points, there is the possibility that the test
point is also removed. In this case, it is no longer a member of the
linked list and will break the output.
The test for re-fracturing a broken polygon can also result in multiple
polygons, rather than only 0 or 1. Skipping the extra polygons will
result in a limited tesselation.
Fixes: lp:1812393
* https://bugs.launchpad.net/kicad/+bug/1812393
The fracture() call may result in zero polygons remaining, which will
cause failure in our tesselation routine, so we need to check whether
this is a valid POLYGON before re-tesselating.
Tesselation can fail for a number of reasons. When this happens, we set
the triangulationValid flag to false to prevent using the broken
triangulation. This will fall back to the slow OpenGL triangulation
when DrawPolygon is called.
Use TesselatePolygon() to draw polygons in Gerbview instead of GLU tesselation, much slower.
Add helper methods in GAL to know if the current GAL engine is Cairo, OpenGL or something else,
useful to optimize drawing code.
This adds a check for contiguous board outlines to the DRC. It also
uses the calculated outline to ensure that traces are not crossing the
outlines.
Fixes: lp:1648055
* https://bugs.launchpad.net/kicad/+bug/1648055
This commit finishes the removal of OpenMP from the KiCad codebase.
Removed in this commit are the OpenMP calls in 3d-viewer and
qa/polygon_triangulation as well as all references in CMakeLists.txt
std::thread is used instead for multithreaded computation
Commit 73c229714 was a bit of a sledgehammer for the associated problem
of degenerate points. This commit replaces that one by only performing
additional simplification of the zone polygons on those polygons that
fail our initial triangulation attempt.
Replaces Poly2Tri with updated code to process polygons faster and more
robustly. Notably, we can now handle overlapping holes in the polygons,
allowing us to cache the triangulation of complex boards
This reverts the previous work-around 7d62f14dd for the RTree splitting
degeneracy that was placed prior to v5. It appears not to have worked
for all systems. In its place, we use doubles instead of floats to
calculate the bounding box when filling the RTree. This keeps maximum
volume items from overlapping with the system boundary in test cases.
This introduces a new util function in geometry_utils which snaps a vector
to axes or 45 degree lines. This can be used whenever you want to
snap a vector to these angles, but still want it to stay on a grid.
This snapping is used for the dimension tool and the ruler tool.
This is substantially simpler for two-point tools that the method
used by the line tool, which uses DIRECTION_45.
Fixes: lp:1780826
https://bugs.launchpad.net/kicad/+bug/1780826
When removing items, we perform a two-pass removal, checking first
for the cached rectangle collision. Then, if we do not find the item
being removed, we perform a second, more expensive pass over the full
tree. This second pass is required as we cannot be certain that an
item's bbox has not been modified between insertion and deletion. In
which case, keeping stale pointers in the tree will lead to segfaults.
Fixes: lp:1777246
* https://bugs.launchpad.net/kicad/+bug/1777246
Changes SHAPE_CONVEX to SHAPE_SIMPLE to better reflect the limitations.
Changes SHAPE_LINE_CHAIN::PointInside calculation to allow points
strictly inside a line chain
Fixes: lp:1768862
* https://bugs.launchpad.net/kicad/+bug/1768862
jean-pierre charras
Seth Hillbrand
Jeff Young
Seth Hillbrand
John Beard
Tomasz Włostowski
Maciej Suminski
Jon Evans