The tests for file existence and write status perform two file stats
which cause performance issues on slow network shares. Now the file
state is determined at load time and stored in the SCH_SCREEN object
so file access is no longer required.
Fixes https://gitlab.com/kicad/code/kicad/-/issues/9343
Make auto save recover all auto saved sheet files not just the root sheet.
This does not fix existing auto save issues because recovery would require
an iterative reload for each automatically saved sheet file and would only
happen once.
Fixes https://gitlab.com/kicad/code/kicad/-/issues/9288
Changes a dot to be a square pixel (linewidth x linewidth). This allows
the removal of IU dependencies and ensures that a dot is always visible
on screen. Also makes sure that cairo is setting the current linewidth
during its stroke routines
Fixes https://gitlab.com/kicad/code/kicad/issues/9362
We already exclude components based on the schematic property when
generating the XML. Don't make a hidden second pass in the python
netlist reader by default
Fixes https://gitlab.com/kicad/code/kicad/issues/9326
Leaving the forced position on results in the auto-line linking to a pin
for far longer than we intend if there are other hits in the area of the
cursor
Fixes https://gitlab.com/kicad/code/kicad/issues/9321
Calling UpdateItem() may be performed by in a common loop, e.g. for(
SCH_ITEM* item : GetScreen()->Items() ) or similar. We cannot call
GetScreen()->Update( SCH_ITEM* ) in this routine as it will remove and
re-add the item to the RTree, invalidating iterators. If needed, the
items need to be cached to an external container before updating
Fixes https://gitlab.com/kicad/code/kicad/issues/9318
Unused variables in function calls can be commented out. Unused
return variables get a new routine `ignore_unused()` that silences the
warnings with zero overhead.
Also cleans up the dialog a bit.
Also changes the workings of NormalizePath to return the input if
it couldn't be shortened which saves open coding that behaviour in
all its callers.
Fixes https://gitlab.com/kicad/code/kicad/issues/9036
When we zoom after centering, the screen does not display the item at
the appropriate location until you center again. Choose the zoom first
to set the appropriate scale before translating the viewport
When the user deletes a pin row, they need to see which row is now
selected for them to delete again. Otherwise, deleting will have
unexpected consequences
Fixes a crash where the screen was destroyed during new loading but
still gets dereferenced by the UI update. Also puts defensive checks in
the UI conditions against null configuration settings
If we do a real drag, the timer will be cancelled. If we don't, then
the user is still holding the mouse button down and would probably like
to see the menu. Otherwise it appears the action is dropped.
In the symbol fields table the reference contains only the letter, so
the number must be appended. Conversely, we don't want the unit suffix
as all units are folded into a single symbol.
Fixes https://gitlab.com/kicad/code/kicad/issues/9238
Keeps the lines connected but doesn't solve the crossed-wire mixups.
This allows common usage (mirroring labels/rotating segments) and leaves
prevention of the crossed wires to the user.
We can't really prevent crossed wires when rotating e.g. a part with
wires connected to 4 sides and users will not expect that. But we
should not disable useful features like mirroring multiple parallel
items to avoid a known impossible situation
Also provides an alternative solution to 345f506f0c that allows us to
rotate groups without destroying connections on hover
Fixes https://gitlab.com/kicad/code/kicad/issues/8403
Fixes https://gitlab.com/kicad/code/kicad/issues/8523
doSelectionMenu returns false to indicate a cancelled menu. This needs
to be handled in selectPoint when directly accessing doSelectionMenu in
the same manner as SelectionMenu() events
Fixes https://gitlab.com/kicad/code/kicad/issues/9218
We have a fixed bottom-align for the text and labels in eeschema.
Therefore, mirroring vertically needs to account for the text's offset
from the bounding box center.
Fixes https://gitlab.com/kicad/code/kicad/issues/9208
The select-single-wire stuff is no longer needed, and the prefer-wire-ends
stuff needs to be distance-based.
Also improved wire hit-testing to include bubbles the size of the dangling
symbol at each end.
Fixes https://gitlab.com/kicad/code/kicad/issues/9187
1) Expand the accuracy to at least the grid spacing
2) Move pin-selects-symbol logic to Selectable()
3) Recode pick-closest algorithm to handle exactHits and inexactHits
separately
4) Remove pin target from bounding box in favour of expanding it to
at least as wide as the pin decoration (this prevents the accuracy
off the end of the pin from growing too large)
Also removes the message bar display of assigned netclass for
buses and bus-to-bus entries.
Also fixes a bug where assigning a netclass via the canvas only
looked at the first level of bus members (and not any nested
members).
Also fixes a bug where the bus name validator tried to validate
a vector bus first -- which doesn't work as a vector bus may be
nested in a group bus.
Also fixes a bug where we were failing to check for illegal
chars in bus definitions which otherwise passed the bus parsers.
See additional comments in the bug report.
Fixes https://gitlab.com/kicad/code/kicad/issues/9160
While it would make more sense to have them both use the old canvas
algorithm (of only assigning to the members), I'd be *very* hesitant
to change the Setup Dialog and underlying machinery this late in the
6.0 release.
Fixes https://gitlab.com/kicad/code/kicad/issues/9160
Escaping the document file (.dcm) alias names breaks the link between the
document file and the library (.lib) file. This causes the information in
the document file to be lost.
Fixes https://gitlab.com/kicad/code/kicad/-/issues/9130
When opening new schematic, the undo/redo list is invalidated.
CleanupSchematic uses it before presenting the new schematic to the
user, so we need to make sure that old data are not preserved here past
their pointer lifetimes
Global labels can optionally have intersheet references. When enabled,
they expand the bounding box of the label. Because they can be far away
from the actual label, the bounding box is not a good approximation for
the hittest
We have two cases where we might be requesting a disambiguation menu.
In one case, we already have the event loop waiting and we need to run
the selectionMenu through the action system. In the other case, the
tool isn't on the stack and won't activate. Here we need to run
directly and we are assured of not crashing because the tool isn't
active.
Fixes https://gitlab.com/kicad/code/kicad/issues/9154
Jeff Young
Wayne Stambaugh
Mikolaj Wielgus
Roberto Fernandez Bautista
Seth Hillbrand
Marek Roszko
Jon Evans
Mike Williams
jean-pierre charras