The drivers need to get passed up and down the hierarchical sheets. In
order to do this, both the sheet pin and the hierarchical pin need to
be in the changed items. However, we only get sheets in the screen
items list while the pins are the elements that get set dirty
Fixes https://gitlab.com/kicad/code/kicad/-/issues/17797
(cherry picked from commit 5f16c5892c)
Fixed point output can create arbitrarily long string representations of
floating points. When this happens, we would like to just represent the
scientific notation to a limited precision
Fixes https://gitlab.com/kicad/code/kicad/-/issues/17890
(cherry picked from commit 24f02e72d0)
Zoom-to-selection should only show the elements that we can see, so
avoiding hidden fields in footprints keeps the zoom window appropriately
sized
Fixes https://gitlab.com/kicad/code/kicad/-/issues/15245
(cherry picked from commit 416033e8e5)
When loading schematics/pcbs, notification of font replacements might be
warranted but in libraries, this warning is not helpful and intrusive
(cherry picked from commit 11c6164934)
Previously, we only allowed pins of the same time as well as passive
pins to be stacked. This change allows NIC pins as well since they are
not internally connected, there is no reason that they cannot be tied to
another pin electrically
(cherry picked from commit 0e0ada8e4e)
- Add a test to be sure a SCHEMATIC exists before trying to use it
- Delete items in Repeat list and undo/redo list before deleting the schematic
- Delete also these items when loading a new schematic (the repeat list was
not cleaned previously)
( include qa compatibility from master )
Before these fixes I saw a use after delete issue due to dangling pointers
stored in Repeat list.
From master branch
Broken parent symbols could lead to invalid return values and flattened
library symbols cannot be derived from a parent symbol.
(cherry picked from commit 360d185e3b)
Ensures that if moving or deleting a netclass directive label, the
netclass on the previously-assigned net doesn't have the stale
netclass re-applied.
Cherry-picked from 0826c10a20
(Along with prior connectivity cherry-pick...)
Fixes https://gitlab.com/kicad/code/kicad/-/issues/18173
Also fixes a bug where the pattern assignment cache was not cleared
if all pattern assignments were deleted in the setup netclasses
panel.
Cherry-picked from 9e1a5eb1bd
We look at the parent when deleting EDA_ITEMs but when clearing the test
sheet, we delete the sheet before the pin, causing a use-after-free.
Avoid this by setting parents to null in this case
Fixes https://gitlab.com/kicad/code/kicad/-/issues/17961
(cherry picked from commit b2e0c2606f)
Cache page numbers during sort.
Don't construct SCH_SHEET_PATH when reference will do.
Don't construct SCH_SHEET_PATH when KIID_PATH will do.
(cherry picked from commit 758974f5aa)
Reference, Value, Footprint and Datasheet all reference elements that
should not use trailing/leading whitespace or similar hidden charaters.
Other fields may utilize leading/trailing whitespace for obscure
purposes but won't break functionality as they are display only
(cherry picked from commit 6dcb9bb664)
The drivers need to get passed up and down the hierarchical sheets. In
order to do this, both the sheet pin and the hierarchical pin need to
be in the changed items. However, we only get sheets in the screen
items list while the pins are the elements that get set dirty
Fixes https://gitlab.com/kicad/code/kicad/-/issues/17797
Cherry-picked from 5f16c5892c
Fixes https://gitlab.com/kicad/code/kicad/-/issues/17984
When moving / deleting a power symbol, in some instances the
symbol is marked dirty, and sometimes the pin (depending on
whether the symbol is the primary edited item, or whether
the pin is identified through an edited item subgraph). If
the pin and the symbol are marked dirty, the pin dirty flag
was not being cleared.
Additionally, not all extracted items were being deleted
from subgraphs correctly.
Both resulted in dirty state and duplicated items in the
subgraphs during incremental connectivity, which resulted
in essentially corrupted subgraph states.
The selection heuristic broke down when one item was a filled
shape. Because all hit tests would succeed with distance 0 for
these shapes, they would always be considered the closest item
to the exclusion of all else, which made it very hard to
click on a graphic inside a filled shape.
Now, recognise when an item would be "dominating" and
decline to promote it to the "closet" spot. It will still
be selectable if there are no other items nearby, or if
there are multiple shapes.
(cherry picked from commit fd4c15517f)
Seth Hillbrand
Ian McInerney
Jeff Young
Alex Shvartzkop
Wayne Stambaugh
jean-pierre charras
JamesJCode
Petri Hodju
Mike Williams
John Beard