Fixes#12580
Additionally does not export no_connect netlist annotation on pins
which are both connected to another pin and a NC item, unless all
connected pins are stacked at the symbol level.
Adds testing of pin types to netlist QA unit tests.
Also fixes a bug so that voltages, currents and power dissipations are
only probed if the flag is set -- this keeps ngspice from throwing an
error if you probe something twice (for instance, if you have .probe
commands in text and turn off the auto-probing).
The pressure relief valve was not useful for common work patterns as it
forced the recalculation on many common actions such as bus expansion.
This caused it to actually feel slower than with the pressure relief
valve off.
For most schematics, realtime is now fast enough to not need the valve
and for those that are extremely complex, removing the valve helps this
run more predictably
We don't do anything with the parse tree, and we have to check it for
directives again afterwards to account for when our parsing failed, so
there's not much point in parsing it to begin with.
We only need to parse the libraries for the Sim Model Editor dialog
(so that we can determine the models to put in the popup). Doing it
for netlisting just opens us up to incorrectly parsing the SPICE, and
returns no value.
This also means that a SIM_LIB_MGR manages a single library, and
never multiple libraries.
This also allows the tidying-up of some error reporting structures to
better inform the user what went wrong.
Fixes https://gitlab.com/kicad/code/kicad/issues/13431
This should be OK now because we should be writing out all parameters
now (even if they're default value).
But perhaps more to the point, if we have the safety in there then
we miss cases where we really can't parse the model (value of "{VCC}")
and we need to fall back to a raw spice model.
Fixes https://gitlab.com/kicad/code/kicad/issues/13444
This moves some stuff to REPORTER APIs. Moving more stuff would be good,
but it probably too high-risk at present. We'll wait for 8.0 for that....
Fixes https://gitlab.com/kicad/code/kicad/issues/13359
1) Use C-locale when converting output of parsers to strings.
2) InferSimModel() must run on local copy of fields, not the fields in
the symbol (which might be different if they've already been edited in
the Symbol Properties Dialog, for instance).
3) InferSimModel() should accept a deviceType already set to RLC.
4) Don't output trailing decimal separators if there's no fraction.
1) if a legacy model references a library then we need to see if said
libraray exists and read model from it if so
2) legacy node ordering is by index, not pin name
3) we can't auto-generate a pin map when we don't know the pin names,
so don't try
Also adds simulation model inference to Symbol Editor's Simulation Model
Editor, and adds write-back from both Simulation Model Editors to their
parent Symbol Fields Editor dialogs.
Fixes https://gitlab.com/kicad/code/kicad/issues/12505
Also removes a bunch of std::string stuff from the file handling in the
simulator. All our file handling, env variable expansion, project path,
etc. stuff is wxString based, and jumping through std::string in between
just makes it more complex and increases the potential bug surface.
Also fixes a bug where you'd get two error messages when a spice model
library wasn't found.
Also fixes a bug where you'd get a spice model library not found error
when the text field was empty.
Also fixes a bug where we'd try to absolutize a path starting with an
unresolved text or environment variable. If the path starts with a
variable it's probably absolute, and tacking on the project path in the
error message just obfuscates things.
Fixes https://gitlab.com/kicad/code/kicad/issues/13082
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.
Also includes fixes for instance data and resolving of textvar
references.
Also includes virtual d'tors for IBIS parser to get rid of all the
compile warnings on CLang.
Fixes https://gitlab.com/kicad/code/kicad/issues/12357
When simulating an empty netlist, calling the `.save all` and `.probe
alli` commands gives a segfault in ngspice. We can avoid this by
skipping these commands in KiCad.
Fixes https://gitlab.com/kicad/code/kicad/issues/12938
Follow-up after the KIBIS and KIBIS GUI merge requests.
- Move KIBIS from Pcbnew to Eeschema space,
- Make KIBIS obtain the Ku/Kd coefficients via the `SPICE_SIMULATOR` class instead of calling the `ngspice` executable via `system()`,
- Allow to toggle between differential and single-ended model in the GUI,
- Various GUI fixes and improvements.
- Some preliminary refactoring,
- Use multiple types for IBIS drivers instead of a wftype variable,
- Remove the m_requiresUIUpdate variable and its accessors.
- Fix parsing .model lines with model names containingnon-alphanumeric
characters like - and _,
- Fix parsing libraries in which EOF is not preceded by a newline.
Fixes https://gitlab.com/kicad/code/kicad/issues/12394
* fix broken float values in countries using a comma as separator (missing LOCALE_IO switch)
* fix regression that ignored the pseudo directive Kxx Lyy Lzz nn (coupling factor)
* fix regression that created empty blocks for .subckt and .control
Only pins from the current symbol part were used: the rest was
incorrectly set to Not Connected.
Also slightly change generation of names for Not Connected nets.
This change removes the incorrect assumption that symbol pin numbers are
integers and are the same as indexes in the vector storing the symbol
pins.
"~" is now used to denote a floating sim model pin.
ba7a06f5 tried to fix a (minor) issue in net names when the name was coming
from a input power pin having a '/' in name. But ba7a06f5 blindly escaped any '/'
in net name, and the hierarchy info was lost.
Now only the pin name is escaped (like any other pin name used in net names)
1) Move a bunch of std::map's to std::unordered_map to get constant-time
look-ups
2) Lengthen progress-reporting intervals to spend more time doing work
and less time talking about it
3) Reverse order of SHAPE_LINE_CHAINs in thermal intersection checks to
make (much) better use of bbox caches
4) Don't re-generate bboxes we already have
5) Fix some autos that weren't by reference (and were therefore copying
large datasets)
6) Rename delta progressDelta so it's easier to search for in future
7) Get rid of a few more autos (because I don't like them)
8) Pass large items to lambdas by reference
Fixes https://gitlab.com/kicad/code/kicad/issues/12130
Unfortunately, Windows headers define a lot of macros for common words,
so we had to rename some enums to not collide.
We also fix some of the many bugs related to the new simulation
architecture and the Spice Model Editor dialog.
- Tab-switching,
- Automatic expansion of categories on tab-switch,
- Various minor simulation improvements,
- Various new simulation-related bugfixes.
JamesJCode
Jeff Young
Fabien Corona
Graham Keeth
Seth Hillbrand
Wayne Stambaugh
Marek Roszko
jean-pierre charras
Mikolaj Wielgus
Sylwester Kocjan