This also removes the GROUP/UNGROUP-specific undo actions.
This also fixes a bunch of undo bugs when duplicating
group members, creating pins, etc.
This also fixes some undo bugs when dividing wires etc.
This also fixes some bugs with new sch items not
being created within an entered group.
The lexicographical sort for hierarchical pages needs to satisfy the
following:
- Ensure a parent comes before all its children
- Ensure all children with the same parent are sorted by page number
- Ensure all of a parent's children come before the parent's next sibling
To do this, we need to compare at each path level before descending
Fixes https://gitlab.com/kicad/code/kicad/-/issues/20701
There should be far fewer hatched objects than other
objects, so we're spending too much effort finding
all the possible damage when we probably should
just be auto-regenerating all the hatching.
This also moves it out of being done during redraw,
which was proving problematic. Plus the refill
zones architecture does it during commit, and has
a lot more miles behind it.
Recommendation is to avoid using the year nomenclature as this
information is already encoded in the git repo. Avoids needing to
repeatly update.
Also updates AUTHORS.txt from current repo with contributor names
Also adds a mode for AUTOADDED fields.
Also fixes a couple of bugs where a manual-level autoplaced
symbol would get reset back to auto-level when autoplaced.
Commit 83de056de9 broke loading legacy schematics that are not shared.
Legacy schematics do not contain instance data "AR Path=" entries when
the schematic file was not shared. The reference and unit information
was only stored in the symbol reference field and unit entries in the
symbol definition.
Fixes https://gitlab.com/kicad/code/kicad/-/issues/18779
This fix adds missing symbol instance data that defaults the instance
data to no annotation which will will trigger a annotation request on
any operations that require a fully annotated schematic.
Fixes https://gitlab.com/kicad/code/kicad/-/issues/18486
Also removes the side-effect that SCH_SHEET_LIST's
c'tor would sort the list (and write virtual page
numbers) anytime the starting sheet was the root.
Also, definitely don't build a SHEET_LIST (sorted or
otherwise) if you're not even going to use it.
Also don't build SCH_SHEET_LISTs on idle events. Better
to just always have the Next Sheet button enabled (we
already beep if you click it and there's no next sheet).
Also, use a SCREEN_LIST when you can. It's much cheaper
to create.
When adding a sheet using a schematic from another project, set the
project name for the new symbol instance data to the current project
rather than the project name from the copied instance data.
A sheetpath is required to correctly resolve text variables.
Depending on currentSheet is rife with bugs.
There are many places where we do *not* want to be prepending
field names to the field values, such as netlisting,
building PDF hypertext menus, etc.
Also, Find/Replace needs to work on unresolved text, as
that's what we're going to display (and if replace nuked
your variable references you wouldn't be happy).
Only update value and footprint fields from instance data if it's not
empty. These fields were not always stored in the instance data so
loading them from instance data that does not contain them will clear
the fields.
https://gitlab.com/kicad/code/kicad/-/issues/13735
Rather than update library symbols one at a time, queue them up by
schematic symbol to prevent multiple updates to the same symbol in
complex hierarchies. This also removes all of the library symbols
first which will clear out all of the library symbol variants that
were created by modifying library symbols in place and/or changes to
the symbol in the library.
Don't add new variant library symbol if an equivalent variant already
exists in the schematic local cache. This prevents duplicate library
symbols from being added to the local cache when the first variant in
the cache does not match that of the symbol being added.
Also moves passive RLC inference out from migration to just-in-time
creation for the simulator or netlisting.
Also fixes a version guard mismatch because the spice migration was
done inside UpdateSymbolInstances (which has its own version guard).
Also changed UpdateSymbolInstances to UpdateSymbolInstanceData so
someone else in the future doesn't think it's a general-purpose symbol
instance updater.
Jeff Young
Seth Hillbrand
Mike Williams
Wayne Stambaugh
John Beard
Jon Evans
Marek Roszko
jean-pierre charras