Christian Menard

Yes, the proposal is to always shutdown at tag (timeout, 0). Only if there are no periodic timers and the runtime runs out of events, it can shutdown earlier. >...

> From what I understood from @cmnrd, this PR provides fixes for a problem that shouldn't exist in the first place (or, rather, should be solved in a different way)....

I updated this branch today (merged in master, also updated the submodule and merged in reactor-c main) and extended CI so that we test both Python 3.10 and 3.11. However,...

Have we made progress in figuring out why the tests do not pass? > I noticed that we're increasing the testing burden by testing both Python 3.10 and 3.11. What...

Since Python 3.12 was released recently, we should actually also add it to the matrix.

There might still be some problems in the gradle configurations. Something I still do not fully understand is when gradle actually evaluates tasks. I am also not sure what "created...

Interesting, except for the `src-gen` directory produced by xtext, we explicitly ignore all `src-gen` and `fed-gen` directories. But perhaps this ignoring is what makes Gradle slow. We have this in...

Sure. Here is the code used to produce the diagram from above. ``` target Python reactor ClientMethodTransactor { @side("west") input request @side("west") output response physical action p logical action l...

Seems closely related to https://github.com/lf-lang/lingua-franca/issues/1919

I wasn't aware of this issue before. In C++ target, reaction names start at 1 (because they are displayed this way in the diagrams). I don't know about the other...