A few methods differed by the number or type of arguments in derived
classes. They were changed to match the base class method signature and
marked as 'override'.
Module placer did not mark the newly placed modules as selected,
but all edit functions (rotate/place/etc.) rely on getting a selected
item. When a rotation/flip command was issued, the newly placed module
was dropped and the one underneath the cursor has been rotated.
To fix this the newly placed modules are marked as selected.
It also simplifies the placer code a bit.
Fixes: lp:1738148
* https://bugs.launchpad.net/kicad/+bug/1738148
Revert() method was still not aware of CHT_DONE flag, therefore
the switch statement was always invalid. To fix it, the commit
type is split to type and flags as is done in Push() method.
Zones: it turned out that for every zone point, the set of hatch lines
has been drawn, resulting in numerous redundant lines. Now the set is
drawn only once per zone.
Custom shape pads: as GAL supports SHAPE_POLY_SETs, there is no need to
create a temporary vector to copy there a list of points to be drawn.
Changed the drawing calls to use SHAPE_POLY_SETs directly.
Thanks to Andreas Buhr patch, code for drawing filled polygons and their
outlines has been unified. Previously it was done by changing the
drawing method (GAL::DrawPolygon() vs GAL::DrawPolyline()), now it is
done by modifying attributes (stroke/fill).
The polygon drawing routing in the OpenGL-GAL did
not plot polygons which are "closed", i.e. have a segment
from their last point to their first point, correctly.
That segment was omitted.
This commit fixes this bug.
If the object changes between a TA_CONTEXT_MENU_UPDATE
event and a TA_CONTEXT_MENU_CHOICE event (usually as
a result of a cancel), the old code would leave the
last item hidden.
Fixes: lp:1737562
* https://bugs.launchpad.net/kicad/+bug/1737562
The OSX implementation needs to be poked to update the
brightening / unbrightening during running of the
disambiguation menu. (FWIW, now that I've finally seen
it in action, it's a REALLY cool feature.)
Fixes: lp:1715440
* https://bugs.launchpad.net/kicad/+bug/1715440
The reason the component tree widget was cleared on Windows is the fact
m_adapter->UpdateSearchString( m_query_ctrl->GetLineText( 0 ) ) is
called on any change of the symbol. But if m_query_ctrl is empty,
m_query_ctrl->GetLineText( 0 ) returns the "hint" (currently the word
"Search") not the actual value, so there is no "candidate" symbol in
list. Using m_query_ctrl->GetValue() returns the actual value and fixes
this issue.
Fixes: lp:1737132
* https://bugs.launchpad.net/kicad/+bug/1737132
Pins are weak_ptrs to the library, so they require a lock() before
accessing. This imposes a performance hit on fast loops that access
the pin list repeatedly. This patch caches the pin position locally
for each component, updating only when needed.
Fixes: lp:1737363
* https://bugs.launchpad.net/kicad/+bug/1737363
COMPONENT_TREE::STATE used wxDataViewItem to store the selection, but it
will point to a deleted object if the selection was removed. Switched to
storing the selection using LIB_ID as it is much safer to use.
The OPENGL_GAL::drawPolygon function and all functions using it
did not respect the isFillEnabled member set by
GAL::SetIsFill. This is fixed by this patch.
The "operator BOX2I() const" of EDA_RECT produced
wrong boxes, usually much to big. It passed its
end-position as second argument to the constructor
of BOX2I. However, BOX2I expects its size as second
argument.
This caused a dramatic performance penalty in some
cases.
Because of how wxMenuItem decodes the label text, the case of
alphabetic hotkeys is not preserved when displaying the menu. So,
alphabetic hotkeys that use the Shift key must be defined this way.
Fixes: lp:1737303
* https://bugs.launchpad.net/kicad/+bug/1737303
Show the NC box only on pins connected via the same component,
indicating a stacked pin group. Do not show NC box on pins for
different components with the same symbol.
There were two problems in the triangulation caching
of SHAPE_POLY_SET:
First there was a double free:
While SHAPE_POLY_SET implements the copy constructor,
it did not implement the operator=, which resulted
in the default operator= being generated by the
compiler. The default operator= copied the member
m_triangulatedPolys, which is a std::vector of pointers.
So after operator= execution, there are two SHAPE_POLY_SET
having pointers to the same TRIANGULATED_POLYGONs, each
of them deleting them in their destructors. This led
to segfaults, because calling
TransformShapeWithClearanceToPolygon on a Zone
uses operator= to copy the contained SHAPE_POLY_SET.
The new SHAPE_POLY_SET then went out of scope and
deleted the cached triangulation within the Zone.
This first problem is fixed by implementing operator=
for SHAPE_POLY_SET.
Second, there was a memory leak: Calling
"CacheTriangulation" on a SHAPE_POLY_SET,
then changing the polygon and then calling
"CacheTriangulation" again led to
leaking the
triangulations generated in the first call.
This second problem is fixed by holding
the cached triangulations in a unique_ptr.
If a schematic contains two symbols with the same name from different
libraries, the cache will contain the last symbol saved with that name.
Prepend the library nickname with the original schematic symbol library
nickname when saving the cache library to prevent name collisions.
Moved a DRC function (BOARD::Test_Drc_Areas_Outlines_To_Areas_Outlines)
from BOARD class to DRC (DRC::TestZoneToZoneOutline). Added BOARD_COMMITs
to handle marker visibility updates.
Fixes: lp:1619124
* https://bugs.launchpad.net/kicad/+bug/1619124
Moved the code handling text overbars to a single function
(ProcessOverbars()) to replace similar pieces of code trying to do the
same thing.
Fixes: lp:1735906
* https://bugs.launchpad.net/kicad/+bug/1735906
jean-pierre charras
Tomasz Włostowski
Seth Hillbrand
Eldar Khayrullin
Maciej Suminski
Andreas Buhr
Jeff Young
Oliver
Jon Evans
Wayne Stambaugh