Re: endless tab?

[Blöchl Bernhard]

Am 06.02.2018 03:15, schrieb Andrew Bernard:
> Hi bb,
>
> Do you simply want 17/16? I use times like this very often. Just use
> time 17/16 and you will get barlines.

Thank you! This is a possibilty if one really wants 17/16. I do not. But 
lilypond does not serve the sloppy user perfectly from using such 
constructs inattentively - lilypond could do!

> But I think you are algorithmically generating the notes and
> durations, in which case, why can't you also generate the /time
> commands for each section where you want a particular time signature?

> My hand written pieces often change time sig every bar, and with any
> sort of signature you like - lilypond handles that just fine, and
> places barlines as expected. I use 15/16 commonly. Not a problem. And
> in fact I hide the time signatures (for reasons not relevant to this
> topic) so you can do that too if you want, and still have barlines.

For this case I think it's better to \omit the bar check.

> This is a deeply nested thread so it's hard to read. Apologies if I
> have entirely missed the point!

It is always good to exchange ideas and discuss about.

> Andrew


Let me add a third experiment:
\version "2.19.80"
\relative c ''{ d4  cis8  c8  b8. ais8  a
            #gis a b c d  e f g a b c
}


and a fourth experiment:
\version "2.19.80"
\relative c ''{ d4  cis8  c8  b8. ais8  a
            d4  cis8  c8  b8. ais8  a
            d4  cis8  c8  b8. ais8  a
}

I find it remarkable that warnings arise "one bar later". I would 
understand this behaviour , if the first bar is not completed and there 
is missing something to fill the bar - bit in this case lilypond is not 
seting a bar line. Butt remeber that the first bar is overstuffed by 
1/16.


Thanks for your mail and regards.

> Again old arguments. Imagine two experiments:
>
> > Case 1: I set an arbitrary bar (the manual calls it "bar check")
> > version "2.19.80"
> > relative { d''4  cis8  c8  b8. ais8  a  gis | }
> > That enables lilypond to checking the bar content and tells me in
> > this case that the check failed ...
> > 17/16 is > 4/4
> >
> > Case 2: lilypond is setting the barline on the same position
> > (wrong/critcal, however you might call it)
> > version "2.19.80"
> > relative { d''4  cis8  c8  b8. ais8  a  gis  }
> > The bar content is 17/16>4/4 again and tells NOTHING!
> > As there is set a barline lilypond can check the bar content and
> > ....
> > Nobody has explained me why in this case lilypond should not be
> > able to use its algorithm to find 17/16 ? The algorithm is existing
> > - use it for this as well!