The basic problem is that the DRC engine does length testing and skew
testing by collecting all the diff pair constituent parts and pairing them
itself. Since each part is collected on its own, we need to ignore the
'B' unit when evaluating any conditional expressions. However, doing this
in general means that when evaluating "OwnClearance()" we also ignore the
'B' unit and return the diff pair CLEARANCE_CONSTRAINT when we shouldn't.
This implements a more discerning test which know what the current requested
constraint is when evaluating expressions.
See also https://forum.kicad.info/t/solved-custom-differencing-rule-not-working-drc/34034/6
Fixes https://gitlab.com/kicad/code/kicad/issues/11314
SIMULATOR has now an Attach() method, which should be called
with proper SIMULATION_MODEL that should be simulated, before calling Run()
Concrete class of SIMULATION_MODEL for ngspice is NGSPICE_CIRCUIT_MODEL,
renamed from NETLIST_EXPORTER_PSPICE_SIM. DIALOG_SIM_SETTINGS relies
on above mentioned object, so it was added as an argument of the constructor.
We only need one bottom to top layer sequence definition.
Plot a sequence of layer IDs (LSEQ) in the order of the sequence.
Add helper method to layer set (LSET) to create a sequence of layer IDs
using another sequence for ordering.
Now that we are dealing with individual connection elements that do not
update their connected elements as well, we can thread the update, just
being careful to guard any remaining updates (bus_enty/busLine) that
need reciprocal updating
Fixes https://gitlab.com/kicad/code/kicad/issues/10974
(cherry picked from commit 6a53e318e5)
If the SCH_ITEM has already been processed, the extra time needed to
iterated over the memberset and get the SCH_CONNECTION when we won't use
it is not neccesary.
Fixes https://gitlab.com/kicad/code/kicad/issues/10974
(cherry picked from commit 776a28a10e)
Normally, you will gain by resursing a nested loop only over the
pairs that are not already handled. In this case, however, you lose
time because you step outside of the cache by adding the reciprocal test
at each step.
Instead, we process one element at a time, keeping it cached and loop
over all other elements to add to the connection. This saves us about
75% of the time for larger loops (e.g. stacked power pins on a large
BGA)
Fixes https://gitlab.com/kicad/code/kicad/issues/10974
(cherry picked from commit 3a98eacdb9)
Fixes:
* Various special cases around junctions on pins and dragging.
* Some rotations of endpoints resulting in a 45 degree rotate.
* Some cases where it was possible to get a line with neither
endpoint selected, and also substractive unselecting only one of two
selected endpoints unselecting both.
* Use line midpoint for sorting. Start and endpoints aren't
consistent in the order they appear on the X or Y axis. So,
we need to use the midpoint for our position for consistent
sorting when dragging groups of parallel lines where some
have the start and end reversed.
Other:
* Rename TEMP_SELECTED TO SELECTED_BY_DRAG. This is the actual meaning
of the flag, and should reduce confusion as
to when it should be used.
* Move usage of TEMP_SELECTED as an algorithmic mark to CANDIDATE
instead.
* Fix mistaken clearing of START_POINT and ENDPOINT.
* Move endpoint setting and clearing out of narrowSelection, and into
selectPoint and selectMultiple.
* Don't show dangling end warnings on new lines
Jeff Young
Marek Roszko
Seth Hillbrand
Sylwester Kocjan
jean-pierre charras
markus-bonk
Wayne Stambaugh
Mike Williams