A state event of a 2.0 ME FMU is located too early

[TorstenBlochwitz]

FMPy locates a state event of an 2.0 ME FMU too early. According to the FMI 2.0 specification (point 3 above below Figure 4) an event instant occurs "where an event indicator zj(t) changes its domain from zj > 0 to zj ≤ 0 or from zj ≤ 0 to zj > 0".

Please have a look at the attached picture:

At the last successful time, the event indicator has negative sign (green boxes). Later on, searching for the next event starts. Then fmi2EnterEventMode is called, after fmi2getEventIndicators returned 0.0 (red boxes). But this is too early!

As a result, the outputs discReal1, discReal2, intOut1 and intOut2 do not jump, but stay constant.

The FMU is treated correctly in Dymola, Simulink, OpenModelica and SimulationX.

I can provide the FMU, please let me know your preferred way to do so.

[t-sommer]

Thanks for reporting this problem. Please attach the FMU to the issue.