wxWidgets' UTF8 mode wasn't thread-safe, so we had to put mutex's on
a bunch of string operations, which meant that we couldn't return
references to strings. Now that we no longer use UTF8 mode anywhere
(and the mutex's were removed long ago), using references is a big
performance win in some places.
Between the time the user loads the footprint preview in cvpcb or Add
Footprint in pcbnew, we don't need to verify cache integrity to display
the footprint. We use the cache copy when displaying until we close the
selection window and re-open. This limits our need to iterate over the
full directory on every footprint display.
Alternate solutions for v7 should include looking into libfswatch for
change detection
Timestamps were checked using a different routine than they were
generated with, leading to the cache always being out of date.
This uses the same routine for both. While it iterates over the
directory twice during the initial load, this will never be dominant
Fixes https://gitlab.com/kicad/code/kicad/issues/7711
The move tool can stack on others, so when we re-enter the previous tool
this caused another immediate action. We flag re-entry in the tool
stack to check for this and avoid unexpected tool starts
When a user clicks on the "Add Component", "Add Power" or "Add Footprint"
icon, they want to add the item. Requiring an extra click on the screen
before selecting the component/footprint is not needed.
We auto-size them in the code and doing both produces nothing but a
mess.
Also improves the auto-sizing for the layers drop-down (particularly
needed for the new, somewhat longer, layer names.
Fixes https://gitlab.com/kicad/code/kicad/issues/7671
Also remove an implicit zone merge that was happening when the user
edited the zone parameters. This appears to have been a feature but it
happened without a) telling the user and b) having the option to prevent
it.
Fixes https://gitlab.com/kicad/code/kicad/issues/7661
We cannot do forward/reverse annotation in Single Top mode because we
depend on shared memory access. This disables the menu options in this
case and replaces the toolbar update pcb icon with the import netlist,
which is more appropriate for single top mode
Fixes https://gitlab.com/kicad/code/kicad/issues/7637
- Rename member variable names to use the 'm_' prefix
- Rename a few object types that clashed with KiCad object names, such
as BOARD and SCHEMATIC, to avoid the use of '::' to access the KiCad
objects.
- Remove some unused code.
Add a reset button that will only show up if the user
edits the JSON configuration for grids. This is a temporary
measure until we have a real grid editor GUI in V7.
Fixes https://gitlab.com/kicad/code/kicad/-/issues/7595
Each symbol unit in eeschema has a unique identifier. But we don't have
a unique identifier for the entire symbol. So changing which symbol
instance was unit A (our default base for matching), changed the UUID
that we were using to match the footprints.
This commit adds all UUIDs to the netlist, allowing us to match symbol
to footprint without worrying about which unit is referenced.
This still does not handle changing different units on different sheets.
Fixes https://gitlab.com/kicad/code/kicad/issues/7604
First issue: the C++ action plugins list should be cleared so that any
plugins that fail to reload do not hang around on the toolbar.
This is fixed on the C++ side.
Second issue: inner dependencies of plugins that are modules were not
reloaded before, which broke reloading for many plugin examples out there.
This is fixed on the Python side by keeping track of dependencies and
clearing them out before reloading a module (unfortunately there doesn't
seem to be a way to do this in the standard library)
Fixes https://gitlab.com/kicad/code/kicad/-/issues/7245
This change is mainly due to the fact Modal dialogs do not work when our
current tool is the tool used to create new item (text, zone, graphic).
If when this tool is active, one try to edit the properties of the new item
being created (or being placed) the modal dialog does not capture the event tool
and hangs (cannot be dismissed)
Our quasi modal dialogs have their own event loop and work.
Fixes#7585https://gitlab.com/kicad/code/kicad/issues/7585
CHANGED: If the user made a selection and then tried dragging it, the
drag would only succeed if the selection contained at least one
footprint. Selections with only tracks and vias would instead make
KiCad drag a new selection box.
This MR allows routing of selections of a single track or via, while moving
selections containing more than one item, whether or not they also contain
a footprint.
Fixes https://gitlab.com/kicad/code/kicad/issues/7539
- Make DIALOG_SHIM::SetPosition working)
- rename PCB_BASE_FRAME::SelectLayer() to PCB_BASE_FRAME::SelectOneLayer()
- PCB_BASE_FRAME::SelectOneLayer(): make dismiss dialog by escape key working.
Fixes#7578https://gitlab.com/kicad/code/kicad/issues/7578
Assume that the user has locked tracks/vias for a reason and would very
much like them to remain in place, regardless of the operation
Fixes https://gitlab.com/kicad/code/kicad/issues/7562
1) Don't remove items from group when the group is being deleted as
well. We need those pointers for undo/redo.
2) Flip the GROUP/UNGROUP undo status when undoing/redoing group and
ungroup actions.
3) Remove PCB_GROUP_T from the rebuild-netlists section. Nothing
about an item's group/ungrouped status changes its netlisting.
4) Add PCB_PAD_T to the rebuild-netlists section. It can probably
only happen in the footprint editor where we don't care about
netlisting, but it's conceptually more correct and who knows what
might change in the future....
Fixes https://gitlab.com/kicad/code/kicad/issues/7540
This simple commit makes the Pcbnew copy to clipboard operation snap the
obtained cursor position to grid before the anchors are calculated. This
is also the way the move operation has been apparently working so far.
To be clear, this will not remove all cases where the copy reference
point is not aligned to the grid (see the linked Gitlab issue for
details about the problem). This will still occur for zone outlines that
are not axis-aligned. But such behavior has been so far also present in
the moving operation, so this may be as designed.
Fixes https://gitlab.com/kicad/code/kicad/issues/7507
For some kinds of issues, we want a way to point them out but
we don't need to keep the user from being able to close the
dialog. A separate Validate button lets the user check for
issues if desired, but these issues are not "fatal" and can
be ignored if the user wishes.
The removal of the pcbnew general section was meant for v6 but passed
the deadline. This removes the duplication of thickness. We do not
allow the thickness to be different between nominal and stackup, so
keeping separate text fields is unnecesarily confusing to users
The automatic wire tool was not working in the gridless mode because the
grid snapping on/off logic present in `EE_GRID_HELPER`, based on the
value of `m_enableGrid`, was not taking into account that another check,
but for the global KiCad setting, would be performed in the `Align()`
method, which is inherited from `GRID_HELPER`.
I've modified all tests of `m_enableGrid` value to also take the global
setting into account, and moved checks for it to `Align()` method, as
it's more consistent this way.
I've also removed an override of the `Align()` method in the
`PCB_GRID_HELPER` class, as it was made redundant by my changes.
Fixes https://gitlab.com/kicad/code/kicad/issues/7402
Some CADSTAR designs use zero sized pads because CADSTAR did not allow
arbitrarily assigning pad numbers: they had to be sequential. The
workaround to this was to use zero sized pads. However this causes DRC
errors in KiCad, so we load them just in a mask layer.
1. Make sure to convert to KiCad units
2. There are some KiCad rules that have no CADSTAR equivalent.
Make a guess as to what an appropriate rule could be.
It just means the outline wasn't malformed. It might still be
empty.
[Edit: my fix collided with JP's, so all this really is now is a
couple of slight performance improvements.]
Keep Through Hole pads on external layers should be ON by default, because
such a pad does not allow soldering/unsoldering, if this option this option is OFF
Also rename dialog_enum_pads.fbp as dialog_enum_pads_base.fbp, as usual
Keeps parent information such as UUID and pad removal options for
elements that are just updated in routing rather than created fresh or
fully destroyed.
Fixes https://gitlab.com/kicad/code/kicad/issues/7460
You had to set the line width in the SVG export dialog in order to
change the displayed line width in the 3d viewer. The line width did
not affect the SVG export. Removing in favor of the Board design
settings for line widths
Some differences are due to difference between libraries being files
in symbol editor and directories in footprint editor, but this unifies
most of the rest.
It also dispenses with the save-to-library vs save-to-document distinction
in the GUI because it makes for too much variability in the GUI, was
implemented differently between the two editors, and isn't needed as much
anymore now that we have the highly visible infobar.
There was also an issue that the save-to-board icon occupies the same
location and has the same size/shape/colours as the Board Setup icon in
PCBNew.
Fixes https://gitlab.com/kicad/code/kicad/issues/7215
Physical layer test needs to be board-visible-layers, not pad-layers.
Also cleans up the tests so that LOD factors in. (When things
disappear due to zoom level they should not be selectable.)
Fixes https://gitlab.com/kicad/code/kicad/issues/7238
Now that footprint properties inherit custom fields from schematic
they are very useful.
Setting custom properties from python will also be handy in plugins
that need to somehow mark their own objects.
Stores a cache of the convex hull similar to the bounding box. Modifies
the hull based on actions to allow complex footprint graphics without delay
Fixes https://gitlab.com/kicad/code/kicad/issues/7341
This no longer returns the first pad, just the first by pad sort order
in output. The call is not used correctly in pcbnew, so removing to
prevent future misuse
For some reason the post-pinned local variable references were replaced
with calls to the points themselves (which are the pre-pinned values).
Fixes https://gitlab.com/kicad/code/kicad/issues/7291
The nettie hack is used in DRC but we also need to use it in zone filler
to allow zones to be used with netties. We limit this to the
appropriate nets in the nettie
Fixes https://gitlab.com/kicad/code/kicad/issues/7351
The conditions prevent use in immediate mode for common actions. While
this makes enable/disable easier in menu items, this fix is a patch
until we refactor the action framework to distinguish between immediate
mode hotkeys that should start without selections and menu items that
need a selection before running.
Fixes https://gitlab.com/kicad/code/kicad/issues/5333
One way to do this would have been to keep the Uuids in the editor
copy. However, this opens us up to errors if we forget a Save As or
export path and end up writing the Uuids out somewhere else.
In the end, it felt safer to store a map of the original Uuids and
restore them if we happen to save the editor footprint back to the
board.
Fixes https://gitlab.com/kicad/code/kicad/issues/7312
... because that's where we do hole-to-copper clearance testing.
Also augments the Clearance Resolution reporter to report on said
hole clearances.
And changes the interpretation of HOLE_CLEARANCE_CONSTRAINT to include
local pad clearance overrides.
Fixes https://gitlab.com/kicad/code/kicad/issues/7325
New strategy isolates all draw/don't draw decisions to the ViewGetLOD
routines, and all dimmed/not dimmed to PCB_RENDER_SETTINGS::GetColor.
The actual drawing in PCB_PAINTER is more-or-less conditon free.
Also adds new layers for pad and via hole walls so that they can be
controlled for high-contrast mode.
Also changes the drawing paradigm so that the pads are drawn even when
not on the high contrast layer, just in low contrast. The hole wall
is drawn in high contrast. This actually makes things clearer to the
user (although to be honest was done to keep from having to re-render
pads when the high contrast layer changes since we have two separate
layers now that we can adjut colours on).
Fixes https://gitlab.com/kicad/code/kicad/issues/7328
Moves rule area triangulation outside of layer iterator and keeps the
triangulation_valid flag false until the triangulation has been
completed.
Fixes https://gitlab.com/kicad/code/kicad/issues/7032
Non-layer elements can be logical centers for the circle and arc tools.
We allow the non-layer to snap for the center points and revert to only
on-layer elements after
Fixes https://gitlab.com/kicad/code/kicad/issues/7327
Zone connections (even half-connections) should trigger the connected
logic in flashPads. This removes the double-standard for painting pads
and imposes thermal standoff based on the net connection to the pad if
it shares a net, not just if connected
CHANGED footprint positions files -> component placement files.
Not all footprints get placed, so component is a more neutral word.
Use of "placement" instead of "positions" removes the double-plural
that makes footprint positions files such a mouth-full.
Fixes https://gitlab.com/kicad/code/kicad/issues/7097
New strategy is to draw the hole wall cylinder if the pad (or via)
isn't flashed on a particular layer. This is (1) more correct,
(2) keeps pads which aren't flashed on any layer from disappearing,
and (3) allows us to remove a bunch of the other special cases.
Or at least I think it does. The proof will be in the (lack of)
follow-on bug reports....
Also fixes a bug where via annular rings weren't highlighted in
high contrast mode.
Fixes https://gitlab.com/kicad/code/kicad/issues/7279
The BuildSteps folder now contains the CMake scripts that are
called during the build process as individual stages, while
the root CMakeModules directory contains the CMake files used
during the configuration phase.
Show the dragging tools in the context selection menu when a single
component is selected. Prevent the dragging tools from appearing when
more than one object is selected, as they will not activate anyway.
Fixes https://gitlab.com/kicad/code/kicad/issues/7258
Also removes the "tip" icon which doesn't work on OSX and isn't as
important as the graphics which actually convey information. Replaced
with the more generic "Note: " string.
Fixes https://gitlab.com/kicad/code/kicad/issues/7281
including wx.h is useless for this file, and include wx.h in a lot of other files.
Include wx.h must be made only when needed in a given file, because on Windows
it include some headers that frequently create collision with kicad declarations
Also fixes some latent bugs with not updating the local ratsnest
visibility in other places, and the netcode in update PCB from schematic.
Fixes https://gitlab.com/kicad/code/kicad/issues/7202
After a rotation or flip the reference point has to be restored to the
previous value to prevent the dragged item from warping to the cursor.
Before this change, the reference point was only cleared, causing odd
behavior when a rotation or flip was performed.
Fixes https://gitlab.com/kicad/code/kicad/issues/7112
They are used for chamfered round rect pads, and can be used for custom shaped pads.
No actual change currently, but the shape rotation of custom pads and chamfered rr pads
can be now used in gerber plots.
ADDED a new pin electrical type "free" for internally unconnected pins.
CHANGED the "unconnected" pin electrical type is now represented by
"no_connect" in files and netlists. (The legacy syntax is also accepted
in files.)
We have been writing footprint versions for a while now, but we were
still interpreting the lack of a version as "current". This changes
lack of a version to mean "last format before we started writing out
versions".
Fixes https://gitlab.com/kicad/code/kicad/issues/7143
Two issues found with the locking system used to prevent access to
stale connectivity data during the zone fill process:
1) a std::mutex has undefined behavior if you try to use it to guard
against access from the same thread. Because of the use of wx event
loops (and coroutines) it is entirely possible, and in some situations
inevitable, that the same thread will try to redraw the ratsnest in the
middle of zone refilling.
2) The mutex was only guarding the ZONE_FILLER::Fill method, but the callers
of that method also do connectivity updates as part of the COMMIT::Push.
Redrawing the ratsnest after the Fill but before the Push will result in
stale connectivity pointers to zone filled areas.
Fixed (1) by switching to a trivial spinlock implementation. Spinlocks would
generally not be desirable if the contention for the connectivity data crossed
thread boundaries, but at the moment I believe it's guaranteed that the reads
and writes to connectivity that are guarded by this lock happen from the main
UI thread. The writes are also quite rare compared to reads, and reads are
generally fast, so I'm not really worried about the UI thread spinning for any
real amount of time.
Fixed (2) by moving the locking location up to the call sites of
ZONE_FILLER::Fill.
This issue was quite difficult to reproduce, but I found a fairly reliable way:
It only happens (for me) on Windows, MSYS2 build, with wxWidgets 3.0
It also only happens if I restrict PcbNew to use 2 CPU cores.
With those conditions, I can reproduce the issue described in #6471 by
repeatedly editing a zone properties and changing its net. The crash is
especially easy to trigger if you press some keys (such as 'e' for edit)
while the progress dialog is displayed. It's easiest to do this in a debug
build as the slower KiCad is running, the bigger the window is to trigger this
bug.
Fixes https://gitlab.com/kicad/code/kicad/-/issues/6471
Fixes https://gitlab.com/kicad/code/kicad/-/issues/7048
It's no longer used for dimensions so doesn't need the line thickness
controls.
The spacing was a bit wonky.
Changed to disabling rather than hiding visibility checkbox. The less
the GUI moves around the better.
std::map is not move-constructable on MSVC because
it is not declared noexcept (and is not required to be
by the standard)
This means that a struct containing graphic_element
cannot be inserted into a vector with move semantics without
compile errors. However, wrapping it in a unique_ptr
works.
Seth Hillbrand
Jeff Young
jean-pierre charras
Tomasz Wlostowski
Marek Roszko
Jon Evans
david-beinder
Roberto Fernandez Bautista
Mikolaj Wielgus
Zoltan Gyarmati
PJM
James Jackson
Wayne Stambaugh
Simon Richter
Thomas Pointhuber
Michael Kavanagh
qu1ck
Ian McInerney