Hi Lukas,

> Please forgive my laziness - grepping for a source line is not something
> I'm used to doing on my phone on a bus ride. :-)

This was not a Vorwurf, it was a confirmation.

> Yes, I think it's safe to assume this, even if the user neglects to put
> a rhythmic event at the beginning of each bar:

I think in fact this needs to be done for the timing translator to actually be 
able to update the measure count, so I’d expect this one to schedule this.

> I think this is ensured by the stop-translation-timestep hook of the
> Timing_translator, if I read the C++ source correctly (something I'm
> always struggling with):
> 
> constautomp
> = from_scm (get_property (this, "measurePosition"), Moment (0))
> .main_part_;
> Rational barleft = 0;
> if(mp < 0)
> barleft = -mp;
> else
> barleft = measure_length (context ()) - mp;
> if((barleft > 0) && isfinite (barleft))
> {
> autonextmom = Moment (now_mom ().main_part_ + barleft);
> find_global_context ()->add_moment_to_process (nextmom);
> }

ah right, same thought. So it works out. This should be quite easily be 
implementable by the edition engraver, but I don’t quite have an overview over 
how this one works, neither do I really have the mental energy to try to 
understand it.

> Lukas

Cheers,
Tina

Attachment: signature.asc
Description: This is a digitally signed message part.

Reply via email to