Lars T. Kyllingstad
Lars T. Kyllingstad
When re-reading it now, I see a flaw in my own argumentation. There is one use case for indexes that I've missed, namely when you want to refer to an...
Not really. I see I've been a bit hasty when writing this, but this was about the C library. There, we're still using separate FMU-unpacking directories for separate slaves, even...
An alternative would be to make our `slave` interface more explicit by expanding it with a `do_iteration()` function which is equivalent to `do_step(0.0)` in FMI 1.0 and is a no-op...
Also worth noting is that `mock_slave` (in `test/cpp/mock_slave.hpp`) does not adhere to an FMI 2.0-like interface in this regard. It does not update any outputs unless `do_step()` is called.
Isn't this mostly a GUI thing?
Possibly relevant: In an FMI 2.0 model description you can specify the *display units* for variables, which may be different from the actual variable value units. Look for `` in...
I'm not too familiar with those, so I'm not sure what you mean. But I believe they are based on code generation, which might make things slightly different? It would...
> What is the feeling about doing what agx do by inlining a default noop in the header? > IMO it's super neat. I still think they should be separate...
> > If a base class is used as an interface, make it a pure abstract class > > Still holds with inline defaults though.. Since they are not apart...
> Semantics I guess, but below `observer` is technically what you call a _pure abstract class_. The default implementations is defined outside the class definition. > > ```c++ > class...