In commit 7cb754dd a call to wxMilliSleep(50) was removed (because it created
lag in editing) but this removal created a 100% CPU core usage.
Using a much small sleep time (1ms) fixes these issues.
Fixes https://gitlab.com/kicad/code/kicad/-/issues/17979
Defer telling the statement about the connection until the
try block, because otherwise it can connect immediately and
potentially throw an exception if the connection fails.
We need the object in m_locale for a bit more things during program cleanup,
especially on macOS
The destructor of PGM_BASE is sufficient allowance of lifetime.
Fixes https://gitlab.com/kicad/code/kicad/-/issues/17369
(cherry picked from commit 9ffa5db667)
Co-authored-by: Marek Roszko <mark.roszko@gmail.com>
* Do not handle tab key event when committing property changes so the
property grid tab navigation works correctly.
* Do not call commit property changes on tab key when no changes are
pending.
* Skip event handling in the base object value change and changing event
handlers in case they do not get overloaded in derived objects.
* Do not force focus to canvas on property change so arrow and tab key
grid navigation work properly. Forcing the user to click the property
grid control to edit a single property doesn't make sense.
* Pass properties panel show event up the event stack.
* Do not set focus to canvas in tool dispatcher if it already has the
focus.
Fixes https://gitlab.com/kicad/code/kicad/-/issues/16957
manully cherry picked from commit 8a1347d2c8
The arraystring reference is lost when we don't keep the wxFileName
variable in scope.
This also adds removing invalid paths and checking for correct version
before sorting
Fixes https://gitlab.com/kicad/code/kicad/-/issues/17749
(cherry picked from commit d2701323a8)
Refactors `SHAPE_POLY_SET::fractureSingle()` to be more efficient, while
not changing the actual algorithm:
* increase cache locality by using contiguous arrays instead of what was
effectively a linked list
* reduce latency and jitter by replacing per-edge allocator calls with
ahead-of-time std::vector reserves
* increase cache efficiency by making the vertex struct smaller
* replace O(n^2) leftmost edge search with O(n log n) std::sort
* sort the polygons instead of the edges
* cut iteration count in half in the remaining O(polygons * edges) part
(cherry picked from commit e98c9f283f)
Logging is useful when we find an area that cannot be triangulated.
This will be used to generated test cases.
Skipping minor untesselated areas means that the polygon will still be
considered fully tesselated (and not sent back again and again) even if
the tesselation misses an area less than the configured limit.
Currently, this is 31^2nm.
(cherry picked from commit 7e7fec69f6)
This was a crude hack that appends the library URN ordinal to the name of
the footprint so that it can be correctly looked up by the "element" node.
The Eagle XML DTD does not make it clear how the URN "ASSET_ID" and
"VERSION" are used to look up the appropriate "ASSET_TYPE" so this is a
best guess and seems to work correctly.
The inferred edge clearance dialog had to be disabled when importing third
party boards because on GTK (and possibly other platforms) the dialog would
completely hang KiCad preventing the imported board and possible schematic
from being saved.
Fixes https://gitlab.com/kicad/code/kicad/-/issues/12897
(cherry picked from commit 34ec57958d)
Don't apply pruning to libraries that don't have
any children, since the point of the pruning is to
hide libraries that don't directly match a search
term if they don't have any children that match.
Fix searching for full LIB_IDs that got broken by
the implementation of "AND"
Fix searching for library names alone
Fixes https://gitlab.com/kicad/code/kicad/-/issues/17205
(cherry picked from commit 0c2a084a47)